text
stringlengths 14
5.22M
| meta
dict | __index_level_0__
int64 0
9.97k
| input_ids
listlengths 128
128
| attention_mask
listlengths 128
128
| labels
listlengths 128
128
|
---|---|---|---|---|---|
Every Year, the rodeo is one of the most anticipated events in Houston. Houston Rodeo Transportation is something to consider when attempting to compete with the anticipated large crowds and inflated parking costs. With our unique fleet, we can transport your company, friends and family to the Rodeo with Ease. From selecting the perfect vehicle, to picking the right date and time, weather its the cook-off, a concert, the Livestock Show or the Carnival, we got you back! Contact us and tell us what you have in mind. We will help you select the vehicle that best matches your needs and budget.
This Year the Houston Rodeo will be held from Monday, Feb.25 to Sunday, March 17 at NRG Park.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 9,195 |
[
128000,
11769,
9941,
11,
279,
41761,
78,
374,
832,
315,
279,
1455,
30199,
4455,
304,
16386,
13,
16386,
432,
536,
78,
30978,
374,
2555,
311,
2980,
994,
19969,
311,
20874,
449,
279,
30199,
3544,
35851,
323,
70954,
13217,
7194,
13,
3161,
1057,
5016,
26155,
11,
584,
649,
7710,
701,
2883,
11,
4885,
323,
3070,
311,
279,
432,
536,
78,
449,
95121,
13,
5659,
27397,
279,
4832,
7458,
11,
311,
21816,
279,
1314,
2457,
323,
892,
11,
9282,
1202,
279,
4394,
12744,
11,
264,
21497,
11,
279,
37311,
478,
1197,
7073,
477,
279,
78844,
11,
584,
2751,
499,
1203,
0,
9359,
603,
323,
3371,
603,
1148,
499,
617,
304,
4059,
13,
1226,
690,
1520,
499,
3373,
279,
7458,
430,
1888,
9248,
701,
3966,
323,
8199,
627,
2028,
9941,
279
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
11769,
9941,
11,
279,
41761,
78,
374,
832,
315,
279,
1455,
30199,
4455,
304,
16386,
13,
16386,
432,
536,
78,
30978,
374,
2555,
311,
2980,
994,
19969,
311,
20874,
449,
279,
30199,
3544,
35851,
323,
70954,
13217,
7194,
13,
3161,
1057,
5016,
26155,
11,
584,
649,
7710,
701,
2883,
11,
4885,
323,
3070,
311,
279,
432,
536,
78,
449,
95121,
13,
5659,
27397,
279,
4832,
7458,
11,
311,
21816,
279,
1314,
2457,
323,
892,
11,
9282,
1202,
279,
4394,
12744,
11,
264,
21497,
11,
279,
37311,
478,
1197,
7073,
477,
279,
78844,
11,
584,
2751,
499,
1203,
0,
9359,
603,
323,
3371,
603,
1148,
499,
617,
304,
4059,
13,
1226,
690,
1520,
499,
3373,
279,
7458,
430,
1888,
9248,
701,
3966,
323,
8199,
627,
2028,
9941,
279,
-100
] |
just tell both the hostess and your server that you are there to support Haynes House. The restaurant will donate 15 percent of the bill to the house. Please RSVP to 518-642-8155 if you plan to attend. Cooper's Cave is located at 2 Sagamore Street in Glens Falls.
If you are interested in hosting your own event to support us please call the house or send us an email!!!
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 5,990 |
[
128000,
4345,
3371,
2225,
279,
3552,
434,
323,
701,
3622,
430,
499,
527,
1070,
311,
1862,
18276,
4978,
4783,
13,
578,
10960,
690,
33009,
220,
868,
3346,
315,
279,
4121,
311,
279,
3838,
13,
5321,
82989,
311,
220,
21312,
12,
22266,
12,
23582,
20,
422,
499,
3197,
311,
9604,
13,
24421,
596,
50492,
374,
7559,
520,
220,
17,
55361,
309,
461,
6825,
304,
8444,
729,
30743,
627,
2746,
499,
527,
8173,
304,
20256,
701,
1866,
1567,
311,
1862,
603,
4587,
1650,
279,
3838,
477,
3708,
603,
459,
2613,
12340,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
] |
[
4345,
3371,
2225,
279,
3552,
434,
323,
701,
3622,
430,
499,
527,
1070,
311,
1862,
18276,
4978,
4783,
13,
578,
10960,
690,
33009,
220,
868,
3346,
315,
279,
4121,
311,
279,
3838,
13,
5321,
82989,
311,
220,
21312,
12,
22266,
12,
23582,
20,
422,
499,
3197,
311,
9604,
13,
24421,
596,
50492,
374,
7559,
520,
220,
17,
55361,
309,
461,
6825,
304,
8444,
729,
30743,
627,
2746,
499,
527,
8173,
304,
20256,
701,
1866,
1567,
311,
1862,
603,
4587,
1650,
279,
3838,
477,
3708,
603,
459,
2613,
12340,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100
] |
Buriedxhq - Just now the info related to Buriedxhq many search by internet cyberspace user. However, there's not web about Buriedxhq totally. I am trust that you 'are attempting to looking for the news relevant to Buriedxhq until found this web.
But, possibly a post about Buriedxhq on this web can't satisfy your want so that still must added comprehensive lettering. that could be due to issued articles at this site 's been made some time ago & has not been updated, so that were not able to satisfy your needs..
These are my articles related to Buriedxhq . You can read one of these. I hoped that the following writings can be a useful for you.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 3,912 |
[
128000,
62339,
1142,
87,
82999,
482,
4702,
1457,
279,
3630,
5552,
311,
12649,
1142,
87,
82999,
1690,
2778,
555,
7757,
9693,
1941,
1330,
1217,
13,
4452,
11,
1070,
596,
539,
3566,
922,
12649,
1142,
87,
82999,
12756,
13,
358,
1097,
7095,
430,
499,
364,
548,
19969,
311,
3411,
369,
279,
3754,
9959,
311,
12649,
1142,
87,
82999,
3156,
1766,
420,
3566,
627,
4071,
11,
11000,
264,
1772,
922,
12649,
1142,
87,
82999,
389,
420,
3566,
649,
956,
27651,
701,
1390,
779,
430,
2103,
2011,
3779,
16195,
6661,
287,
13,
430,
1436,
387,
4245,
311,
11136,
9908,
520,
420,
2816,
364,
82,
1027,
1903,
1063,
892,
4227,
612,
706,
539,
1027,
6177,
11,
779,
430,
1051,
539,
3025,
311,
27651,
701,
3966,
35047,
9673,
527,
856,
9908,
5552,
311
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
62339,
1142,
87,
82999,
482,
4702,
1457,
279,
3630,
5552,
311,
12649,
1142,
87,
82999,
1690,
2778,
555,
7757,
9693,
1941,
1330,
1217,
13,
4452,
11,
1070,
596,
539,
3566,
922,
12649,
1142,
87,
82999,
12756,
13,
358,
1097,
7095,
430,
499,
364,
548,
19969,
311,
3411,
369,
279,
3754,
9959,
311,
12649,
1142,
87,
82999,
3156,
1766,
420,
3566,
627,
4071,
11,
11000,
264,
1772,
922,
12649,
1142,
87,
82999,
389,
420,
3566,
649,
956,
27651,
701,
1390,
779,
430,
2103,
2011,
3779,
16195,
6661,
287,
13,
430,
1436,
387,
4245,
311,
11136,
9908,
520,
420,
2816,
364,
82,
1027,
1903,
1063,
892,
4227,
612,
706,
539,
1027,
6177,
11,
779,
430,
1051,
539,
3025,
311,
27651,
701,
3966,
35047,
9673,
527,
856,
9908,
5552,
311,
-100
] |
Collaborate with experts and executives in Disney's businesses to identify research problems worthy of research investment.
Independently conceive of and propose research problems that are relevant to Disney's businesses.
Conduct research inspired by those problems.
Conduct research directed by an assigned Senior Research Scientist or Research Scientist.
Present research problems, plans, and results.
Document research in internal reports, peer-reviewed publications, and patent applications.
Work with Disney's businesses to commercialize research results.
Provide expert consulting to various business units of The Walt Disney Company.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 621 |
[
128000,
15623,
4422,
349,
449,
11909,
323,
29509,
304,
16795,
596,
9873,
311,
10765,
3495,
5435,
28390,
315,
3495,
9341,
627,
64499,
4501,
77151,
315,
323,
30714,
3495,
5435,
430,
527,
9959,
311,
16795,
596,
9873,
627,
1128,
1076,
3495,
14948,
555,
1884,
5435,
627,
1128,
1076,
3495,
15910,
555,
459,
12893,
19903,
8483,
68409,
477,
8483,
68409,
627,
21886,
3495,
5435,
11,
6787,
11,
323,
3135,
627,
7676,
3495,
304,
5419,
6821,
11,
14734,
79804,
29085,
11,
323,
25589,
8522,
627,
6919,
449,
16795,
596,
9873,
311,
8518,
553,
3495,
3135,
627,
61524,
6335,
31831,
311,
5370,
2626,
8316,
315,
578,
36367,
16795,
8351,
13,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
] |
[
15623,
4422,
349,
449,
11909,
323,
29509,
304,
16795,
596,
9873,
311,
10765,
3495,
5435,
28390,
315,
3495,
9341,
627,
64499,
4501,
77151,
315,
323,
30714,
3495,
5435,
430,
527,
9959,
311,
16795,
596,
9873,
627,
1128,
1076,
3495,
14948,
555,
1884,
5435,
627,
1128,
1076,
3495,
15910,
555,
459,
12893,
19903,
8483,
68409,
477,
8483,
68409,
627,
21886,
3495,
5435,
11,
6787,
11,
323,
3135,
627,
7676,
3495,
304,
5419,
6821,
11,
14734,
79804,
29085,
11,
323,
25589,
8522,
627,
6919,
449,
16795,
596,
9873,
311,
8518,
553,
3495,
3135,
627,
61524,
6335,
31831,
311,
5370,
2626,
8316,
315,
578,
36367,
16795,
8351,
13,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100
] |
Markus Soomets (born 2 March 2000) is an Estonian professional footballer who plays as a midfielder for Meistriliiga club Flora.
Club career
On 17 January 2020, he returned from Italy to Estonia and signed a 3-year contract with Flora.
International career
He made his debut for Estonia national football team on 11 November 2020 in a friendly game against Italy. He played the full game.
References
External links
2000 births
Living people
Sportspeople from Tartu
Estonian footballers
Association football midfielders
Esiliiga players
Meistriliiga players
Tartu JK Tammeka players
FC Flora players
Serie C players
Estonia youth international footballers
Estonia under-21 international footballers
Estonia international footballers
Estonian expatriate footballers
Expatriate footballers in Italy
Estonian expatriate sportspeople in Italy
|
{
"redpajama_set_name": "RedPajamaWikipedia"
}
| 92 |
[
128000,
9126,
355,
2100,
316,
1441,
320,
16381,
220,
17,
5587,
220,
1049,
15,
8,
374,
459,
54423,
1122,
6721,
9141,
261,
889,
11335,
439,
264,
45067,
369,
2206,
3843,
4008,
16960,
6469,
3061,
6347,
382,
54849,
7076,
198,
1966,
220,
1114,
6186,
220,
2366,
15,
11,
568,
6052,
505,
15704,
311,
69462,
323,
8667,
264,
220,
18,
4771,
5226,
449,
3061,
6347,
382,
34746,
7076,
198,
1548,
1903,
813,
17755,
369,
69462,
5426,
9141,
2128,
389,
220,
806,
6841,
220,
2366,
15,
304,
264,
11919,
1847,
2403,
15704,
13,
1283,
6476,
279,
2539,
1847,
382,
32812,
271,
27002,
7902,
271,
1049,
15,
66332,
198,
45605,
1274,
198,
42079,
16455,
505,
78641,
84,
198,
36,
7876,
1122,
9141,
388,
198,
64561,
9141,
30789,
388,
198,
17812,
4008,
16960
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
9126,
355,
2100,
316,
1441,
320,
16381,
220,
17,
5587,
220,
1049,
15,
8,
374,
459,
54423,
1122,
6721,
9141,
261,
889,
11335,
439,
264,
45067,
369,
2206,
3843,
4008,
16960,
6469,
3061,
6347,
382,
54849,
7076,
198,
1966,
220,
1114,
6186,
220,
2366,
15,
11,
568,
6052,
505,
15704,
311,
69462,
323,
8667,
264,
220,
18,
4771,
5226,
449,
3061,
6347,
382,
34746,
7076,
198,
1548,
1903,
813,
17755,
369,
69462,
5426,
9141,
2128,
389,
220,
806,
6841,
220,
2366,
15,
304,
264,
11919,
1847,
2403,
15704,
13,
1283,
6476,
279,
2539,
1847,
382,
32812,
271,
27002,
7902,
271,
1049,
15,
66332,
198,
45605,
1274,
198,
42079,
16455,
505,
78641,
84,
198,
36,
7876,
1122,
9141,
388,
198,
64561,
9141,
30789,
388,
198,
17812,
4008,
16960,
-100
] |
The Burney Baronetcy, of Preston Candover in the County of Southampton, is a title in the Baronetage of the United Kingdom. It was created on 27 January 1921 for Admiral of the Fleet Sir Cecil Burney. He was second-in-command of the Grand Fleet from 1914 to 1916 and Second Sea Lord from 1916 to 1917. He was succeeded by his only son, the second baronet. He was an aeronautical engineer and Conservative politician. His son, the third baronet, was a businessman and a member of the legislative councils of Rhodesia and Zambia. As of 2007 the title is held by the latter's eldest son, the fourth baronet, who succeeded in 2002.
Burney baronets, of Preston Candover (1921)
Sir Cecil Burney, 1st Baronet (1858–1929)
Sir (Charles) Dennistoun Burney, 2nd Baronet (1888–1968)
Sir Cecil Dennistoun Burney, 3rd Baronet (1923–2002)
Sir Nigel Dennistoun Burney, 4th Baronet (born 1959)
References
www.thepeerage.com
Burney
|
{
"redpajama_set_name": "RedPajamaWikipedia"
}
| 7,173 |
[
128000,
791,
18530,
1216,
54007,
295,
11377,
11,
315,
63325,
70654,
2017,
304,
279,
6406,
315,
62251,
11,
374,
264,
2316,
304,
279,
54007,
295,
425,
315,
279,
3723,
15422,
13,
1102,
574,
3549,
389,
220,
1544,
6186,
220,
5926,
16,
369,
59094,
315,
279,
44555,
17177,
90227,
18530,
1216,
13,
1283,
574,
2132,
3502,
49519,
315,
279,
10517,
44555,
505,
220,
7529,
19,
311,
220,
7529,
21,
323,
10657,
15379,
10425,
505,
220,
7529,
21,
311,
220,
7529,
22,
13,
1283,
574,
26399,
555,
813,
1193,
4538,
11,
279,
2132,
3703,
69786,
13,
1283,
574,
459,
264,
20110,
68267,
24490,
323,
30071,
37038,
13,
5414,
4538,
11,
279,
4948,
3703,
69786,
11,
574,
264,
47378,
323,
264,
4562,
315,
279,
27743,
61783,
315,
57711,
689,
323,
95104
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
791,
18530,
1216,
54007,
295,
11377,
11,
315,
63325,
70654,
2017,
304,
279,
6406,
315,
62251,
11,
374,
264,
2316,
304,
279,
54007,
295,
425,
315,
279,
3723,
15422,
13,
1102,
574,
3549,
389,
220,
1544,
6186,
220,
5926,
16,
369,
59094,
315,
279,
44555,
17177,
90227,
18530,
1216,
13,
1283,
574,
2132,
3502,
49519,
315,
279,
10517,
44555,
505,
220,
7529,
19,
311,
220,
7529,
21,
323,
10657,
15379,
10425,
505,
220,
7529,
21,
311,
220,
7529,
22,
13,
1283,
574,
26399,
555,
813,
1193,
4538,
11,
279,
2132,
3703,
69786,
13,
1283,
574,
459,
264,
20110,
68267,
24490,
323,
30071,
37038,
13,
5414,
4538,
11,
279,
4948,
3703,
69786,
11,
574,
264,
47378,
323,
264,
4562,
315,
279,
27743,
61783,
315,
57711,
689,
323,
95104,
-100
] |
14K Yellow gold Cubic Zirconia Earrings, app. 3 4 ct. tw.
10K Yellow gold Diamond Bentley Pendant Mens Flying B Logo Wing Charm 0.59 Ct.
10K Yellow gold Diamond Pyrex Measuring Glass Cup Pendant 1.80 Charm 2.24 CT.
10K Yellow gold Enhance Cognac Brown Diamond Bridal Wedding Engagement Ring5 8CT – a proud sponsor of the MCM.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 3,011 |
[
128000,
975,
42,
26541,
6761,
18521,
292,
1901,
89376,
689,
469,
52598,
11,
917,
13,
220,
18,
220,
19,
20864,
13,
4483,
627,
605,
42,
26541,
6761,
25328,
74520,
81501,
24157,
47152,
426,
31152,
35622,
58600,
220,
15,
13,
2946,
63379,
627,
605,
42,
26541,
6761,
25328,
5468,
38639,
2206,
69774,
21396,
11098,
81501,
220,
16,
13,
1490,
58600,
220,
17,
13,
1187,
19084,
627,
605,
42,
26541,
6761,
29757,
685,
356,
4021,
582,
10690,
25328,
32437,
278,
31917,
60463,
22249,
20,
220,
23,
1182,
1389,
264,
12691,
25438,
315,
279,
386,
10190,
13,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
] |
[
975,
42,
26541,
6761,
18521,
292,
1901,
89376,
689,
469,
52598,
11,
917,
13,
220,
18,
220,
19,
20864,
13,
4483,
627,
605,
42,
26541,
6761,
25328,
74520,
81501,
24157,
47152,
426,
31152,
35622,
58600,
220,
15,
13,
2946,
63379,
627,
605,
42,
26541,
6761,
25328,
5468,
38639,
2206,
69774,
21396,
11098,
81501,
220,
16,
13,
1490,
58600,
220,
17,
13,
1187,
19084,
627,
605,
42,
26541,
6761,
29757,
685,
356,
4021,
582,
10690,
25328,
32437,
278,
31917,
60463,
22249,
20,
220,
23,
1182,
1389,
264,
12691,
25438,
315,
279,
386,
10190,
13,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100
] |
module FullTextSearch
class UpsertTargetJob < ActiveJob::Base
queue_as :full_text_search
queue_with_priority 15
discard_on ActiveRecord::RecordNotFound
def perform(mapper_class_name, source_id)
mapper_class = mapper_class_name.constantize
source = mapper_class.redmine_class.find(source_id)
mapper = mapper_class.redmine_mapper(source)
mapper.upsert_fts_target
end
end
end
|
{
"redpajama_set_name": "RedPajamaGithub"
}
| 4,656 |
[
128000,
4450,
8797,
1199,
6014,
198,
220,
538,
72416,
531,
6531,
12524,
366,
10106,
12524,
487,
4066,
198,
262,
7325,
12162,
551,
9054,
4424,
10947,
198,
262,
7325,
6753,
39261,
220,
868,
271,
262,
38967,
4570,
18256,
487,
6607,
10588,
271,
262,
711,
2804,
1278,
3183,
4895,
1292,
11,
2592,
851,
340,
415,
25514,
4895,
284,
25514,
4895,
1292,
31130,
553,
198,
415,
2592,
284,
25514,
4895,
17787,
6095,
4895,
2725,
12719,
851,
340,
415,
25514,
284,
25514,
4895,
17787,
6095,
77764,
12719,
340,
415,
25514,
18119,
6175,
42672,
82,
11370,
198,
262,
842,
198,
220,
842,
198,
408,
198,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
] |
[
4450,
8797,
1199,
6014,
198,
220,
538,
72416,
531,
6531,
12524,
366,
10106,
12524,
487,
4066,
198,
262,
7325,
12162,
551,
9054,
4424,
10947,
198,
262,
7325,
6753,
39261,
220,
868,
271,
262,
38967,
4570,
18256,
487,
6607,
10588,
271,
262,
711,
2804,
1278,
3183,
4895,
1292,
11,
2592,
851,
340,
415,
25514,
4895,
284,
25514,
4895,
1292,
31130,
553,
198,
415,
2592,
284,
25514,
4895,
17787,
6095,
4895,
2725,
12719,
851,
340,
415,
25514,
284,
25514,
4895,
17787,
6095,
77764,
12719,
340,
415,
25514,
18119,
6175,
42672,
82,
11370,
198,
262,
842,
198,
220,
842,
198,
408,
198,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100
] |
Over the last few weeks and months, I have been promoting the use of the chatroom. I would love to see the chatroom become as vibrant, fun and dynamic as the forum.
I think it's fair to say that all who dare to enter have a good time.
Let's just say the most likely time to find the room occupied is 8pm onwards.
It's also noticeable that when people drop in and find the room empty, are quickly leaving, not to return. Don't be it off if the room is empty. Stay a while, give others a chance to join you.
The conversation need not be crude. We just have a good laugh and because it's instant we truly are laughing together.
If your nervous just come in and take a seat, you are sure to be welcomed, I don't mind, I don't think others will either. Although it is good etiquette to say "Hi" when you enter.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 416 |
[
128000,
1959,
279,
1566,
2478,
5672,
323,
4038,
11,
358,
617,
1027,
22923,
279,
1005,
315,
279,
6369,
3039,
13,
358,
1053,
3021,
311,
1518,
279,
6369,
3039,
3719,
439,
34076,
11,
2523,
323,
8915,
439,
279,
12111,
627,
40,
1781,
433,
596,
6762,
311,
2019,
430,
682,
889,
33304,
311,
3810,
617,
264,
1695,
892,
627,
10267,
596,
1120,
2019,
279,
1455,
4461,
892,
311,
1505,
279,
3130,
25366,
374,
220,
23,
5298,
60525,
627,
2181,
596,
1101,
43426,
430,
994,
1274,
6068,
304,
323,
1505,
279,
3130,
4384,
11,
527,
6288,
9564,
11,
539,
311,
471,
13,
4418,
956,
387,
433,
1022,
422,
279,
3130,
374,
4384,
13,
29837,
264,
1418,
11,
3041,
3885,
264,
6140,
311,
5249,
499,
627,
791,
10652,
1205,
539,
387,
35750
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
1959,
279,
1566,
2478,
5672,
323,
4038,
11,
358,
617,
1027,
22923,
279,
1005,
315,
279,
6369,
3039,
13,
358,
1053,
3021,
311,
1518,
279,
6369,
3039,
3719,
439,
34076,
11,
2523,
323,
8915,
439,
279,
12111,
627,
40,
1781,
433,
596,
6762,
311,
2019,
430,
682,
889,
33304,
311,
3810,
617,
264,
1695,
892,
627,
10267,
596,
1120,
2019,
279,
1455,
4461,
892,
311,
1505,
279,
3130,
25366,
374,
220,
23,
5298,
60525,
627,
2181,
596,
1101,
43426,
430,
994,
1274,
6068,
304,
323,
1505,
279,
3130,
4384,
11,
527,
6288,
9564,
11,
539,
311,
471,
13,
4418,
956,
387,
433,
1022,
422,
279,
3130,
374,
4384,
13,
29837,
264,
1418,
11,
3041,
3885,
264,
6140,
311,
5249,
499,
627,
791,
10652,
1205,
539,
387,
35750,
-100
] |
After sidelining Gov. Deval Patrick's proposal to make alcohol licensing a purely local responsibility, the House and Senate have approved a few local licensing requests.
The House passed a pair of bills based on legislation proposed by Acton Sen. James Eldridge and Holliston Rep. Carolyn Dykema. One bill (H 4311) specifies a license for Panzano Market in Southborough and the other (H 4312) authorizes a license for Sperry's Country Market Beer and Wine. Both businesses are located on Turnpike Road.
The Senate gave the green light to a bill allowing the town of Easton to grant one more license, which is limited to North Easton Village under a proposal based on legislation filed by Rep. Claire Cronin. The Consumer Protection and Professional Licensure Committee is still weighing a Rep. Kate Hogan bill to grant five alcohol licenses to people operating businesses at Highland Commons in Hudson (H 4313) and a Rep. Jay Barrows bill (H 4361) specifying licenses for the Mansfield Crossing Pouring License Area and the Mansfield Marketplace/The Pavilion Pouring License Area.
The governor in April proposed as part of his economic development bill a plan to place authority for granting liquor licenses in the hands of municipalities, which the governor said would "allow local communities to make responsible decisions regarding their economic development and growth." Patrick is reviewing a redrafted version of the so-called jobs bill that does not contain that proposal.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 2,310 |
[
128000,
6153,
86177,
5859,
25428,
13,
1611,
838,
20199,
596,
14050,
311,
1304,
13200,
28506,
264,
32227,
2254,
12014,
11,
279,
4783,
323,
10092,
617,
12054,
264,
2478,
2254,
28506,
7540,
627,
791,
4783,
5946,
264,
6857,
315,
19123,
3196,
389,
13543,
11223,
555,
3298,
263,
5476,
13,
7957,
27965,
11431,
323,
25518,
59919,
3402,
13,
83358,
43048,
74,
9355,
13,
3861,
4121,
320,
39,
220,
19852,
16,
8,
30202,
264,
5842,
369,
11233,
100098,
8152,
304,
4987,
35179,
323,
279,
1023,
320,
39,
220,
19852,
17,
8,
3229,
4861,
264,
5842,
369,
3165,
5515,
596,
14438,
8152,
34484,
323,
32926,
13,
11995,
9873,
527,
7559,
389,
12268,
65546,
9728,
627,
791,
10092,
6688,
279,
6307,
3177,
311,
264,
4121,
10923,
279,
6424,
315,
6460,
263,
311,
13500
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
6153,
86177,
5859,
25428,
13,
1611,
838,
20199,
596,
14050,
311,
1304,
13200,
28506,
264,
32227,
2254,
12014,
11,
279,
4783,
323,
10092,
617,
12054,
264,
2478,
2254,
28506,
7540,
627,
791,
4783,
5946,
264,
6857,
315,
19123,
3196,
389,
13543,
11223,
555,
3298,
263,
5476,
13,
7957,
27965,
11431,
323,
25518,
59919,
3402,
13,
83358,
43048,
74,
9355,
13,
3861,
4121,
320,
39,
220,
19852,
16,
8,
30202,
264,
5842,
369,
11233,
100098,
8152,
304,
4987,
35179,
323,
279,
1023,
320,
39,
220,
19852,
17,
8,
3229,
4861,
264,
5842,
369,
3165,
5515,
596,
14438,
8152,
34484,
323,
32926,
13,
11995,
9873,
527,
7559,
389,
12268,
65546,
9728,
627,
791,
10092,
6688,
279,
6307,
3177,
311,
264,
4121,
10923,
279,
6424,
315,
6460,
263,
311,
13500,
-100
] |
News Corp, the international media conglomerate spearheaded by Rupert Murdoch, is hiving off its STAR China's TV business to China Media Capital (CMC). In an announcement News Corp mentions of signing an agreement with CMC wherein latter will acquire controlling stake in News Corporation's Xing Kong, Xing Kong International, Channel [V] Mainland China along with its Fortune Star Chinese movie library.
CMC is China's first private equity fund with a focus on investment in the media industry. The company was formed in 2009 and is backed by China's second leading media firm Shanghai Media Group. The agreement with News Corporation represents CMC's first investment project.
Xing Kong and Xing Kong International are general entertainment channels while Channel [V] Mainland Channel is a popular music brand among the youth audiences. The Fortune Star Chinese movie library has a total of 757 Chinese-language titles, a number of which have worldwide perpetual rights. The content reaches Chinese communities across the globe.
The move is indicative of News Corp's gradual movement out of China. During all these years the company has come across several complications in operating in the China's tightly controlled media environment. In 2009 News Corp shifted its scaled down its operations in Hong Kong and shifted focus on operations in the Indian region. The move was seen to be in similar direction at that time.
Other terms of deal and financials were not disclosed.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 304 |
[
128000,
14710,
22621,
11,
279,
6625,
3772,
88751,
349,
41963,
63600,
555,
87269,
79709,
11,
374,
305,
2299,
1022,
1202,
57845,
5734,
596,
6007,
2626,
311,
5734,
7972,
18880,
320,
10190,
34,
570,
763,
459,
17480,
5513,
22621,
34945,
315,
16351,
459,
9306,
449,
356,
11865,
42418,
15629,
690,
21953,
26991,
18783,
304,
5513,
13332,
596,
1630,
287,
18711,
11,
1630,
287,
18711,
7327,
11,
13740,
510,
53,
60,
4802,
1974,
5734,
3235,
449,
1202,
46331,
7834,
8620,
5818,
6875,
627,
10190,
34,
374,
5734,
596,
1176,
879,
25452,
3887,
449,
264,
5357,
389,
9341,
304,
279,
3772,
5064,
13,
578,
2883,
574,
14454,
304,
220,
1049,
24,
323,
374,
22126,
555,
5734,
596,
2132,
6522,
3772,
7626,
38147,
7972,
5856,
13,
578,
9306,
449,
5513,
13332,
11105
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
14710,
22621,
11,
279,
6625,
3772,
88751,
349,
41963,
63600,
555,
87269,
79709,
11,
374,
305,
2299,
1022,
1202,
57845,
5734,
596,
6007,
2626,
311,
5734,
7972,
18880,
320,
10190,
34,
570,
763,
459,
17480,
5513,
22621,
34945,
315,
16351,
459,
9306,
449,
356,
11865,
42418,
15629,
690,
21953,
26991,
18783,
304,
5513,
13332,
596,
1630,
287,
18711,
11,
1630,
287,
18711,
7327,
11,
13740,
510,
53,
60,
4802,
1974,
5734,
3235,
449,
1202,
46331,
7834,
8620,
5818,
6875,
627,
10190,
34,
374,
5734,
596,
1176,
879,
25452,
3887,
449,
264,
5357,
389,
9341,
304,
279,
3772,
5064,
13,
578,
2883,
574,
14454,
304,
220,
1049,
24,
323,
374,
22126,
555,
5734,
596,
2132,
6522,
3772,
7626,
38147,
7972,
5856,
13,
578,
9306,
449,
5513,
13332,
11105,
-100
] |
The Mapmaker Chronicles: Breath of the Dragon
The Mapmaker Chronicles: Breath of the Dragon by A.L.Tait (Lothian/Hachette)
PB RRP $14.99 ISBN 9780734415813
Reviewed by Hilary Smillie
The third and last book in The Mapmaker Chronicles brings to an exciting conclusion the adventures of Verdanian, Quinn Freeman, mapmaker on board the Libertas, one of three ships vying for the prize offered by the king of Verdania for the best map of the world. A.L. Tait has skillfully recapped the previous two books in such a way that children reading Breath of the Dragon will enjoy it as a stand- alone title.
The story opens with Quinn and his fellow voyagers, Zain the captain, Ash his friend who masquerades as a boy, Tomas, and the rest of the crew all weak and close to starvation. The ship reaches a desolate landfall where no tree or sign of life is apparent other than a mountain belching smoke. Despite their lack of confidence, the only way forward is to explore and hope some form of life and therefore provisions will be discovered. A boat is launched from the Libertas and thus new adventures begin.
A strange group of dusty, goblin-like men shovelling out dirt and rocks on the mountain side, stop their work as Quinn, Tomas and Zain appear. They look blank as Zain then Quinn try various languages to communicate with them, but it isn't until Tomas speaks in Prate - the pirate language - that comprehension appears on the mob's faces. Egunon, one of the sand goblins (as Quinn calls them) steps forward and says he is in charge. Tomas introduces Zain as Hayreddin, a compatriot of the notorious pirate, the Golden Serpent, from the Black Hawk where Quinn was once held captive.
The mention of the Golden Serpent has a dramatic effect on Egunon. He quickly offers hospitality. But treachery threatens and the crew are fortunate to escape both the sand goblins and the might of the erupting volcano with its heat and stifling air the strange goblins call "dragon's breath". But escape they do, taking with them provisions and a sack of stones that Quinn thinks valuable and he is soon proved right. The contents of the rocks should give the Libertas extra credit towards earning the king's prize.
With food in their bellies, everyone is strengthened except for the old cleric who is still weak and ill. The cleric is the only one to know Quinn has two maps and the genuine one is hidden in his cabin. But will the confidence held in the true map prove to be misplaced?
There are more challenges ahead before the mapmaker and those on board the Libertas can return home: capture by marauding Deslonders, serious injuries and imprisonment for Zain and clashes with their rivals to name a few. Perhaps one of the more intriguing problems is Quinn struggling with loss of memory recall which is impacting on his capabilities.
I found the depth of the hatred and revenge Quinn feels towards Ira, at whose hands he suffered so much, a shock. Despite a welcome benefit from the fight which ensues, I am not a fan of these emotions and was pleased to find the author balanced this scenario later with the calm of Zain's mature, adult approach and wise influence on Quinn when a great injustice is done to the Libertas' crew. It is good for young readers to see that patience and restraint is preferable to violence.
The author brings her series to a satisfying conclusion with plenty of interest and unexpected events to surprise the reader right up to the last pages. The Mapmaker Chronicles showcases a vivid imagination and gift for storytelling which marries fantasy and reality together to great effect and is bound to be a favourite series on the bookshelf.
Labels: AL Tait, Goblins, Hilary Smillie, Lothian/Hachette, ship, storytelling gift, The Mapmaker Chronicles: Breath of the Dragon
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 5,202 |
[
128000,
791,
5135,
26850,
66416,
25,
58292,
315,
279,
16537,
198,
791,
5135,
26850,
66416,
25,
58292,
315,
279,
16537,
555,
362,
1236,
844,
1339,
320,
4357,
339,
1122,
24240,
613,
6672,
340,
41737,
432,
22394,
400,
975,
13,
1484,
34526,
220,
17272,
25779,
18495,
23864,
18,
198,
80772,
555,
39696,
661,
4487,
484,
648,
198,
791,
4948,
323,
1566,
2363,
304,
578,
5135,
26850,
66416,
12716,
311,
459,
13548,
17102,
279,
32078,
315,
6383,
36255,
1122,
11,
41623,
50664,
11,
2472,
26850,
389,
4580,
279,
52181,
300,
11,
832,
315,
2380,
18198,
348,
7169,
369,
279,
22643,
9076,
555,
279,
11734,
315,
93807,
9345,
369,
279,
1888,
2472,
315,
279,
1917,
13,
362,
1236,
13,
350,
1339,
706,
10151,
3725,
1421,
5795,
279,
3766,
1403,
6603,
304
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
791,
5135,
26850,
66416,
25,
58292,
315,
279,
16537,
198,
791,
5135,
26850,
66416,
25,
58292,
315,
279,
16537,
555,
362,
1236,
844,
1339,
320,
4357,
339,
1122,
24240,
613,
6672,
340,
41737,
432,
22394,
400,
975,
13,
1484,
34526,
220,
17272,
25779,
18495,
23864,
18,
198,
80772,
555,
39696,
661,
4487,
484,
648,
198,
791,
4948,
323,
1566,
2363,
304,
578,
5135,
26850,
66416,
12716,
311,
459,
13548,
17102,
279,
32078,
315,
6383,
36255,
1122,
11,
41623,
50664,
11,
2472,
26850,
389,
4580,
279,
52181,
300,
11,
832,
315,
2380,
18198,
348,
7169,
369,
279,
22643,
9076,
555,
279,
11734,
315,
93807,
9345,
369,
279,
1888,
2472,
315,
279,
1917,
13,
362,
1236,
13,
350,
1339,
706,
10151,
3725,
1421,
5795,
279,
3766,
1403,
6603,
304,
-100
] |
Edge Towers will be a brand new master-planned community comprised of three towers located at the southwest corner of Hurontario St & Elm Drive in Mississauga. Built by Solmar Developments, the buildings are strategically located just steps to Square One Shopping Centre, Sheridan College, the Living Arts Centre, Celebration Square as well as the new LRT which will connect Brampton to the Port Credit GO Train Station.
Edge will be a 3 building community with over 1200 condominium residences that will include the convenience of retail space, a daycare, and an exciting new community green space. The first phase of 293 units will set a new standard in elegant living in Mississauga's prime downtown neighbourhood. Pricing is expected to begin in the mid $300,000s. Live life on the Edge!
Walking distance to the Mississauga-Brampton LRT which has received provincial funding which will run from the Brampton GO all the way down to the Port Credit GO.
Developer: Solmar Developments Walking distance to the Mississauga-Brampton LRT which has received provincial funding which will run from the Brampton GO all the way down to the Port Credit GO. Steps to Square One. Luxury Condos in Mississauga Close to Sheridan College which has gone through a massive expansion that will add an additional 3,200 students to its campus. Total enrollment at Sheridan can now hold up to 5,500 Students.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 5,601 |
[
128000,
11918,
68457,
690,
387,
264,
6883,
502,
7491,
33207,
7443,
4029,
40056,
315,
2380,
40825,
7559,
520,
279,
54085,
9309,
315,
21670,
546,
3370,
800,
612,
65329,
16542,
304,
9083,
1056,
90624,
13,
34154,
555,
11730,
5730,
8000,
1392,
11,
279,
14016,
527,
68774,
7559,
1120,
7504,
311,
15992,
3861,
30064,
14821,
11,
95223,
9304,
11,
279,
19048,
17979,
14821,
11,
75200,
15992,
439,
1664,
439,
279,
502,
445,
5463,
902,
690,
4667,
71884,
19271,
311,
279,
5896,
16666,
12890,
27217,
17040,
627,
11918,
690,
387,
264,
220,
18,
4857,
4029,
449,
927,
220,
4364,
15,
86076,
85871,
430,
690,
2997,
279,
19679,
315,
11040,
3634,
11,
264,
100254,
11,
323,
459,
13548,
502,
4029,
6307,
3634,
13,
578,
1176,
10474,
315,
220,
17313,
8316,
690,
743
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
11918,
68457,
690,
387,
264,
6883,
502,
7491,
33207,
7443,
4029,
40056,
315,
2380,
40825,
7559,
520,
279,
54085,
9309,
315,
21670,
546,
3370,
800,
612,
65329,
16542,
304,
9083,
1056,
90624,
13,
34154,
555,
11730,
5730,
8000,
1392,
11,
279,
14016,
527,
68774,
7559,
1120,
7504,
311,
15992,
3861,
30064,
14821,
11,
95223,
9304,
11,
279,
19048,
17979,
14821,
11,
75200,
15992,
439,
1664,
439,
279,
502,
445,
5463,
902,
690,
4667,
71884,
19271,
311,
279,
5896,
16666,
12890,
27217,
17040,
627,
11918,
690,
387,
264,
220,
18,
4857,
4029,
449,
927,
220,
4364,
15,
86076,
85871,
430,
690,
2997,
279,
19679,
315,
11040,
3634,
11,
264,
100254,
11,
323,
459,
13548,
502,
4029,
6307,
3634,
13,
578,
1176,
10474,
315,
220,
17313,
8316,
690,
743,
-100
] |
<?php
namespace Phrest\SDK\Generator;
use Phrest\SDK\Generator;
abstract class AbstractGenerator implements GeneratorInterface
{
/**
* @var string
*/
protected $version;
/**
* @var string
*/
protected $entityName;
/**
* AbstractGenerator constructor.
*
* @param string $version
* @param string $entityName
*/
public function __construct($version, $entityName)
{
$this->version = $version;
$this->entityName = $entityName;
}
/**
* @param string $version
*
* @return AbstractGenerator
*/
public function setVersion($version)
{
$this->version = $version;
return $this;
}
/**
* @return string
*/
public function getVersion()
{
return $this->version;
}
/**
* @param string $entityName
*
* @return AbstractGenerator
*/
public function setEntityName($entityName)
{
$this->entityName = $entityName;
return $this;
}
/**
* @return string
*/
public function getEntityName()
{
return $this->entityName;
}
}
|
{
"redpajama_set_name": "RedPajamaGithub"
}
| 2,129 |
[
128000,
1340,
1230,
271,
2280,
2405,
4014,
59,
32634,
59,
12846,
401,
817,
2405,
4014,
59,
32634,
59,
12846,
401,
16647,
538,
13822,
12846,
5280,
29458,
5160,
198,
517,
220,
1583,
256,
353,
571,
959,
925,
198,
256,
740,
220,
2682,
400,
4464,
401,
220,
1583,
256,
353,
571,
959,
925,
198,
256,
740,
220,
2682,
400,
3069,
678,
401,
220,
1583,
256,
353,
13822,
12846,
4797,
627,
256,
1235,
256,
353,
571,
913,
925,
400,
4464,
198,
256,
353,
571,
913,
925,
400,
3069,
678,
198,
256,
740,
220,
586,
734,
1328,
7750,
703,
4464,
11,
400,
3069,
678,
340,
220,
341,
262,
400,
576,
405,
4464,
284,
400,
4464,
280,
262,
400,
576,
405,
3069,
678,
284,
400,
3069,
678,
280,
220,
557,
220,
1583,
256
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
1340,
1230,
271,
2280,
2405,
4014,
59,
32634,
59,
12846,
401,
817,
2405,
4014,
59,
32634,
59,
12846,
401,
16647,
538,
13822,
12846,
5280,
29458,
5160,
198,
517,
220,
1583,
256,
353,
571,
959,
925,
198,
256,
740,
220,
2682,
400,
4464,
401,
220,
1583,
256,
353,
571,
959,
925,
198,
256,
740,
220,
2682,
400,
3069,
678,
401,
220,
1583,
256,
353,
13822,
12846,
4797,
627,
256,
1235,
256,
353,
571,
913,
925,
400,
4464,
198,
256,
353,
571,
913,
925,
400,
3069,
678,
198,
256,
740,
220,
586,
734,
1328,
7750,
703,
4464,
11,
400,
3069,
678,
340,
220,
341,
262,
400,
576,
405,
4464,
284,
400,
4464,
280,
262,
400,
576,
405,
3069,
678,
284,
400,
3069,
678,
280,
220,
557,
220,
1583,
256,
-100
] |
目錄列表器允許您列出& 打印文件夾的內容,即創建,然後保存,打印或通過電子郵件發送從硬盤上,CD-ROM,DVD-ROM,軟盤,USB 存儲和網絡共享選定的文件夾中的文件列表。列表可以是 HTML,文本,Microsoft Excel,CSV 格式或直接存儲到數據庫中。目錄列表程序是來自各種目錄打印機的最佳應用程序。目錄列表也可以集成到 Windows 資源管理器的上下文菜單中,所以你甚至不需要打開應用程序來生成列表。命令行界面支持從 Windows Task Scheduler 運行的自動化列表.
目錄列表程序 64 位,您還可以找出給定的目錄大小是什麼,按文件夾大小進行排序並檢查哪些文件夾佔用了磁盤上最多的空間。您還可以使用尺寸過濾器選項在 PC 上找到最大的文件.
目錄 Lister 在以下版本的 Windows 10 中受到 KRKsoft 的支持– Windows 10 Pro,Windows 10 Education 和 Windows 10 Enterprise。目錄列表支持 Windows 10 的支持服務的分支機構,包括 - 當前分支機構,當前分支機構以及以下長期服務分支機構 - Windows 10 Enterprise LTSB.
- Added "Vertical resolution" option and renamed current "Resolution" into "Horizontal resolution"
- Fixed database output to SQL Server when date columns were selected.
Directory Lister Enterprise 2.35 with Patch and Key. 31/01/2019 CRACKSurl ... Directory Lister Pro Installer – 32-bit · Directory Lister Pro Installer – 64-bit.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 6,237 |
[
128000,
30832,
110346,
45277,
32648,
120162,
110008,
88126,
32218,
20834,
5,
126795,
104371,
27996,
8192,
122,
9554,
105134,
37729,
11,
92776,
109832,
26892,
11,
61994,
74482,
47585,
11,
76537,
104371,
58291,
126954,
120115,
31938,
113,
14558,
103106,
37705,
110039,
110567,
114457,
17905,
11,
6620,
99483,
28365,
12757,
99483,
11,
122279,
114457,
11,
27123,
54597,
246,
101268,
110,
34208,
105550,
117139,
55999,
72843,
102647,
118078,
27996,
8192,
122,
105363,
27996,
45277,
1811,
45277,
74770,
21043,
9492,
11,
17161,
22656,
11,
13068,
21705,
11,
45309,
124518,
29430,
58291,
109467,
25359,
101268,
110,
28037,
104214,
114065,
107086,
16325,
1811,
30832,
110346,
45277,
75320,
21043,
102993,
37026,
102208,
104091,
30832,
110346,
76537,
104371,
101513,
9554,
126579,
106877,
11883,
75320,
1811,
30832,
110346,
45277,
75863,
74770,
43167,
13153,
28037,
5632,
110124,
229,
39084
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
30832,
110346,
45277,
32648,
120162,
110008,
88126,
32218,
20834,
5,
126795,
104371,
27996,
8192,
122,
9554,
105134,
37729,
11,
92776,
109832,
26892,
11,
61994,
74482,
47585,
11,
76537,
104371,
58291,
126954,
120115,
31938,
113,
14558,
103106,
37705,
110039,
110567,
114457,
17905,
11,
6620,
99483,
28365,
12757,
99483,
11,
122279,
114457,
11,
27123,
54597,
246,
101268,
110,
34208,
105550,
117139,
55999,
72843,
102647,
118078,
27996,
8192,
122,
105363,
27996,
45277,
1811,
45277,
74770,
21043,
9492,
11,
17161,
22656,
11,
13068,
21705,
11,
45309,
124518,
29430,
58291,
109467,
25359,
101268,
110,
28037,
104214,
114065,
107086,
16325,
1811,
30832,
110346,
45277,
75320,
21043,
102993,
37026,
102208,
104091,
30832,
110346,
76537,
104371,
101513,
9554,
126579,
106877,
11883,
75320,
1811,
30832,
110346,
45277,
75863,
74770,
43167,
13153,
28037,
5632,
110124,
229,
39084,
-100
] |
Tag archive for 'chicken pox'
By Staff On Monday, March 30th, 2015
Second chickenpox case reported by New Hope-Solebury School District
The New Hope-Solebury School District Health Office Monday notified local parents and guardians that a second case of chickenpox has been diagnosed in the district, this one involving a student attending the middle More...
By Staff On Friday, March 13th, 2015
Chickenpox case reported at New Hope-Solebury High School
The New Hope-Solebury School District today emailed a notice to local parents and guardians informing them that "there has been a case of Varicella (Chicken Pox) diagnosed in a student at the High School." The More...
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 2,478 |
[
128000,
5786,
18624,
369,
364,
331,
9890,
3273,
87,
1270,
1383,
17381,
1952,
7159,
11,
5587,
220,
966,
339,
11,
220,
679,
20,
198,
16041,
16553,
79,
5241,
1162,
5068,
555,
1561,
18231,
6354,
1286,
20176,
6150,
11182,
198,
791,
1561,
18231,
6354,
1286,
20176,
6150,
11182,
6401,
8410,
7159,
30316,
2254,
6699,
323,
72881,
430,
264,
2132,
1162,
315,
16553,
79,
5241,
706,
1027,
29704,
304,
279,
9474,
11,
420,
832,
16239,
264,
5575,
24096,
279,
6278,
4497,
9522,
1383,
17381,
1952,
6740,
11,
5587,
220,
1032,
339,
11,
220,
679,
20,
198,
84963,
79,
5241,
1162,
5068,
520,
1561,
18231,
6354,
1286,
20176,
5234,
6150,
198,
791,
1561,
18231,
6354,
1286,
20176,
6150,
11182,
3432,
49614,
264,
5406,
311,
2254,
6699,
323,
72881,
63025,
1124,
430
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
5786,
18624,
369,
364,
331,
9890,
3273,
87,
1270,
1383,
17381,
1952,
7159,
11,
5587,
220,
966,
339,
11,
220,
679,
20,
198,
16041,
16553,
79,
5241,
1162,
5068,
555,
1561,
18231,
6354,
1286,
20176,
6150,
11182,
198,
791,
1561,
18231,
6354,
1286,
20176,
6150,
11182,
6401,
8410,
7159,
30316,
2254,
6699,
323,
72881,
430,
264,
2132,
1162,
315,
16553,
79,
5241,
706,
1027,
29704,
304,
279,
9474,
11,
420,
832,
16239,
264,
5575,
24096,
279,
6278,
4497,
9522,
1383,
17381,
1952,
6740,
11,
5587,
220,
1032,
339,
11,
220,
679,
20,
198,
84963,
79,
5241,
1162,
5068,
520,
1561,
18231,
6354,
1286,
20176,
5234,
6150,
198,
791,
1561,
18231,
6354,
1286,
20176,
6150,
11182,
3432,
49614,
264,
5406,
311,
2254,
6699,
323,
72881,
63025,
1124,
430,
-100
] |
news, national,
ONE of the deadliest climbing seasons on Mount Everest is failing to deter climber Steve Ellery who has the peak on his "to-do list". So far this season, 11 people have died on the world's highest peak with most believed to have suffered from altitude sickness, which is caused by low amounts of oxygen at high elevation. Mr Ellery, from Bathurst, NSW, is an adventurer at heart. Tell him you don't think he can do it and that's the quickest way to help him succeed. He has already climbed the Kokoda Track (2190 metres), Machu Picchu (2430 metres) and his biggest so far is Mount Kilimanjaro (5895 metres). "I've always wanted to climb it," Mr Ellery said of the 8848 m Mount Everest. "If you're worried about dying you'll never do anything, if you're worried about the danger you won't do a thing. "Part of the lure of doing these sorts of things is that there is a little bit of danger to it. It makes it exciting, the fact that there's the potential to die if you've not trained well enough or you're not prepared well enough." Mr Ellery straight up says climbing Kilimanjaro was the hardest thing he has ever done. It took four days to reach base camp and then his group of 17 climbed to the summit in the middle of the night so they could watch the sunrise from the peak. "We were pushing through at minus 17 degrees ... it's like the walking dead," he said. "I passed out going up there, well a few of us passed out, but you've just got to walk slow to keep you heart rate low so you don't use too much oxygen. "The best way to describe it is like trying to breath through a straw. "There was saying on Kilimanjaro called 'poly poly' which means slowly slowly because as soon as you start tying to go too fast you heart rate would increase, your body would use more oxygen and you pass out. "I was delirious, but it was really good, the guys made a promise to each other down the bottom that we were going to get each other to the top, no matter what." Mr Ellery has already set himself a new challenge - Aconcagua in Argentinian. It's the highest peak in South America, at 6962 metres. "Then Everest will be the one after that," he said. A recently released photo of dozens of climbers lined up waiting to reach Everest's summit only serves as a cautionary tale to Mr Ellery to be as prepared as possible. He said delays could be life-threatening when climbing a mountain. "That's the biggest fear, you get stuck on that mountain and a storm comes," he said. "You've only got a certain window of time when you can go up, the weather can change so quick up there. If the weather changed all those people sitting in that queue would die." Despite the risks associated with mountain climbing, Mr Ellery said his sense of satisfaction and addiction to climbing was greater. "You want the next higher one," he said of his search for a higher peak. Western Advocate
https://nnimgt-a.akamaihd.net/transform/v1/crop/frm/yqbYpxNMru7TBX8VR5QF63/c99bf696-0911-40dc-b854-c90ccfb1c40f.jpg/r0_144_960_686_w1200_h678_fmax.jpg
Mount Everest: Deaths of 11 climbers fails to deter Aussie adventurer
Nadine Morton
Mount Everest is still on my to-do list, says adventurer
A picture by climber Nirmal Purja shows heavy traffic of mountain climbers lining up to stand at the summit of Mount Everest. PHOTO: NIRMAL PURJA, PROJECT POSSIBLE
BIG AMBITIONS: Steve Ellery's dream is to one day climb Mount Everest.
SKY HIGH: Steve Ellery was among a group to reach the summit of Mount Kilimanjaro in 2016. Photo: SUPPLIED
SKY HIGH: Steve Ellery photographed the curvature of the Earth from the top of Mount Kilimanjaro in 2016. Photo: SUPPLIED
GREAT VIEW: Steve Ellery was among a group who climbed Machu Picchu in 2014. Photo: SUPPLIED
BROTHERS IN ARMS: Steve Ellery, Ben O'Connor, brothers Marcus and Adam Schembri, and Dean Mobbs during their training session ahead of walking the Kokoda Track. Photo: NADINE MORTON 061717nmkokoda1
PHOTOS: Mountain high's the best place to be challenged
ONE of the deadliest climbing seasons on Mount Everest is failing to deter climber Steve Ellery who has the peak on his "to-do list".
So far this season, 11 people have died on the world's highest peak with most believed to have suffered from altitude sickness, which is caused by low amounts of oxygen at high elevation.
Mr Ellery, from Bathurst, NSW, is an adventurer at heart. Tell him you don't think he can do it and that's the quickest way to help him succeed.
He has already climbed the Kokoda Track (2190 metres), Machu Picchu (2430 metres) and his biggest so far is Mount Kilimanjaro (5895 metres).
"I've always wanted to climb it," Mr Ellery said of the 8848 m Mount Everest.
"If you're worried about dying you'll never do anything, if you're worried about the danger you won't do a thing.
"Part of the lure of doing these sorts of things is that there is a little bit of danger to it. It makes it exciting, the fact that there's the potential to die if you've not trained well enough or you're not prepared well enough."
Mr Ellery straight up says climbing Kilimanjaro was the hardest thing he has ever done.
It took four days to reach base camp and then his group of 17 climbed to the summit in the middle of the night so they could watch the sunrise from the peak.
"We were pushing through at minus 17 degrees ... it's like the walking dead," he said.
"I passed out going up there, well a few of us passed out, but you've just got to walk slow to keep you heart rate low so you don't use too much oxygen.
"The best way to describe it is like trying to breath through a straw.
"There was saying on Kilimanjaro called 'poly poly' which means slowly slowly because as soon as you start tying to go too fast you heart rate would increase, your body would use more oxygen and you pass out.
"I was delirious, but it was really good, the guys made a promise to each other down the bottom that we were going to get each other to the top, no matter what."
You've only got a certain window of time when you can go up, the weather can change so quick up there. If the weather changed all those people sitting in that queue would die.
Steve Ellery
Mr Ellery has already set himself a new challenge - Aconcagua in Argentinian. It's the highest peak in South America, at 6962 metres.
"Then Everest will be the one after that," he said.
A recently released photo of dozens of climbers lined up waiting to reach Everest's summit only serves as a cautionary tale to Mr Ellery to be as prepared as possible.
He said delays could be life-threatening when climbing a mountain.
GREAT VIEW: Steve Ellery (wearing green) was among a group who climbed Machu Picchu in 2014. Photo: SUPPLIED
"That's the biggest fear, you get stuck on that mountain and a storm comes," he said.
"You've only got a certain window of time when you can go up, the weather can change so quick up there. If the weather changed all those people sitting in that queue would die."
Despite the risks associated with mountain climbing, Mr Ellery said his sense of satisfaction and addiction to climbing was greater.
"You want the next higher one," he said of his search for a higher peak.
Western Advocate
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 4,794 |
[
128000,
10189,
11,
5426,
345,
5338,
315,
279,
99469,
30608,
15956,
389,
10640,
87578,
374,
22109,
311,
4130,
11323,
655,
14129,
13852,
727,
889,
706,
279,
16557,
389,
813,
330,
998,
30659,
1160,
3343,
2100,
3117,
420,
3280,
11,
220,
806,
1274,
617,
8636,
389,
279,
1917,
596,
8592,
16557,
449,
1455,
11846,
311,
617,
16654,
505,
36958,
54950,
11,
902,
374,
9057,
555,
3428,
15055,
315,
24463,
520,
1579,
27255,
13,
4491,
13852,
727,
11,
505,
19710,
33712,
11,
39008,
11,
374,
459,
100235,
520,
4851,
13,
25672,
1461,
499,
1541,
956,
1781,
568,
649,
656,
433,
323,
430,
596,
279,
81610,
1648,
311,
1520,
1461,
12265,
13,
1283,
706,
2736,
45519,
279,
65815,
14320,
20371,
320,
13762,
15,
37356,
705,
18798,
84,
26987,
66852,
320,
14052
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
10189,
11,
5426,
345,
5338,
315,
279,
99469,
30608,
15956,
389,
10640,
87578,
374,
22109,
311,
4130,
11323,
655,
14129,
13852,
727,
889,
706,
279,
16557,
389,
813,
330,
998,
30659,
1160,
3343,
2100,
3117,
420,
3280,
11,
220,
806,
1274,
617,
8636,
389,
279,
1917,
596,
8592,
16557,
449,
1455,
11846,
311,
617,
16654,
505,
36958,
54950,
11,
902,
374,
9057,
555,
3428,
15055,
315,
24463,
520,
1579,
27255,
13,
4491,
13852,
727,
11,
505,
19710,
33712,
11,
39008,
11,
374,
459,
100235,
520,
4851,
13,
25672,
1461,
499,
1541,
956,
1781,
568,
649,
656,
433,
323,
430,
596,
279,
81610,
1648,
311,
1520,
1461,
12265,
13,
1283,
706,
2736,
45519,
279,
65815,
14320,
20371,
320,
13762,
15,
37356,
705,
18798,
84,
26987,
66852,
320,
14052,
-100
] |
1708 ABA Journal Illinois articles.
Fired partner is accused of double billing, charging personal expenses to firm
A former Vedder Price partner in Chicago is accused in an ethics complaint of double billing for legal services and charging the firm for personal expenses.
Jan 15, 2020 10:26 AM CST
'Transient' charged in stabbing death of asbestos lawyer who is being hailed as a hero
Police have charged a man described as a "transient" in the home invasion and stabbing death of an Illinois lawyer Saturday.
Jan 7, 2020 12:15 PM CST
Afternoon Briefs: Biden likes Obama for Supreme Court; ex-law student among successful GirlsDoPorn plaintiffs
Biden would consider Obama as Supreme Court nominee
Democratic presidential candidate and former Vice President Joe Biden said Dec. 28 he would appoint former President Barack Obama to the U.S.…
Jan 6, 2020 4:32 PM CST
First Amendment defense claims could threaten 'revenge pornography' statutes
Dec 19, 2019 6:30 AM CST
Lawyer who allowed client to write appeals brief is sanctioned for 'incoherent' document
A federal appeals court has ordered an Aurora, Illinois, solo practitioner to pay attorney fees and double costs to his opponent for allowing his client to write and submit an "incoherent" appellate brief using his electronic signature.
Dec 18, 2019 2:22 PM CST
How do the states stack up? Lawyers make the most and least money in these areas
Lawyers make the lowest average salary in Montana and the highest average salary in California, according to data from the U.S. Bureau of Labor Statistics.
Nov 20, 2019 10:55 AM CST
Former FBI Director Comey encourages law students to be 'ethical leaders'
It was never one of former FBI Director James Comey's career goals to be an "unemployed B-list celebrity," he said, addressing a crowd at the University of Chicago Law School on Tuesday. During a lecture he gave on leadership roles, he said it became his mission to communicate what being an ethical leader means after President Donald Trump fired him in May 2017.
Oct 31, 2019 12:40 PM CDT
Federal judge vacates $16M settlement after parties claim press release violated secrecy provision
A federal judge in Chicago has tossed a $16 million settlement between Kraft Foods Group Inc. and the U.S. Commodity Futures Trading Commission after Kraft Foods claimed that the CFTC's public statements about the deal violated its confidentiality provision.
Judge is removed from the bench for false statements about firing gun, retaliation
An Illinois judge has been removed from the bench based on accusations that he lied about accidentally firing a gun in his apartment and retaliated against two court employees who complained about sexual harassment.
Oct 2, 2019 11:34 AM CDT
Police officers subpoena former governor to testify about commuted sentence
Lawyers for 11 Chicago police officers have subpoenaed former Illinois Gov. Pat Quinn to testify about his reasons for commuting the sentence of a man who has sued them for allegedly railroading the case against him.
Sep 24, 2019 10:31 AM CDT
Lawyer accused of smelling of alcohol at client's felony trial gets suspension
An Ohio lawyer has been suspended after he was accused of showing up for a client's felony trial smelling of alcohol and refusing to submit to a breath-alcohol test.
Federal judge tosses suit by gay lawyer who says his co-counsel outed him to al-Qaida client
A federal judge in Chicago has ruled that a mitigation specialist who worked on the defense team for accused Sept. 11 mastermind Khalid Sheikh Mohammed can't sue a co-counsel accused of outing him as a gay man to the client.
Sep 17, 2019 2:05 PM CDT
Want encouragement in the justice arena? Think about what's changed, says Justice Ruth Bader Ginsburg
The country has come a long way, said U.S. Supreme Court Justice Ruth Bader Ginsburg, and that's what keeps her optimistic for the future. "It can be hard to do anything as a loner, but if you get together with like-minded people, you can be a force for change," Ginsburg said at a discussion Monday at the University of Chicago.
Suspended lawyer can't shield home in bankruptcy from malpractice judgment, 7th Circuit says
A suspended Illinois lawyer who had owned his home as a tenant by the entirety before his wife's death won't be able to shield the property in bankruptcy from a former client who won a default malpractice judgment.
Aug 6, 2019 10:13 AM CDT
In-House Ethics: Top state court says ethics rules haven't 'evolved with the profession' and lawyer-client relationships
A fired in-house lawyer may sue his former employer for breach of contract and violation of public policy, despite an ethics rule that "robustly protects" a client's right to terminate a lawyer at any time, for any reason, the Washington Supreme Court has ruled in a 6-3 decision.
Jul 25, 2019 3:31 PM CDT
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 8,444 |
[
128000,
8258,
23,
362,
7209,
10139,
19174,
9908,
627,
37,
2757,
8427,
374,
13487,
315,
2033,
34631,
11,
23468,
4443,
18512,
311,
7626,
198,
32,
4846,
75887,
1126,
8650,
8427,
304,
10780,
374,
13487,
304,
459,
32008,
12458,
315,
2033,
34631,
369,
5897,
3600,
323,
23468,
279,
7626,
369,
4443,
18512,
627,
18820,
220,
868,
11,
220,
2366,
15,
220,
605,
25,
1627,
6912,
60039,
198,
17773,
34489,
1188,
6,
11684,
304,
72022,
4648,
315,
77235,
15779,
889,
374,
1694,
64895,
439,
264,
12084,
198,
22993,
617,
11684,
264,
893,
7633,
439,
264,
330,
1485,
1188,
1,
304,
279,
2162,
30215,
323,
72022,
4648,
315,
459,
19174,
15779,
7884,
627,
18820,
220,
22,
11,
220,
2366,
15,
220,
717,
25,
868,
5975,
60039,
198,
6153,
12684,
37618,
82
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
8258,
23,
362,
7209,
10139,
19174,
9908,
627,
37,
2757,
8427,
374,
13487,
315,
2033,
34631,
11,
23468,
4443,
18512,
311,
7626,
198,
32,
4846,
75887,
1126,
8650,
8427,
304,
10780,
374,
13487,
304,
459,
32008,
12458,
315,
2033,
34631,
369,
5897,
3600,
323,
23468,
279,
7626,
369,
4443,
18512,
627,
18820,
220,
868,
11,
220,
2366,
15,
220,
605,
25,
1627,
6912,
60039,
198,
17773,
34489,
1188,
6,
11684,
304,
72022,
4648,
315,
77235,
15779,
889,
374,
1694,
64895,
439,
264,
12084,
198,
22993,
617,
11684,
264,
893,
7633,
439,
264,
330,
1485,
1188,
1,
304,
279,
2162,
30215,
323,
72022,
4648,
315,
459,
19174,
15779,
7884,
627,
18820,
220,
22,
11,
220,
2366,
15,
220,
717,
25,
868,
5975,
60039,
198,
6153,
12684,
37618,
82,
-100
] |
Mark Pantoni still fueling Ohio State's high-octane recruiting machine https://t.co/mkHs21gAa3
— Lettermen Row (@LettermenRow) July 17, 2019
B🅾️🅾️🅾️🅾️🅾️🅾️MMMMMM!!!!! pic.twitter.com/vyFmsCdwFH
— Mark Pantoni (@markpantoni) October 13, 2019
B🅾️🅾️🅾️🅾️🅾️MMMMM!!!! pic.twitter.com/pWN88CziJU
OSUBasketballJunkie, Nov 1, 2019
Mark Pantoni spent about 40 minutes talking Ohio State recruiting this morning. Everything he hit on today: https://t.co/tQcmgcDU7E
— Colin Hass-Hill (@chasshill) April 3, 2020
Ohio State continues to benefit from the decision to keep the infrastructure Mickey Marotti and Mark Pantoni helped build when Ryan Day took over. https://t.co/euMT3dOnib pic.twitter.com/StoqsqgxMW
— Eleven Warriors (@11W) June 16, 2020
PLAY BALL‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️‼️
— Mark Pantoni (@markpantoni) September 16, 2020
OSUBasketballJunkie, Sep 16, 2020
MOOD... pic.twitter.com/a2u8hLAcb3
— Mark Pantoni (@markpantoni) November 20, 2020
OSUBasketballJunkie, Nov 19, 2020
MOOD‼️‼️‼️ pic.twitter.com/u4l7msAOnk
B🅾️🅾️🅾️🅾️🅾️MMMMM‼️‼️‼️ pic.twitter.com/p3b6vyiaXb
Back to Business...
Then (2012) Now pic.twitter.com/YNWuqSXRoZ
— Mark Pantoni (@markpantoni) December 4, 2020
OSUBasketballJunkie, Dec 4, 2020
B🅾️🅾️🅾️🅾️🅾️🅾️🅾️🅾️MMM‼️‼️‼️ pic.twitter.com/fmpuEbYA6h
— Mark Pantoni (@markpantoni) February 28, 2021
First day of spring practice. Let's go‼️ pic.twitter.com/LEnhwCKiH0
WEEKEND MOOD‼️‼️ pic.twitter.com/TkMuE8NJTi
— Mark Pantoni (@markpantoni) June 4, 2021
B🅾️🅾️🅾️🅾️🅾️🅾️MMMM‼️‼️‼️‼️ pic.twitter.com/Toqb6ibyjt
Fun times yesterday 😎 pic.twitter.com/nzJoiqmjbS
The Mark Pantoni effect: How the recruiting chief keeps Ohio State football talent rich https://t.co/ufIQM04fhg
— BuckeyeXtra.com (@BuckeyeXtra) August 29, 2021
Spending signing day watching film of players in the transfer portal…
Welcome to college football 2021 😏
— Mark Pantoni (@markpantoni) December 15, 2021
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 9,733 |
[
128000,
9126,
54222,
21446,
2103,
10633,
287,
14689,
3314,
596,
1579,
16405,
302,
2194,
32959,
5780,
3788,
1129,
83,
6973,
3262,
74,
39,
82,
1691,
70,
32,
64,
18,
198,
2345,
6914,
5098,
268,
11035,
8165,
10267,
5098,
268,
3179,
8,
5887,
220,
1114,
11,
220,
679,
24,
198,
33,
9468,
227,
122,
31643,
9468,
227,
122,
31643,
9468,
227,
122,
31643,
9468,
227,
122,
31643,
9468,
227,
122,
31643,
9468,
227,
122,
31643,
73402,
8195,
70900,
10532,
16535,
916,
5574,
88,
37,
1026,
34,
30506,
84414,
198,
2345,
4488,
54222,
21446,
8165,
4075,
79,
519,
21446,
8,
6664,
220,
1032,
11,
220,
679,
24,
198,
33,
9468,
227,
122,
31643,
9468,
227,
122,
31643,
9468,
227,
122,
31643,
9468,
227,
122,
31643,
9468,
227,
122,
31643,
73402
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
9126,
54222,
21446,
2103,
10633,
287,
14689,
3314,
596,
1579,
16405,
302,
2194,
32959,
5780,
3788,
1129,
83,
6973,
3262,
74,
39,
82,
1691,
70,
32,
64,
18,
198,
2345,
6914,
5098,
268,
11035,
8165,
10267,
5098,
268,
3179,
8,
5887,
220,
1114,
11,
220,
679,
24,
198,
33,
9468,
227,
122,
31643,
9468,
227,
122,
31643,
9468,
227,
122,
31643,
9468,
227,
122,
31643,
9468,
227,
122,
31643,
9468,
227,
122,
31643,
73402,
8195,
70900,
10532,
16535,
916,
5574,
88,
37,
1026,
34,
30506,
84414,
198,
2345,
4488,
54222,
21446,
8165,
4075,
79,
519,
21446,
8,
6664,
220,
1032,
11,
220,
679,
24,
198,
33,
9468,
227,
122,
31643,
9468,
227,
122,
31643,
9468,
227,
122,
31643,
9468,
227,
122,
31643,
9468,
227,
122,
31643,
73402,
-100
] |
January is nobody's favourite month. I don't know about where you are, but here in Scotland it's cold, it's dark, and our wallets are now as empty as our Christmas stockings. So, if you're as skint as I am, feast your eyes on the riches of this Top 10 (plus bonus #11) list I made, just for you. They're all brilliant, and they're all free. The library. New books, old books, classic books, cult books, books by authors you follow on Twitter….
At BookTalk, I discussed death, autobiography, licking batteries, and what makes a fairytale.
Do something nice for yourself today: take some time to look at these photos from Russia in the early 1900s and listen to Leonard Cohen's new album, 'Old Ideas'. Perfect for a Sunday afternoon.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 6,811 |
[
128000,
33327,
374,
19093,
596,
19214,
2305,
13,
358,
1541,
956,
1440,
922,
1405,
499,
527,
11,
719,
1618,
304,
19627,
433,
596,
9439,
11,
433,
596,
6453,
11,
323,
1057,
54408,
527,
1457,
439,
4384,
439,
1057,
10280,
76705,
13,
2100,
11,
422,
499,
2351,
439,
1940,
396,
439,
358,
1097,
11,
53268,
701,
6548,
389,
279,
70934,
315,
420,
7054,
220,
605,
320,
7284,
12306,
674,
806,
8,
1160,
358,
1903,
11,
1120,
369,
499,
13,
2435,
2351,
682,
20333,
11,
323,
814,
2351,
682,
1949,
13,
578,
6875,
13,
1561,
6603,
11,
2362,
6603,
11,
11670,
6603,
11,
4612,
6603,
11,
6603,
555,
12283,
499,
1833,
389,
6405,
1981,
627,
1688,
6017,
37310,
11,
358,
14407,
4648,
11,
91537,
11,
67231,
27360,
11,
323,
1148
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
33327,
374,
19093,
596,
19214,
2305,
13,
358,
1541,
956,
1440,
922,
1405,
499,
527,
11,
719,
1618,
304,
19627,
433,
596,
9439,
11,
433,
596,
6453,
11,
323,
1057,
54408,
527,
1457,
439,
4384,
439,
1057,
10280,
76705,
13,
2100,
11,
422,
499,
2351,
439,
1940,
396,
439,
358,
1097,
11,
53268,
701,
6548,
389,
279,
70934,
315,
420,
7054,
220,
605,
320,
7284,
12306,
674,
806,
8,
1160,
358,
1903,
11,
1120,
369,
499,
13,
2435,
2351,
682,
20333,
11,
323,
814,
2351,
682,
1949,
13,
578,
6875,
13,
1561,
6603,
11,
2362,
6603,
11,
11670,
6603,
11,
4612,
6603,
11,
6603,
555,
12283,
499,
1833,
389,
6405,
1981,
627,
1688,
6017,
37310,
11,
358,
14407,
4648,
11,
91537,
11,
67231,
27360,
11,
323,
1148,
-100
] |
Magazine Article (1)
(-) Andrew Gordon (1)
(-) Mitsuo Fuchida and Masatake Okumiya Foreword by Raymond A Spruance (1)
The Rules of the Game
Jutland and British Naval Command
By Andrew Gordon
Foreword by Admiral Sir John Woodward. When published in hardcover in 1997, this book was praised for providing an engrossing education not only in naval strategy and tactics but in Victorian social attitudes and the influence of character on history. In juxtaposing an operational with a cultural theme, the author comes closer than any historian yet to explaining what was ...
The Battle That Doomed Japan, the Japanese Navy's Story
By Mitsuo Fuchida and Masatake Okumiya; Foreword by Raymond A. Spruance
This landmark study was first published in English by the Naval Institute in 1955 and was added to the Classics of Naval Literature series in 1992. Widely acknowledged for its valuable Japanese insights into the battle that turned that tide of war in the Pacific, the book has made a great impact on American readers over the years. Two Japanese ...
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 9,686 |
[
128000,
34015,
10119,
13659,
320,
16,
340,
4172,
8,
13929,
26952,
320,
16,
340,
4172,
8,
60676,
24012,
435,
1412,
4849,
323,
20459,
460,
441,
7777,
25330,
7911,
8371,
1178,
555,
43670,
362,
15883,
84,
685,
320,
16,
340,
791,
23694,
315,
279,
4140,
198,
41,
332,
1974,
323,
8013,
46829,
7498,
198,
1383,
13929,
26952,
198,
11339,
1178,
555,
59094,
17177,
3842,
87218,
13,
3277,
4756,
304,
2653,
3773,
304,
220,
2550,
22,
11,
420,
2363,
574,
37475,
369,
8405,
459,
2995,
2177,
287,
6873,
539,
1193,
304,
46398,
8446,
323,
26411,
719,
304,
44277,
3674,
33726,
323,
279,
10383,
315,
3752,
389,
3925,
13,
763,
98953,
14759,
459,
25605,
449,
264,
13042,
7057,
11,
279,
3229,
4131,
12401,
1109,
904,
44068,
3686,
311,
26073,
1148,
574
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
34015,
10119,
13659,
320,
16,
340,
4172,
8,
13929,
26952,
320,
16,
340,
4172,
8,
60676,
24012,
435,
1412,
4849,
323,
20459,
460,
441,
7777,
25330,
7911,
8371,
1178,
555,
43670,
362,
15883,
84,
685,
320,
16,
340,
791,
23694,
315,
279,
4140,
198,
41,
332,
1974,
323,
8013,
46829,
7498,
198,
1383,
13929,
26952,
198,
11339,
1178,
555,
59094,
17177,
3842,
87218,
13,
3277,
4756,
304,
2653,
3773,
304,
220,
2550,
22,
11,
420,
2363,
574,
37475,
369,
8405,
459,
2995,
2177,
287,
6873,
539,
1193,
304,
46398,
8446,
323,
26411,
719,
304,
44277,
3674,
33726,
323,
279,
10383,
315,
3752,
389,
3925,
13,
763,
98953,
14759,
459,
25605,
449,
264,
13042,
7057,
11,
279,
3229,
4131,
12401,
1109,
904,
44068,
3686,
311,
26073,
1148,
574,
-100
] |
On the evening of Wednesday 6th March, PCS Dance & Drama took 45 students to watch Chicago School edition at the Old Market Theatre in Hove.
This was performed by Brighton Theatre Group-Youth, and it was a phenomenal performance. They have children from 11-18 within their company and they all performed their characters exceptionally.
We also had three current students in the play: James, last seen playing Danny in the PCS performance of Grease, performing the lead role of Billy Flynn. James did a fantastic job in this role showing off his skills in acting, singing and dancing.
Bailey,who also appeared in Grease as Kenickie, showed his versatility playing the other male lead: sorrowful, loyal Amos Hart gaining sympathy 'awws' from the audience, and Evie Turner, who played Frenchie in the PCS production of Grease, was one of the dancers performing quick and challenging dances.
It was also fantastic to see many of our ex-students (Beth, Emma, Beth) who also appeared in featured roles. The whole show had such professionalism, with high production values all round. Their next show is Chitty Chitty Bang Bang which will be performed at The Theatre Royal in August.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 9,225 |
[
128000,
1966,
279,
11714,
315,
8079,
220,
21,
339,
5587,
11,
92197,
30704,
612,
47569,
3952,
220,
1774,
4236,
311,
3821,
10780,
6150,
14002,
520,
279,
10846,
8152,
27315,
304,
473,
1009,
627,
2028,
574,
10887,
555,
58519,
27315,
5856,
12,
2675,
339,
11,
323,
433,
574,
264,
57492,
5178,
13,
2435,
617,
2911,
505,
220,
806,
12,
972,
2949,
872,
2883,
323,
814,
682,
10887,
872,
5885,
48298,
627,
1687,
1101,
1047,
2380,
1510,
4236,
304,
279,
1514,
25,
7957,
11,
1566,
3970,
5737,
33699,
304,
279,
92197,
5178,
315,
13842,
521,
11,
16785,
279,
3063,
3560,
315,
33919,
40486,
13,
7957,
1550,
264,
14964,
2683,
304,
420,
3560,
9204,
1022,
813,
7512,
304,
15718,
11,
26139,
323,
29066,
627,
59927,
18280,
11,
14965,
1101,
9922,
304
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
1966,
279,
11714,
315,
8079,
220,
21,
339,
5587,
11,
92197,
30704,
612,
47569,
3952,
220,
1774,
4236,
311,
3821,
10780,
6150,
14002,
520,
279,
10846,
8152,
27315,
304,
473,
1009,
627,
2028,
574,
10887,
555,
58519,
27315,
5856,
12,
2675,
339,
11,
323,
433,
574,
264,
57492,
5178,
13,
2435,
617,
2911,
505,
220,
806,
12,
972,
2949,
872,
2883,
323,
814,
682,
10887,
872,
5885,
48298,
627,
1687,
1101,
1047,
2380,
1510,
4236,
304,
279,
1514,
25,
7957,
11,
1566,
3970,
5737,
33699,
304,
279,
92197,
5178,
315,
13842,
521,
11,
16785,
279,
3063,
3560,
315,
33919,
40486,
13,
7957,
1550,
264,
14964,
2683,
304,
420,
3560,
9204,
1022,
813,
7512,
304,
15718,
11,
26139,
323,
29066,
627,
59927,
18280,
11,
14965,
1101,
9922,
304,
-100
] |
It's no secret that Cambodia is a highly dollarised country, with certain estimates suggesting a more than 80 per cent penetration of the greenback. A large section of FDI and the urban economy in the Kingdom is fuelled by the US dollar, with little indication that things are going to change soon.
While economists believe that the current dollar rally, which started in late 2013, may not accelerate at the pace it has in the past 18 months, it is expected to continue to strengthen this year. The dollar has seen a 15 per cent appreciation against a basket of foreign currencies, including the euro and yen, in the past six months, with only the euro showing some resilience in the past few days.
Cambodia, while expected to be largely unaffected in the short term, could see some downsides to this trend in the long run.
Grant Knuckey, CEO of ANZ Royal, said that the country will be shielded from this appreciation in some respects, but there can be pressure on agricultural exports, especially to Europe, because a depreciating euro will make exports costlier to the economic bloc.
"For exports to Europe – mainly rice and garments – there can be an impact, although the effect may be lagged depending on the term of the export contract," said Knuckey.
He added that another problem area could be tourism, where depreciation in local currencies makes it more expensive for European and Asian travellers, who currently make up the bulk of tourists to the Kingdom.
This decrease in export competitiveness also poses a future risk to garments, which account for around 70 per cent of outbound shipments, said Hana Becickova, senior economist for the Mekong region at Mantis BV, a risk quantification and forecasting firm in the Netherlands.
Nations like Thailand and Vietnam, whose currencies are floated, have the option of adjusting their currencies through monetary tools, where as Cambodia cannot because it is so strongly pegged to the US dollar.
According to Becickova, the wage hike last year was a "good thing", but given the current economic situation, it puts added pressure on Cambodia's competitiveness, particularly when the Kingdom is unable to use monetary policy to mitigate the rising cost of exports.
"Cambodia currently has really a higher wage level, in nominal terms, compared to other garment manufacturers within the region," she said. "Garment wages is not the only thing that matters for garment investors. When we look at electricity prices, Cambodia is one of the costlier countries in terms of electricity. So again this is something not really favourable for the garment sector," Becickova added.
On the upside, Cambodia will not have to grapple with local currency depreciation, asset volatility and higher priced imports, given that the country is a net importer.
Hiroshi Suzuki, CEO at the Business Research Institute for Cambodia, said one unintended benefit is repayment of loans made in other currencies, because an appreciation in the dollar makes these repayment amounts smaller.
"In case of Japanese ODA loans, Cambodian government could enjoy big savings in the repayment amount in US dollar because all the loans are denominated in Japanese yen and the yen has been depreciated around 40 per cent within the two years," Suzuki said.
Looking at the region, Suzuki feels that Asian economies are better prepared this time around to deal with a cyclical dollar appreciation, with Japan enjoying increased export competitiveness and better returns on overseas investment.
"Many ASEAN countries have not been affected so much because they are mainly using US dollar for the foreign trade currency," said Suzuki.
While the eurozone and Japan have extended quantitative easing, and other nations are exercising the option of adjusting their currencies to maintain competitiveness, these options are not available to Cambodia. With the high rate of dollarisation, Cambodia has little control over its monetary policy and has to rely on the US Federal Reserve to set the dollar's path.
In the past, the government and National Bank of Cambodia have spoken of carrying out de-dollarisation, but the process has been slow and piecemeal.
Mey Kalyan, senior adviser to the Supreme National Economic Council, said that the government has a policy for de-dollarisation, but it hasn't been implemented fast enough.
He added that as the economy gets stronger the government will have to grapple with the right amount of regulatory touch it needs to put in place, while still maintaining free-market principles.
"I think looking back, using the dollar was good for stability – macro stability. But now it is time to de-dollarise as the economy gets stronger and bigger. The problem is not so much about growth anymore," Kalyan said.
He added that Cambodia needs to address this issue with an action plan that moves towards greater local monetary control. "It's like flying a plane with one engine. If the engine is off it can go down," he said.
Today's Stories March 30 , 2015 Subsribe Now !
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 4,537 |
[
128000,
2181,
596,
912,
6367,
430,
62138,
374,
264,
7701,
18160,
4147,
3224,
11,
449,
3738,
17989,
23377,
264,
810,
1109,
220,
1490,
824,
2960,
44596,
315,
279,
6307,
1445,
13,
362,
3544,
3857,
315,
435,
18091,
323,
279,
16036,
8752,
304,
279,
15422,
374,
10633,
839,
555,
279,
2326,
18160,
11,
449,
2697,
28137,
430,
2574,
527,
2133,
311,
2349,
5246,
627,
8142,
45415,
4510,
430,
279,
1510,
18160,
19768,
11,
902,
3940,
304,
3389,
220,
679,
18,
11,
1253,
539,
43880,
520,
279,
18338,
433,
706,
304,
279,
3347,
220,
972,
4038,
11,
433,
374,
3685,
311,
3136,
311,
20259,
420,
1060,
13,
578,
18160,
706,
3970,
264,
220,
868,
824,
2960,
35996,
2403,
264,
14351,
315,
7362,
36702,
11,
2737,
279,
18140,
323,
58440,
11
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
2181,
596,
912,
6367,
430,
62138,
374,
264,
7701,
18160,
4147,
3224,
11,
449,
3738,
17989,
23377,
264,
810,
1109,
220,
1490,
824,
2960,
44596,
315,
279,
6307,
1445,
13,
362,
3544,
3857,
315,
435,
18091,
323,
279,
16036,
8752,
304,
279,
15422,
374,
10633,
839,
555,
279,
2326,
18160,
11,
449,
2697,
28137,
430,
2574,
527,
2133,
311,
2349,
5246,
627,
8142,
45415,
4510,
430,
279,
1510,
18160,
19768,
11,
902,
3940,
304,
3389,
220,
679,
18,
11,
1253,
539,
43880,
520,
279,
18338,
433,
706,
304,
279,
3347,
220,
972,
4038,
11,
433,
374,
3685,
311,
3136,
311,
20259,
420,
1060,
13,
578,
18160,
706,
3970,
264,
220,
868,
824,
2960,
35996,
2403,
264,
14351,
315,
7362,
36702,
11,
2737,
279,
18140,
323,
58440,
11,
-100
] |
Day 2 – WTM London, the business continues
In Press Releases, WTM London.
The second day of WTM London 2021 saw a busy agenda of business meetings, speed networking, high-level conferences, awards and press briefings.
One of the highlights was the annual ministers' summit, held in partnership with the UNWTO and WTTC.
Tourism ministers from around the world pledged to work towards the recovery of the sector with a focus on investing in sustainable travel.
Meanwhile, the International Tourism and Investment Conference at WTM – dubbed Invest, Finance and Restart – continued into its second day at Excel, building on the progress made on Monday.
Elsewhere on the exhibition floor, tourist boards from destinations such as Saudi Arabia, Jordan, Slovenia and the Maldives updated the international media about their countries' tourism strategies.
UNWTO, WTTC & WTM Ministers' Summit
Investing in Tourism's Sustainable Future was the agenda for ministers from countries including the UK, Italy, Portugal, Sierra Leone, Egypt, Jordan and the Philippines.
Moderated by the BBC's Zeinab Badawi, the summit also heard from Zurab Pololikashvili, Secretary-General at UNWTO, and Julia Simpson, WTTC Chief Executive, plus representatives from the private sector, such as the bosses of the Radisson Hotel Group and airport technology firm SITA.
The UNWTO will also partner with Arabian Travel Market – the sister event of WTM London – which will take place on 8-11 May in Dubai.Awards and Competition
There were two WTM awards presentations – the WTM Responsible Tourism Awards, celebrating best practice across the globe, and the WTM London 2021 Best Stand Awards.
The Responsible Tourism Awards honoured winners from countries such as India, the Maldives and South Africa. Meanwhile, destinations such as the Canary Islands, Ireland and Saudi Arabia among the winners of the Best Stand Awards.
Elsewhere, at Travel Forward – the travel technology event that is co-located with WTM London – the winners of the Front Runners Travel Tech Competition for start-ups were revealed as: hotel rebooking specialists from Israel, Pruvo and UK-based Hotelmize; Nepal-based Community Homestay Network, which creates jobs for locals by offering and promoting authentic local experiences; and Motiontag, a German mobility analytics company helping to fight the climate emergency by providing green incentive programmes.
(Front Runners Competition – Dedicated press release attached)
Agile travel companies look to the future
The travel industry is past the worst, a leading analyst told delegates.
During a session entitled: Fixing Travel's Broken Business Model, David Goodger MD Europe and Middle East at consultancy Oxford Economics, said: "We are optimistic that we are past the worst.
"The economy is rebounding. It's a consumer-led boom that we are looking at. People have had forced savings and they are spending it on travel. The year ahead, we do expect to see much stronger activity."
Fellow panellist Sophie Dekkers, Chief Commercial Officer at easyJet, said the airline "will come out stronger" from the pandemic, but added: "But it has been painful. We have had to be much more agile."
Currently, 70% of easyJet's bookings are for travel within the next three months and the airline is hoping for a favourable ski season.
"Fingers crossed for snow as skiers are desperate to go skiing," she said.
Adam Armstrong, CEO of Contiki – the adventure holidays and group travel specialist –also said the operator had had to be "agile in our products" as Covid rules had meant it has so far been unable to do multi-centre trips.
But he said more flexibility and pay options meant "there has never been a better time for a consumer to buy a travel product", adding: "I don't think that will ever change."
Meanwhile, Advantage Travel Partnership Chief Executive Julia Lo Bue-Said said there will "unfortunately be some more failures" in the industry.
"We have still got a tough winter to get through. You can't undo 20 months of devastation," she said.
However, she said the recently announced reopening of travel to the US "has dramatically changed our booking patterns overnight".
TikTok as a travel marketing tool
Travel companies can access a suite of advertising products on the TikTok platform from as little as £50 a day, Hannah Bennett, Industry Lead Travel Brand Partnership at the company told WTM delegates.
In a session entitled Is TikTok the Best Platform for Travel Brands and Content Creators?, she described Tik Tok as an entertainment platform rather than a social platform and said its mission is to inspire creativity. In September, the platform hit a record one billion users.
Bennett attributed some of the growth to the lockdown, saying parents, siblings and grandparents all got involved.
Google Stats
One in four consumers spend more time researching holidays than they did before the pandemic, but the majority of searches are for late bookings, according to Google.
The company's Agency Development Manager, Jad Khalil, said 24% of consumers are spending more time researching because of Covid and currently, 60% of domestic searches on Expedia are for travel within 0-21 days.
He said there is travel optimism because of the vaccine rollout but remaining uncertainty over new Covid variants. This has led to a rise in trips closer to home, a trend towards workcations and a boost in digital marketing.
Meanwhile, Steve Hyde, CEO of digital ad agency Push Group, said 70% total ad spend is on digital.
"If you are not aware, it's here, it's growing, it's speeding up," he said. "If you do the math, digital is good value compared to other forms of media" he said, but added that firms have not migrated away as quickly as the consumers have gone online.
Jordan Kingdom of Time – Press Conference
Jordan used WTM to reveal a new marketing brand, Jordan Kingdom of Time.
The slogan was announced during a press conference in which WTM delegates heard that income from overseas tourists dropped from £4.5bn in 2019 to £1.3bn in 2020.
Representatives said the country is "working together to send the right message that we are ready and welcoming our wonderful travellers".
Slovenia Press Conference
An eight-day ski touring route has opened in Slovenia's Julian Alps region, which encourages confident skiers to explore undeveloped areas, in keeping with the destination's plan to expand its sustainable tourism offering.
Klemen Langus, Director of Tourism Bottinj, one of the tourist boards in the Julian Alps region, said: "Everything is ready. We are just waiting for the snow to come.
Ski touring is quite developed in Slovakia and want to influence visitors to follow our sustainable vision while enjoying Triglav National Park on the Juliana SkiTour."
Maldives Press Conference
Destination aims to be leading meetings and events destination.
The Maldives unveiled a new initiative to become a leading meetings and events destination.
The Indian Ocean nation will also offer homestay holidays from January 2022 as part of a strategy to diversify its offering to visitors.
Mohamed Thoyyib, Managing Director of Maldives Marketing and Public Relations Corporation (MMPRC), the national tourism promotion board, showed a video entitled 'Boardroom on the beach', showcasing the range of activities available for MICE [meetings, incentives, conference and events] delegates in the Maldives.
As well as the new homestay products, the destination also offers liveaboards and guest-houses but Thoyyib said the Maldives will retain its focus on premium luxury accommodation for which it is known.
The destination will launch a marketing campaign, called 'the story of the Maldives', "bringing the stories of our islands' culture and folklore" to consumers and the trade, he added.
Last year, the Maldives welcomed 500,000 despite the problems of the pandemic, and it has already surpassed the one-million mark in 2021.
Dr Abdulla Mausoom, Maldives' Minister of Tourism, said the nation has already recorded 10.7 million bed nights so far this year – which is better than in 2019, because people are staying longer.
He told delegates there will be a year-long celebration in 2022 marking the 50th anniversary of the start of tourism in the Maldives, with events and festivals taking place across the 200-island nation.
The minister told delegates that he was very glad to be at WTM, adding: "We have missed this for a very, very long time – being glued to our screens is not good."
Tagged wtml-pr.
Previous: Day 2 – UNWTO head tells ministers: 'The climate emergency is a bigger threat than Covid'
Next:Day 2 – Maldives aims to be leading meetings and events destination
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 6,742 |
[
128000,
10369,
220,
17,
1389,
468,
22809,
7295,
11,
279,
2626,
9731,
198,
644,
8612,
62115,
11,
468,
22809,
7295,
627,
791,
2132,
1938,
315,
468,
22809,
7295,
220,
2366,
16,
5602,
264,
13326,
18909,
315,
2626,
16659,
11,
4732,
29130,
11,
1579,
11852,
38614,
11,
23146,
323,
3577,
10015,
826,
627,
4054,
315,
279,
22020,
574,
279,
9974,
35050,
6,
30048,
11,
5762,
304,
15664,
449,
279,
6781,
54,
5319,
323,
59199,
7905,
627,
56050,
2191,
35050,
505,
2212,
279,
1917,
43347,
311,
990,
7119,
279,
13654,
315,
279,
10706,
449,
264,
5357,
389,
26012,
304,
22556,
5944,
627,
26054,
11,
279,
7327,
52652,
323,
33350,
15217,
520,
468,
22809,
1389,
42160,
13038,
11,
23261,
323,
57901,
1389,
8738,
1139,
1202,
2132,
1938,
520,
21705,
11,
4857
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
10369,
220,
17,
1389,
468,
22809,
7295,
11,
279,
2626,
9731,
198,
644,
8612,
62115,
11,
468,
22809,
7295,
627,
791,
2132,
1938,
315,
468,
22809,
7295,
220,
2366,
16,
5602,
264,
13326,
18909,
315,
2626,
16659,
11,
4732,
29130,
11,
1579,
11852,
38614,
11,
23146,
323,
3577,
10015,
826,
627,
4054,
315,
279,
22020,
574,
279,
9974,
35050,
6,
30048,
11,
5762,
304,
15664,
449,
279,
6781,
54,
5319,
323,
59199,
7905,
627,
56050,
2191,
35050,
505,
2212,
279,
1917,
43347,
311,
990,
7119,
279,
13654,
315,
279,
10706,
449,
264,
5357,
389,
26012,
304,
22556,
5944,
627,
26054,
11,
279,
7327,
52652,
323,
33350,
15217,
520,
468,
22809,
1389,
42160,
13038,
11,
23261,
323,
57901,
1389,
8738,
1139,
1202,
2132,
1938,
520,
21705,
11,
4857,
-100
] |
Before hiring any deck company you will ensure first that the corporate is legally registered and have a business license. Along with this you end up being well acquainted with the terms and conditions of the particular and the corporation should provides you with proof of business liability medical insurance. This will help you when any scratches to your home was to happen. Meeting the skillful staff of occasions company allows you to discuss the deck design in great detail and finalize an contract.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 7,306 |
[
128000,
10438,
24009,
904,
9722,
2883,
499,
690,
6106,
1176,
430,
279,
13166,
374,
26267,
9879,
323,
617,
264,
2626,
5842,
13,
32944,
449,
420,
499,
842,
709,
1694,
1664,
73734,
449,
279,
3878,
323,
4787,
315,
279,
4040,
323,
279,
27767,
1288,
5825,
499,
449,
11311,
315,
2626,
24305,
6593,
8276,
13,
1115,
690,
1520,
499,
994,
904,
66839,
311,
701,
2162,
574,
311,
3621,
13,
30155,
279,
10151,
1285,
5687,
315,
25975,
2883,
6276,
499,
311,
4358,
279,
9722,
2955,
304,
2294,
7872,
323,
55190,
459,
5226,
13,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
] |
[
10438,
24009,
904,
9722,
2883,
499,
690,
6106,
1176,
430,
279,
13166,
374,
26267,
9879,
323,
617,
264,
2626,
5842,
13,
32944,
449,
420,
499,
842,
709,
1694,
1664,
73734,
449,
279,
3878,
323,
4787,
315,
279,
4040,
323,
279,
27767,
1288,
5825,
499,
449,
11311,
315,
2626,
24305,
6593,
8276,
13,
1115,
690,
1520,
499,
994,
904,
66839,
311,
701,
2162,
574,
311,
3621,
13,
30155,
279,
10151,
1285,
5687,
315,
25975,
2883,
6276,
499,
311,
4358,
279,
9722,
2955,
304,
2294,
7872,
323,
55190,
459,
5226,
13,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100
] |
(2) mowing regime (2 and 4times/ year).
The design consists of 150 plots, each 1.5 m by 1.5 m in size, with every factorial combination replicated five times. Manipulations are applied in a modified Latin Square design; mowing regimes are blocked and randomly assigned within each manipulation.
1. minimal variability (CM): compare weekly precipitation sums with long-term average for the same period and irrigate if smaller (CM as carried out last year). Applied between 1.4. and 30.9.
All 5 manipulations will add up to the same total sum of precipitation, but differ in the variability. Intensity of drougth is based on the local 1000-year extreme event resulting in a drought period of 42 days. Drougth ist sumulated usind rain.-out shelters.
Performance of communities and single species is measured with a variety of parameters, covering all important stages in the life cicle of plants (cover, above and belowground biomass porduction, phenology, seed production, germinabilityor photosynthetic actovity).
Performance of communities and single species is measured with a variety of parameters, covering all important stages in the life cycle of plants (cover, above and belowground biomass production, phenology, seed production, germinability or photosynthetic activity). Furthermore, abiotic parameters like soil moisture, precipitation and temperature are recorded on site. Through cooperation partners, other important ecosystem parameters are analyzed, e.g. microbial activity, soil fauna, N-cycle, stable isotopes.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 6,516 |
[
128000,
7,
17,
8,
296,
24060,
17942,
320,
17,
323,
220,
19,
15487,
14,
1060,
4390,
791,
2955,
17610,
315,
220,
3965,
31794,
11,
1855,
220,
16,
13,
20,
296,
555,
220,
16,
13,
20,
296,
304,
1404,
11,
449,
1475,
54062,
10824,
72480,
4330,
3115,
13,
62011,
7607,
527,
9435,
304,
264,
11041,
20023,
15992,
2955,
26,
296,
24060,
61911,
527,
19857,
323,
27716,
12893,
2949,
1855,
34786,
627,
16,
13,
17832,
54709,
320,
10190,
1680,
9616,
17496,
61050,
37498,
449,
1317,
9860,
5578,
369,
279,
1890,
4261,
323,
53105,
349,
422,
9333,
320,
10190,
439,
11953,
704,
1566,
1060,
570,
43608,
1990,
220,
16,
13,
19,
13,
323,
220,
966,
13,
24,
627,
2460,
220,
20,
14951,
7607,
690,
923,
709,
311,
279,
1890,
2860,
2694
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
7,
17,
8,
296,
24060,
17942,
320,
17,
323,
220,
19,
15487,
14,
1060,
4390,
791,
2955,
17610,
315,
220,
3965,
31794,
11,
1855,
220,
16,
13,
20,
296,
555,
220,
16,
13,
20,
296,
304,
1404,
11,
449,
1475,
54062,
10824,
72480,
4330,
3115,
13,
62011,
7607,
527,
9435,
304,
264,
11041,
20023,
15992,
2955,
26,
296,
24060,
61911,
527,
19857,
323,
27716,
12893,
2949,
1855,
34786,
627,
16,
13,
17832,
54709,
320,
10190,
1680,
9616,
17496,
61050,
37498,
449,
1317,
9860,
5578,
369,
279,
1890,
4261,
323,
53105,
349,
422,
9333,
320,
10190,
439,
11953,
704,
1566,
1060,
570,
43608,
1990,
220,
16,
13,
19,
13,
323,
220,
966,
13,
24,
627,
2460,
220,
20,
14951,
7607,
690,
923,
709,
311,
279,
1890,
2860,
2694,
-100
] |
Fork Shoals, Fountain Inn, Local, Mauldin, Moonville, Piedmont, Politics, Simpsonville
Vice Presidents Mauldin visit recalled
On a March day in 1988, time stood still for Knollwood, one of the largest and oldest communities in Mauldin. Vice-President George H.W. Bush visited that neighborhood for a presidential fund-raiser. Bush would go on to claim the Republican Party's nomination for President of the United States and win a decisive victory over Michael Dukakis, the Democratic nominee for that office.
RC Jones, Mauldin mayor at the time, recalls the event being held at Dick Greer's home in the Knollwood community. Mr. Greer, who had close connections with Carroll Campbell, S.C. Governor at that time, and the Republican Party, hosted the hotdog supper where over a thousand people attended.
Vice-President Bush arrived in a bus but did not disembark until a host of Secret Service had thoroughly searched through Mr. Greer's home and surrounding areas. Taft Matney, current Mauldin Council member and lifelong resident of this community, was a high school student when Bush came to town. He remembers that the Secret Service commandeered his friend's game room and they had to find another place to hang out. Although he served baked beans at the event, he did not meet Bush personally but does recall satellite dishes and TV cameras everywhere as well as cars lined up as as far as you could see.
Vice-President Bush was accompanied by S.C. Governor Carroll Campbell and Cheryl Ladd, famed at the time for her role in "Charlie's Angels". Bush came outside and mingled in the crowd while Ms. Ladd said a few words. Afterwards, Bush himself gave a speech and returned to the area between the house and the crowd. It was here that Mr. Jones presented Bush a key to the city of Mauldin.
Former Mauldin Mayor RC Jones
A Crowded Affair
National and local news satellite dishes set up in adjacent field
Line of traffic heading to the event as Secret Service Agents check ID's
A young Taft Matney on the serving line
Vice-President Bush, with Governor Campbell and Mayor Jones
During the visit, Mr. Jones spent thirty minutes or so speaking with Bush. He says that the man was "down-to-earth", very humble, and interested in what others had to say. Promising to send a picture of the event, the then Vice-President thanked the mayor for the welcome and the key. Mr. Jones laughed as he recalled doubting that such a busy man would ever remember to do that. However, to his surprise, after President George H.W. Bush's inauguration as President of the United States, Mr. Green received a package from the White House. Inside, he found a letter and a photograph of himself with Governor Carroll Campbell and President George H.W Bush as he presented the key to the city of Mauldin. The photograph remains one of the mayor's most treasured possessions.
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 4,048 |
[
128000,
37,
672,
64040,
1147,
11,
78324,
17382,
11,
8949,
11,
94301,
73911,
11,
17781,
8078,
11,
84430,
21047,
11,
35979,
11,
35766,
8078,
198,
53,
560,
79739,
94301,
73911,
4034,
32230,
198,
1966,
264,
5587,
1938,
304,
220,
3753,
23,
11,
892,
14980,
2103,
369,
13934,
980,
6798,
11,
832,
315,
279,
7928,
323,
24417,
10977,
304,
94301,
73911,
13,
23270,
86481,
10058,
473,
1196,
13,
14409,
12263,
430,
12818,
369,
264,
13621,
3887,
59401,
12329,
13,
14409,
1053,
733,
389,
311,
3802,
279,
9540,
8722,
596,
29804,
369,
4900,
315,
279,
3723,
4273,
323,
3243,
264,
51391,
12845,
927,
8096,
423,
3178,
76877,
11,
279,
11650,
29311,
369,
430,
5274,
627,
7532,
12201,
11,
94301,
73911,
17352,
520,
279,
892,
11,
41231,
279,
1567,
1694,
5762
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
37,
672,
64040,
1147,
11,
78324,
17382,
11,
8949,
11,
94301,
73911,
11,
17781,
8078,
11,
84430,
21047,
11,
35979,
11,
35766,
8078,
198,
53,
560,
79739,
94301,
73911,
4034,
32230,
198,
1966,
264,
5587,
1938,
304,
220,
3753,
23,
11,
892,
14980,
2103,
369,
13934,
980,
6798,
11,
832,
315,
279,
7928,
323,
24417,
10977,
304,
94301,
73911,
13,
23270,
86481,
10058,
473,
1196,
13,
14409,
12263,
430,
12818,
369,
264,
13621,
3887,
59401,
12329,
13,
14409,
1053,
733,
389,
311,
3802,
279,
9540,
8722,
596,
29804,
369,
4900,
315,
279,
3723,
4273,
323,
3243,
264,
51391,
12845,
927,
8096,
423,
3178,
76877,
11,
279,
11650,
29311,
369,
430,
5274,
627,
7532,
12201,
11,
94301,
73911,
17352,
520,
279,
892,
11,
41231,
279,
1567,
1694,
5762,
-100
] |
<?xml version="1.0" encoding="utf-8"?>
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN"
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
<html xmlns="http://www.w3.org/1999/xhtml" lang="en" xml:lang="en">
<head>
<!-- 2016-10-10 Mon 15:48 -->
<meta http-equiv="Content-Type" content="text/html;charset=utf-8" />
<meta name="viewport" content="width=device-width, initial-scale=1" />
<title>ACM Computing Seminar C++ Guide</title>
<meta name="generator" content="Org-mode" />
<meta name="author" content="Matt Hancock" />
<style type="text/css">
<!--/*--><![CDATA[/*><!--*/
.title { text-align: center;
margin-bottom: .2em; }
.subtitle { text-align: center;
font-size: medium;
font-weight: bold;
margin-top:0; }
.todo { font-family: monospace; color: red; }
.done { font-family: monospace; color: green; }
.priority { font-family: monospace; color: orange; }
.tag { background-color: #eee; font-family: monospace;
padding: 2px; font-size: 80%; font-weight: normal; }
.timestamp { color: #bebebe; }
.timestamp-kwd { color: #5f9ea0; }
.org-right { margin-left: auto; margin-right: 0px; text-align: right; }
.org-left { margin-left: 0px; margin-right: auto; text-align: left; }
.org-center { margin-left: auto; margin-right: auto; text-align: center; }
.underline { text-decoration: underline; }
#postamble p, #preamble p { font-size: 90%; margin: .2em; }
p.verse { margin-left: 3%; }
pre {
border: 1px solid #ccc;
box-shadow: 3px 3px 3px #eee;
padding: 8pt;
font-family: monospace;
overflow: auto;
margin: 1.2em;
}
pre.src {
position: relative;
overflow: visible;
padding-top: 1.2em;
}
pre.src:before {
display: none;
position: absolute;
background-color: white;
top: -10px;
right: 10px;
padding: 3px;
border: 1px solid black;
}
pre.src:hover:before { display: inline;}
pre.src-sh:before { content: 'sh'; }
pre.src-bash:before { content: 'sh'; }
pre.src-emacs-lisp:before { content: 'Emacs Lisp'; }
pre.src-R:before { content: 'R'; }
pre.src-perl:before { content: 'Perl'; }
pre.src-java:before { content: 'Java'; }
pre.src-sql:before { content: 'SQL'; }
table { border-collapse:collapse; }
caption.t-above { caption-side: top; }
caption.t-bottom { caption-side: bottom; }
td, th { vertical-align:top; }
th.org-right { text-align: center; }
th.org-left { text-align: center; }
th.org-center { text-align: center; }
td.org-right { text-align: right; }
td.org-left { text-align: left; }
td.org-center { text-align: center; }
dt { font-weight: bold; }
.footpara { display: inline; }
.footdef { margin-bottom: 1em; }
.figure { padding: 1em; }
.figure p { text-align: center; }
.inlinetask {
padding: 10px;
border: 2px solid gray;
margin: 10px;
background: #ffffcc;
}
#org-div-home-and-up
{ text-align: right; font-size: 70%; white-space: nowrap; }
textarea { overflow-x: auto; }
.linenr { font-size: smaller }
.code-highlighted { background-color: #ffff00; }
.org-info-js_info-navigation { border-style: none; }
#org-info-js_console-label
{ font-size: 10px; font-weight: bold; white-space: nowrap; }
.org-info-js_search-highlight
{ background-color: #ffff00; color: #000000; font-weight: bold; }
/*]]>*/-->
</style>
<link rel="stylesheet" type="text/css" href="../css/main.css">
<script src="../js/main.js"></script>
<script type="text/javascript">
/*
@licstart The following is the entire license notice for the
JavaScript code in this tag.
Copyright (C) 2012-2013 Free Software Foundation, Inc.
The JavaScript code in this tag is free software: you can
redistribute it and/or modify it under the terms of the GNU
General Public License (GNU GPL) as published by the Free Software
Foundation, either version 3 of the License, or (at your option)
any later version. The code is distributed WITHOUT ANY WARRANTY;
without even the implied warranty of MERCHANTABILITY or FITNESS
FOR A PARTICULAR PURPOSE. See the GNU GPL for more details.
As additional permission under GNU GPL version 3 section 7, you
may distribute non-source (e.g., minimized or compacted) forms of
that code without the copy of the GNU GPL normally required by
section 4, provided you include this license notice and a URL
through which recipients can access the Corresponding Source.
@licend The above is the entire license notice
for the JavaScript code in this tag.
*/
<!--/*--><![CDATA[/*><!--*/
function CodeHighlightOn(elem, id)
{
var target = document.getElementById(id);
if(null != target) {
elem.cacheClassElem = elem.className;
elem.cacheClassTarget = target.className;
target.className = "code-highlighted";
elem.className = "code-highlighted";
}
}
function CodeHighlightOff(elem, id)
{
var target = document.getElementById(id);
if(elem.cacheClassElem)
elem.className = elem.cacheClassElem;
if(elem.cacheClassTarget)
target.className = elem.cacheClassTarget;
}
/*]]>*///-->
</script>
<script type="text/x-mathjax-config">
MathJax.Hub.Config({
displayAlign: "center",
displayIndent: "0em",
"HTML-CSS": { scale: 100,
linebreaks: { automatic: "false" },
webFont: "TeX"
},
SVG: {scale: 100,
linebreaks: { automatic: "false" },
font: "TeX"},
NativeMML: {scale: 100},
TeX: { equationNumbers: {autoNumber: "AMS"},
MultLineWidth: "85%",
TagSide: "right",
TagIndent: ".8em"
}
});
</script>
<script type="text/javascript"
src="http://cdn.mathjax.org/mathjax/latest/MathJax.js?config=TeX-AMS_HTML"></script>
</head>
<body>
<div id="content">
<h1 class="title">ACM Computing Seminar C++ Guide</h1>
<div id="table-of-contents">
<h2>Table of Contents</h2>
<div id="text-table-of-contents">
<ul>
<li><a href="#orgheadline4">1. Introduction</a>
<ul>
<li><a href="#orgheadline1">1.1. A little about the language</a></li>
<li><a href="#orgheadline2">1.2. Additional references</a></li>
<li><a href="#orgheadline3">1.3. License, source, etc.</a></li>
</ul>
</li>
<li><a href="#orgheadline8">2. Getting started</a>
<ul>
<li><a href="#orgheadline5">2.1. Text editors</a></li>
<li><a href="#orgheadline6">2.2. Compilers</a></li>
<li><a href="#orgheadline7">2.3. Writing a program</a></li>
</ul>
</li>
<li><a href="#orgheadline24">3. Data types</a>
<ul>
<li><a href="#orgheadline11">3.1. The <code>bool</code> type</a></li>
<li><a href="#orgheadline12">3.2. <code>int</code> types</a></li>
<li><a href="#orgheadline13">3.3. Floating point types</a></li>
<li><a href="#orgheadline14">3.4. Casting</a></li>
<li><a href="#orgheadline15">3.5. The <code>const</code> modifier</a></li>
<li><a href="#orgheadline16">3.6. The <code>typedef</code> keyword</a></li>
<li><a href="#orgheadline20">3.7. Pointers and references</a>
<ul>
<li><a href="#orgheadline17">3.7.1. Pointers</a></li>
<li><a href="#orgheadline19">3.7.2. References</a></li>
</ul>
</li>
<li><a href="#orgheadline23">3.8. Arrays</a>
<ul>
<li><a href="#orgheadline21">3.8.1. Fixed length arrays</a></li>
<li><a href="#orgheadline22">3.8.2. Dynamic length arrays</a></li>
</ul>
</li>
</ul>
</li>
<li><a href="#orgheadline34">4. Control structures</a>
<ul>
<li><a href="#orgheadline10">4.1. Conditionals</a>
<ul>
<li><a href="#orgheadline25">4.1.1. Example: <code>if</code> / <code>else</code> and random number generation</a></li>
<li><a href="#orgheadline26">4.1.2. Example: <code>if</code> / <code>else if</code> / <code>else</code></a></li>
</ul>
</li>
<li><a href="#orgheadline33">4.2. Loops</a>
<ul>
<li><a href="#orgheadline28">4.2.1. The <code>for</code> loop</a>
<ul>
<li><a href="#orgheadline27">4.2.1.1. Example: row-major matrix</a></li>
</ul>
</li>
<li><a href="#orgheadline31">4.2.2. The <code>while</code> loop</a>
<ul>
<li><a href="#orgheadline29">4.2.2.1. Example: truncated sum</a></li>
<li><a href="#orgheadline30">4.2.2.2. Example: estimating machine epsilon</a></li>
</ul>
</li>
<li><a href="#orgheadline32">4.2.3. The <code>break</code> keyword</a></li>
</ul>
</li>
</ul>
</li>
<li><a href="#orgheadline39">5. Input / Output</a>
<ul>
<li><a href="#orgheadline35">5.1. Inputs to <code>main</code>: <code>argc</code> and <code>argv</code></a></li>
<li><a href="#orgheadline38">5.2. Filestreams</a>
<ul>
<li><a href="#orgheadline36">5.2.1. Reading data from a file</a></li>
<li><a href="#orgheadline37">5.2.2. Writing data to a file</a></li>
</ul>
</li>
</ul>
</li>
<li><a href="#orgheadline18">6. Functions</a>
<ul>
<li><a href="#orgheadline41">6.1. Writing a function</a>
<ul>
<li><a href="#orgheadline40">6.1.1. Example: <code>linspace</code>: generating a set of equally-spaced points</a></li>
</ul>
</li>
<li><a href="#orgheadline45">6.2. Header and implementation files</a>
<ul>
<li><a href="#orgheadline42">6.2.1. The header file</a></li>
<li><a href="#orgheadline43">6.2.2. The implementation file</a></li>
<li><a href="#orgheadline44">6.2.3. The file containing <code>main</code></a></li>
</ul>
</li>
<li><a href="#orgheadline48">6.3. Function pointers</a>
<ul>
<li><a href="#orgheadline46">6.3.1. Example: Newton's method for rootfinding</a></li>
<li><a href="#orgheadline47">6.3.2. Example: The midpoint rule for definite integrals</a></li>
</ul>
</li>
</ul>
</li>
<li><a href="#orgheadline9">7. Object-oriented programming</a>
<ul>
<li><a href="#orgheadline59">7.1. Example: a vector <code>class</code></a>
<ul>
<li><a href="#orgheadline49">7.1.1. The header file</a></li>
<li><a href="#orgheadline50">7.1.2. The implementation file</a></li>
<li><a href="#orgheadline51">7.1.3. Example usage</a></li>
<li><a href="#orgheadline54">7.1.4. Operator overloading</a>
<ul>
<li><a href="#orgheadline52">7.1.4.1. Overloading <code>operator[]</code></a></li>
<li><a href="#orgheadline53">7.1.4.2. Overloading <code>operator=</code></a></li>
</ul>
</li>
<li><a href="#orgheadline55">7.1.5. The copy constructor</a></li>
<li><a href="#orgheadline58">7.1.6. Friend functions</a>
<ul>
<li><a href="#orgheadline56">7.1.6.1. Overloading <code>operator*</code></a></li>
<li><a href="#orgheadline57">7.1.6.2. Overloading <code>operator<<</code></a></li>
</ul>
</li>
</ul>
</li>
<li><a href="#orgheadline61">7.2. Templating: a matrix class</a>
<ul>
<li><a href="#orgheadline60">7.2.1. Overloading <code>operator*</code></a></li>
</ul>
</li>
<li><a href="#orgheadline62">7.3. Inheritance</a></li>
</ul>
</li>
</ul>
</div>
</div>
<div id="main">
<div id="footer"><a href="../../../">ACM Computing Seminar Home</a> — This work is licensed under a <a rel="license" href="http://creativecommons.org/licenses/by/3.0/deed.en_US">Creative Commons Attribution 3.0 Unported License</a>.</div>
<a href="javascript:;" id="toc-open">☰</a>
<a href="javascript:;" id="toc-close">×</a>
<div id="outline-container-orgheadline4" class="outline-2">
<h2 id="orgheadline4"><span class="section-number-2">1</span> Introduction</h2>
<div class="outline-text-2" id="text-1">
<p>
This manual is a guide for quickly learning C++ for mathematical and
scientific computing applications. The goal of this guide is not to
make you a C++ expert, but to quickly teach you enough of the C++
fundamentals and design patterns to help you off the ground. If you
should like to go beyond this guide, a few references are listed below.
</p>
</div>
<div id="outline-container-orgheadline1" class="outline-3">
<h3 id="orgheadline1"><span class="section-number-3">1.1</span> A little about the language</h3>
<div class="outline-text-3" id="text-1-1">
<p>
Before you dive in, here is a little about the C++ programming language:
</p>
<p>
C++ is an extension of the C programming language. Both C and C++ are
<b>statically-typed</b> and <b>compiled</b> languages, which means that the <b>type</b>
of variables used in your source code must be declared explicitly and is
checked when the program is compiled (i.e., translated into a machine
executable file).
</p>
<p>
One key difference between C++ and C, however, is that C++ provides
many mechanisms to allow for the <a href="https://en.wikipedia.org/wiki/Object-oriented_programming">object-oriented programming paradigm</a>. This
essentially allows the software writer to create custom, complex,
reusable data structures. The object-oriented paradigm is extremely useful,
but we will only touch the surface of it in this guide.
</p>
</div>
</div>
<div id="outline-container-orgheadline2" class="outline-3">
<h3 id="orgheadline2"><span class="section-number-3">1.2</span> Additional references</h3>
<div class="outline-text-3" id="text-1-2">
<ul class="org-ul">
<li><a href="http://cppreference.com/">C++ reference</a></li>
<li><a href="http://www.cplusplus.com/doc/tutorial">C++ tutorials</a></li>
<li><a href="https://en.wikipedia.org/wiki/C++">C++ wiki</a></li>
<li><a href="http://stackoverflow.com/questions/388242/the-definitive-c-book-guide-and-list">A compiled list of C++ textbooks from stackoverflow</a></li>
</ul>
</div>
</div>
<div id="outline-container-orgheadline3" class="outline-3">
<h3 id="orgheadline3"><span class="section-number-3">1.3</span> License, source, etc.</h3>
<div class="outline-text-3" id="text-1-3">
<p>
This document was created using <a href="https://www.gnu.org/software/emacs/">Emacs</a> <a href="http://orgmode.org/">org mode</a> with some custom css and
javascript. You can find the license, view the source, and contribute
to this document here:
</p>
<p>
<a href="https://github.com/notmatthancock/acm-computing-seminar"><a href="https://github.com/notmatthancock/acm-computing-seminar">https://github.com/notmatthancock/acm-computing-seminar</a></a>
</p>
</div>
</div>
</div>
<div id="outline-container-orgheadline8" class="outline-2">
<h2 id="orgheadline8"><span class="section-number-2">2</span> Getting started</h2>
<div class="outline-text-2" id="text-2">
<p>
The particular programming tools that you choose to use will
likely be largely influenced by the operating system that you use.
We will use free tools (often developed for GNU / Linux systems) in
this guide. These tools are mostly available in other operating
systems as well. For example, on Windows, you could use <a href="https://www.cygwin.com/">Cygwin</a>, or
install a dual boot with some Linux distribution (e.g., <a href="http://www.ubuntu.com/">Ubuntu</a>). On the
other hand, MAC OSX, being a BSD-derived system, has many of the
required tools already available (although, a command line utility,
<a href="http://brew.sh/">Brew</a>, makes building and installing other tools very simple).
</p>
<p>
In the following two sections, we'll talk about the two basic types of
software that you'll need to begin writing C++ programs.
</p>
</div>
<div id="outline-container-orgheadline5" class="outline-3">
<h3 id="orgheadline5"><span class="section-number-3">2.1</span> Text editors</h3>
<div class="outline-text-3" id="text-2-1">
<p>
The text editor that you choose to use should be be any program
capable of editing plain text files. However, you may find that it's
more productive to write in an editor that offers features such as
syntax highlighting, code-completion, bracket-matching, or other features.
Here are some popular free text editors:
</p>
<ul class="org-ul">
<li><a href="https://atom.io/">Atom</a> is a recently open-sourced GUI editor which some have compared to the very popular non-free editor, <a href="https://sublimetext.com/">Sublime Text</a>.</li>
<li><a href="https://www.gnu.org/software/emacs/">Emacs</a> is another powerful editor, which allows for highly optimized workflows.</li>
<li><a href="https://wiki.gnome.org/Apps/Gedit">Gedit</a> is a nice and simple GUI editor, which is the default in GNOME desktop environment.</li>
<li><a href="https://kate-editor.org/">Kate</a> is a another simple GUI editor, which is the default in the KDE desktop environment.</li>
<li><a href="http://www.vim.org/">Vim</a> is a modal editor with a steep learning curve. It offers highly efficient means to edit text, and is available (or it's predecessor, vi) by default on nearly all UNIX-like operating systems.</li>
</ul>
</div>
</div>
<div id="outline-container-orgheadline6" class="outline-3">
<h3 id="orgheadline6"><span class="section-number-3">2.2</span> Compilers</h3>
<div class="outline-text-3" id="text-2-2">
<p>
Second, you'll need a program called a <b>compiler</b>. A compiler translates
the high-level C++ language into an executable program. In this guide, we
will use the <code>g++</code> compiler which is freely available through the
<a href="https://gcc.gnu.org/">gnu compiler collection (gcc)</a>.
</p>
<p>
<code>g++</code> is a program which you typically call from the command line, which
takes as input, your C++ source code file, and produces as output, a
binary executable file.
</p>
</div>
</div>
<div id="outline-container-orgheadline7" class="outline-3">
<h3 id="orgheadline7"><span class="section-number-3">2.3</span> Writing a program</h3>
<div class="outline-text-3" id="text-2-3">
<p>
Let's create our first C++ program, the obligatory "Hello, world!". First,
fire-up your text editor and create a file called, <code>hello.cpp</code>, with the
following contents:
</p>
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr">1: </span><span style="color: #0072b2;">#include</span> <span style="color: #848ea9;"><iostream></span>
<span class="linenr">2: </span>
<span class="linenr">3: </span><span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #d55e00;">main</span>() {
<span class="linenr">4: </span> <span style="color: #d55e00; font-weight: bold;">std</span>::cout << <span style="color: #848ea9;">"Hello, world!"</span>;
<span class="linenr">5: </span> <span style="color: #56b4e9; font-weight: bold;">return</span> 0;
<span class="linenr">6: </span>}
</pre>
</div>
<p>
Now, to compile the program, execute:
</p>
<div class="org-src-container">
<pre class="src src-sh">g++ hello.cpp
</pre>
</div>
<p>
Followed by:
</p>
<div class="org-src-container">
<pre class="src src-sh">./a.out
</pre>
</div>
<p>
By default, <code>g++</code> names the resulting binary executable, <code>a.out</code>, but
you can specify a different output file name by using the <code>-o</code> flag:
</p>
<div class="org-src-container">
<pre class="src src-sh">g++ hello.cpp -o my_executable_file.bin
</pre>
</div>
<p>
Note that in Windows, executable files typically end in <code>.exe</code>. In Unix-like
systems, there is no particular file-extension type for executables.
</p>
</div>
</div>
</div>
<div id="outline-container-orgheadline24" class="outline-2">
<h2 id="orgheadline24"><span class="section-number-2">3</span> Data types</h2>
<div class="outline-text-2" id="text-3">
<p>
As we <a href="#orgheadline1">mentioned previously</a>, you must explicitly declare the type of a
variable. So, in this section, we'll talk about the main variable
types you'll use. In the section on <a href="#orgheadline9">object-oriented programming</a>,
we'll discuss how to build more complex data types.
</p>
</div>
<div id="outline-container-orgheadline11" class="outline-3">
<h3 id="orgheadline11"><span class="section-number-3">3.1</span> The <code>bool</code> type</h3>
<div class="outline-text-3" id="text-3-1">
<p>
A boolean data type is either <code>true</code> or <code>false</code>. There are a number of
operators between these types, illustrated in the code snippet below
(note that lines starting with <code>//</code> are comments and are ignored by the
compiler):
</p>
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr"> 1: </span><span style="color: #0072b2; font-weight: bold;">bool</span> <span style="color: #e69f00; font-weight: bold;">a</span>,<span style="color: #e69f00; font-weight: bold;">b</span>,<span style="color: #e69f00; font-weight: bold;">c</span>; <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Declare the type of variables a, b, and c.</span>
<span class="linenr"> 2: </span>a = <span style="color: #d55e00; font-weight: bold;">true</span>;
<span class="linenr"> 3: </span>b = <span style="color: #d55e00; font-weight: bold;">false</span>;
<span class="linenr"> 4: </span>
<span class="linenr"> 5: </span><span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">! is logical negation when applied to a single variable.</span>
<span class="linenr"> 6: </span>c = !a; <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">c is false.</span>
<span class="linenr"> 7: </span>
<span class="linenr"> 8: </span><span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">&& is logical and.</span>
<span class="linenr"> 9: </span>c = (a && b); <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">c is false.</span>
<span class="linenr">10: </span>
<span class="linenr">11: </span><span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">|| is logical or.</span>
<span class="linenr">12: </span>c = (a || b); <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">c is true.</span>
</pre>
</div>
<p>
We don't often use boolean variables by themselves, but rather as a result
of comparing two other data types (such as comparing if one integer is less
than another integer). Expressions that result in boolean values are mostly
used in <a href="#orgheadline10">conditional control structures</a>.
</p>
</div>
</div>
<div id="outline-container-orgheadline12" class="outline-3">
<h3 id="orgheadline12"><span class="section-number-3">3.2</span> <code>int</code> types</h3>
<div class="outline-text-3" id="text-3-2">
<p>
There are <a href="http://en.cppreference.com/w/cpp/language/types">a variety of integer types</a> in C++. Below, we illustrate a
couple. These can be modified further using the <code>short</code> and <code>long</code>
keywords, changing the number of bytes occupied by the variable (and
hence the maximum and minimum size the variable can take on).
</p>
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr"> 1: </span><span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">a</span> = 6; <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">initialize a to 6.</span>
<span class="linenr"> 2: </span><span style="color: #0072b2; font-weight: bold;">unsigned</span> <span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">b</span> = 7; <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">initialize b to 7.</span>
<span class="linenr"> 3: </span><span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">c</span>; <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">declare c to be an integer variable.</span>
<span class="linenr"> 4: </span>
<span class="linenr"> 5: </span>a = 6;
<span class="linenr"> 6: </span>b = 7;
<span class="linenr"> 7: </span>
<span class="linenr"> 8: </span>c = a / b; <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">c is 0</span>
<span class="linenr"> 9: </span>c = b / a; <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">c is 1</span>
<span class="linenr">10: </span>c = b % a; <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">c is 1 (% is the integer remainder or modulo operator)</span>
<span class="linenr">11: </span>c = a - b; <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">c is -1</span>
<span class="linenr">12: </span>c = a > b; <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">c is 0 (boolean gets cast to integer)</span>
<span class="linenr">13: </span>c = a < b; <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">c is 1 (boolean gets cast to integer)</span>
<span class="linenr">14: </span>c++; <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">c is 2 (++ is shorthand for c = c + 1)</span>
<span class="linenr">15: </span>
<span class="linenr">16: </span>b = a - b; <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">b is 4294967295 (-1 gets cast to unsigned)</span>
<span class="linenr">17: </span>b = b + 1; <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">b is 0 (b was previously the largest unsigned,</span>
<span class="linenr">18: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">so adding one circles it back to zero.)</span>
<span class="linenr">19: </span>b += 7; <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">b is 7 (+= is shorthand for b = b + 7;</span>
</pre>
</div>
<p>
In the above, we've illustrated the use of signed and unsigned
integer types and the operators between them. It is important to
take care when you assign a result to a variable that doesn't match
the type of the result. In many cases, the result gets implicitly
cast to the type of variable being assigned to. The result may or
may not match your expectations, as shown above.
</p>
</div>
</div>
<div id="outline-container-orgheadline13" class="outline-3">
<h3 id="orgheadline13"><span class="section-number-3">3.3</span> Floating point types</h3>
<div class="outline-text-3" id="text-3-3">
<p>
There are two main floating point data types in C++, <code>float</code> and <code>double</code>,
which correspond to <a href="https://en.wikipedia.org/wiki/IEEE_floating_point">IEEE 32- and 64-bit floating point types</a>.
</p>
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr"> 1: </span><span style="color: #0072b2;">#include</span> <span style="color: #848ea9;"><iostream></span>
<span class="linenr"> 2: </span><span style="color: #0072b2;">#include</span> <span style="color: #848ea9;"><limits></span>
<span class="linenr"> 3: </span>
<span class="linenr"> 4: </span><span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #d55e00;">main</span>() {
<span class="linenr"> 5: </span> <span style="color: #0072b2; font-weight: bold;">float</span> <span style="color: #e69f00; font-weight: bold;">a</span>; <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Declare a single precision float.</span>
<span class="linenr"> 6: </span> <span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #e69f00; font-weight: bold;">b</span>; <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Declare a double precision float.</span>
<span class="linenr"> 7: </span>
<span class="linenr"> 8: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Print the max value of a float type.</span>
<span class="linenr"> 9: </span> <span style="color: #d55e00; font-weight: bold;">std</span>::cout << <span style="color: #d55e00; font-weight: bold;">std</span>::<span style="color: #d55e00; font-weight: bold;">numeric_limits</span><<span style="color: #0072b2; font-weight: bold;">float</span>>::max() << <span style="color: #d55e00; font-weight: bold;">std</span>::endl;
<span class="linenr">10: </span>
<span class="linenr">11: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Print the max value of a double type.</span>
<span class="linenr">12: </span> <span style="color: #d55e00; font-weight: bold;">std</span>::cout << <span style="color: #d55e00; font-weight: bold;">std</span>::<span style="color: #d55e00; font-weight: bold;">numeric_limits</span><<span style="color: #0072b2; font-weight: bold;">double</span>>::max() << <span style="color: #d55e00; font-weight: bold;">std</span>::endl;
<span class="linenr">13: </span>
<span class="linenr">14: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Print machine epsilon of a float type.</span>
<span class="linenr">15: </span> <span style="color: #d55e00; font-weight: bold;">std</span>::cout << <span style="color: #d55e00; font-weight: bold;">std</span>::<span style="color: #d55e00; font-weight: bold;">numeric_limits</span><<span style="color: #0072b2; font-weight: bold;">float</span>>::epsilon() << <span style="color: #d55e00; font-weight: bold;">std</span>::endl;
<span class="linenr">16: </span>
<span class="linenr">17: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Print machine epsilon of a double type.</span>
<span class="linenr">18: </span> <span style="color: #d55e00; font-weight: bold;">std</span>::cout << <span style="color: #d55e00; font-weight: bold;">std</span>::<span style="color: #d55e00; font-weight: bold;">numeric_limits</span><<span style="color: #0072b2; font-weight: bold;">double</span>>::epsilon() << <span style="color: #d55e00; font-weight: bold;">std</span>::endl;
<span class="linenr">19: </span>
<span class="linenr">20: </span> <span style="color: #56b4e9; font-weight: bold;">return</span> 0;
<span class="linenr">21: </span>}
</pre>
</div>
<pre class="example">
3.40282e+38
1.79769e+308
1.19209e-07
2.22045e-16
</pre>
</div>
</div>
<div id="outline-container-orgheadline14" class="outline-3">
<h3 id="orgheadline14"><span class="section-number-3">3.4</span> Casting</h3>
<div class="outline-text-3" id="text-3-4">
<p>
Sometimes it is useful to explicitly cast one variable type as another.
This can be done like the following:
</p>
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr">1: </span><span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">a</span>; <span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #e69f00; font-weight: bold;">b</span> = 3.14159;
<span class="linenr">2: </span>
<span class="linenr">3: </span>a = (<span style="color: #0072b2; font-weight: bold;">int</span>) b;
<span class="linenr">4: </span>
<span class="linenr">5: </span><span style="color: #d55e00; font-weight: bold;">std</span>::cout << a << <span style="color: #d55e00; font-weight: bold;">std</span>::endl;
</pre>
</div>
<pre class="example">
3
</pre>
</div>
</div>
<div id="outline-container-orgheadline15" class="outline-3">
<h3 id="orgheadline15"><span class="section-number-3">3.5</span> The <code>const</code> modifier</h3>
<div class="outline-text-3" id="text-3-5">
<p>
If the value of some variable should not change, you can use the <code>const</code>
keyword to protect its status. It is typical to denote <code>const</code> variables
with all caps. Try to compile the following program:
</p>
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr">1: </span><span style="color: #56b4e9; font-weight: bold;">const</span> <span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #e69f00; font-weight: bold;">PI</span> = 3.14159;
<span class="linenr">2: </span>
<span class="linenr">3: </span>PI = 3.0;
</pre>
</div>
<p>
You will see an error like, <code>error: assignment of read-only variable 'PI'</code>.
</p>
</div>
</div>
<div id="outline-container-orgheadline16" class="outline-3">
<h3 id="orgheadline16"><span class="section-number-3">3.6</span> The <code>typedef</code> keyword</h3>
<div class="outline-text-3" id="text-3-6">
<p>
Suppose you have a large numerical experiment, where all your code used
floating point of type <code>double</code>. Your curious about how the results will
be affected by changing the floating point type to single precision <code>float</code>
type. One solution would be to run a "find and replace" in your editor, but
something about that doesn't feel right.
</p>
<p>
Instead, we can use the <code>typedef</code> statement to define types:
</p>
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr"> 1: </span><span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Define "int_type" to be a short int.</span>
<span class="linenr"> 2: </span><span style="color: #56b4e9; font-weight: bold;">typedef</span> <span style="color: #0072b2; font-weight: bold;">short</span> <span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #0072b2; font-weight: bold;">int_type</span>;
<span class="linenr"> 3: </span>
<span class="linenr"> 4: </span><span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Define "float_type" to be single precision float.</span>
<span class="linenr"> 5: </span><span style="color: #56b4e9; font-weight: bold;">typedef</span> <span style="color: #0072b2; font-weight: bold;">float</span> <span style="color: #0072b2; font-weight: bold;">float_type</span>;
<span class="linenr"> 6: </span>
<span class="linenr"> 7: </span><span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Define "array_index_type" to be unsigned long int.</span>
<span class="linenr"> 8: </span><span style="color: #56b4e9; font-weight: bold;">typedef</span> <span style="color: #0072b2; font-weight: bold;">unsigned</span> <span style="color: #0072b2; font-weight: bold;">long</span> <span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #0072b2; font-weight: bold;">array_index_type</span>;
<span class="linenr"> 9: </span>
<span class="linenr">10: </span><span style="color: #0072b2; font-weight: bold;">int_type</span> <span style="color: #e69f00; font-weight: bold;">a</span> = -17;
<span class="linenr">11: </span><span style="color: #0072b2; font-weight: bold;">float_type</span> <span style="color: #e69f00; font-weight: bold;">b</span> = 1.14;
<span class="linenr">12: </span><span style="color: #0072b2; font-weight: bold;">array_index_type</span> <span style="color: #e69f00; font-weight: bold;">c</span> = 9;
</pre>
</div>
</div>
</div>
<div id="outline-container-orgheadline20" class="outline-3">
<h3 id="orgheadline20"><span class="section-number-3">3.7</span> Pointers and references</h3>
<div class="outline-text-3" id="text-3-7">
</div><div id="outline-container-orgheadline17" class="outline-4">
<h4 id="orgheadline17"><span class="section-number-4">3.7.1</span> Pointers</h4>
<div class="outline-text-4" id="text-3-7-1">
<p>
Pointers are variables that hold the <b>memory address</b> for a variable
of a specific type. Pointers are declared by specifying the variable
type, followed by the <code>*</code> symbol, followed by the name of the pointer
variable, e.g., <code>double * x</code> defines a "pointer to double" variable.
The variable, <code>x</code>, therefore, does not hold the value of a <code>double</code>
type, but rather, the memory address for a variable of type, <code>double</code>.
The memory address for a variable can be obtained by the <code>&</code> operator.
</p>
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr">1: </span><span style="color: #0072b2; font-weight: bold;">double</span> * <span style="color: #e69f00; font-weight: bold;">a</span>;
<span class="linenr">2: </span><span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #e69f00; font-weight: bold;">b</span> = 7;
<span class="linenr">3: </span>
<span class="linenr">4: </span><span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">This obtains the memory address of `b`.</span>
<span class="linenr">5: </span>a = &b;
<span class="linenr">6: </span>
<span class="linenr">7: </span><span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Prints some memory address (starts with 0x)</span>
<span class="linenr">8: </span><span style="color: #d55e00; font-weight: bold;">std</span>::cout << a << <span style="color: #d55e00; font-weight: bold;">std</span>::endl;
</pre>
</div>
<pre class="example">
0x7ffe0d98f7b8
</pre>
<p>
Similar to obtaining the memory address from a regular variable, using the
<code>&</code> operator, you can use the <code>*</code> symbol before a pointer to access the
variable value held at the memory location of the pointer. In this context,
the <code>*</code> symbol is called the <b>dereference operator</b>. This is probably better
understood with a short example:
</p>
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr"> 1: </span><span style="color: #0072b2; font-weight: bold;">double</span> * <span style="color: #e69f00; font-weight: bold;">a</span>;
<span class="linenr"> 2: </span><span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #e69f00; font-weight: bold;">b</span> = 7.3;
<span class="linenr"> 3: </span><span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #e69f00; font-weight: bold;">c</span>;
<span class="linenr"> 4: </span>
<span class="linenr"> 5: </span><span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Now `a` holds the memory address of `b`.</span>
<span class="linenr"> 6: </span>a = &b;
<span class="linenr"> 7: </span>
<span class="linenr"> 8: </span><span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">`*a` obtains the value of the variable</span>
<span class="linenr"> 9: </span><span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">at the memory address held by `a`.</span>
<span class="linenr">10: </span><span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">So, `c` is 7.3.</span>
<span class="linenr">11: </span>c = *a;
<span class="linenr">12: </span>
<span class="linenr">13: </span><span style="color: #d55e00; font-weight: bold;">std</span>::cout << c << <span style="color: #848ea9;">"\n"</span>;
</pre>
</div>
<pre class="example">
7.3
</pre>
</div>
</div>
<div id="outline-container-orgheadline19" class="outline-4">
<h4 id="orgheadline19"><span class="section-number-4">3.7.2</span> References</h4>
<div class="outline-text-4" id="text-3-7-2">
<p>
A reference is a sort of like a pointer, but not quite. <a href="https://en.wikipedia.org/wiki/Reference_(C++)">There are differences</a>.
A good analogy, which you can find in the previous link, is that a reference
is similar to a symbolic link, or "shortcut" if you're on Windows. You can
treat it more-or-less like the original variable, but it's not the original.
</p>
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr"> 1: </span><span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #e69f00; font-weight: bold;">a</span> = 1.1;
<span class="linenr"> 2: </span><span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">`b` is a reference to `a`.</span>
<span class="linenr"> 3: </span><span style="color: #0072b2; font-weight: bold;">double</span> & <span style="color: #e69f00; font-weight: bold;">b</span> = a;
<span class="linenr"> 4: </span>
<span class="linenr"> 5: </span><span style="color: #d55e00; font-weight: bold;">std</span>::cout << <span style="color: #848ea9;">"a: "</span> << a << <span style="color: #848ea9;">", b: "</span> << b << <span style="color: #848ea9;">"\n"</span>;
<span class="linenr"> 6: </span>
<span class="linenr"> 7: </span>a = 2.1;
<span class="linenr"> 8: </span>
<span class="linenr"> 9: </span><span style="color: #d55e00; font-weight: bold;">std</span>::cout << <span style="color: #848ea9;">"a: "</span> << a << <span style="color: #848ea9;">", b: "</span> << b << <span style="color: #848ea9;">"\n"</span>;
<span class="linenr">10: </span>
<span class="linenr">11: </span>b = 3.1;
<span class="linenr">12: </span>
<span class="linenr">13: </span><span style="color: #d55e00; font-weight: bold;">std</span>::cout << <span style="color: #848ea9;">"a: "</span> << a << <span style="color: #848ea9;">", b: "</span> << b << <span style="color: #848ea9;">"\n"</span>;
<span class="linenr">14: </span>
<span class="linenr">15: </span><span style="color: #d55e00; font-weight: bold;">std</span>::cout << <span style="color: #848ea9;">"\n\n"</span>;
<span class="linenr">16: </span><span style="color: #d55e00; font-weight: bold;">std</span>::cout << <span style="color: #848ea9;">"&a: "</span> << &a << <span style="color: #848ea9;">"\n"</span> << <span style="color: #848ea9;">"&b: "</span> << &b << <span style="color: #848ea9;">"\n"</span>;
</pre>
</div>
<pre class="example">
a: 1.1, b: 1.1
a: 2.1, b: 2.1
a: 3.1, b: 3.1
&a: 0x7ffcfbe7b1e8
&b: 0x7ffcfbe7b1e8
</pre>
<p>
References are useful for passing around large objects, so that the object
doesn't need to be copied. References are also useful as a return type for
functions <a href="#orgheadline18">(to be discussed later)</a> because it allows to assign to assign a
value to a function, which is useful if the function, for example, returns
a reference to the element of an array.
</p>
</div>
</div>
</div>
<div id="outline-container-orgheadline23" class="outline-3">
<h3 id="orgheadline23"><span class="section-number-3">3.8</span> Arrays</h3>
<div class="outline-text-3" id="text-3-8">
<p>
The length of an array can be fixed or dynamic, and how you
declare the array depends on this. Array indexing starts at 0 in C++
(compared to start at 1, for example, in Fortran or Matlab).
</p>
</div>
<div id="outline-container-orgheadline21" class="outline-4">
<h4 id="orgheadline21"><span class="section-number-4">3.8.1</span> Fixed length arrays</h4>
<div class="outline-text-4" id="text-3-8-1">
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr">1: </span><span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #e69f00; font-weight: bold;">a</span>[5];
<span class="linenr">2: </span>
<span class="linenr">3: </span>a[0] = 1.0;
<span class="linenr">4: </span><span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">etc.</span>
</pre>
</div>
</div>
</div>
<div id="outline-container-orgheadline22" class="outline-4">
<h4 id="orgheadline22"><span class="section-number-4">3.8.2</span> Dynamic length arrays</h4>
<div class="outline-text-4" id="text-3-8-2">
<p>
Dynamic length arrays are made possible through pointers:
</p>
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr"> 1: </span><span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">This allocates memory for 5 double types.</span>
<span class="linenr"> 2: </span><span style="color: #0072b2; font-weight: bold;">double</span> * <span style="color: #e69f00; font-weight: bold;">a</span> = <span style="color: #56b4e9; font-weight: bold;">new</span> <span style="color: #0072b2; font-weight: bold;">double</span>[5];
<span class="linenr"> 3: </span>
<span class="linenr"> 4: </span><span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Afterwards, you can treat `a` like a normal array.</span>
<span class="linenr"> 5: </span>a[0] = 1.0;
<span class="linenr"> 6: </span><span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">etc...</span>
<span class="linenr"> 7: </span>
<span class="linenr"> 8: </span><span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Whenever you use the `new` keyword, you must</span>
<span class="linenr"> 9: </span><span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">delete the memory allocated when you're done by hand.</span>
<span class="linenr">10: </span><span style="color: #56b4e9; font-weight: bold;">delete</span> [] a;
<span class="linenr">11: </span>
<span class="linenr">12: </span><span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">We can change the size of `a`.</span>
<span class="linenr">13: </span>a = <span style="color: #56b4e9; font-weight: bold;">new</span> <span style="color: #0072b2; font-weight: bold;">double</span> [10];
<span class="linenr">14: </span>
<span class="linenr">15: </span>a[0] = 2.0;
<span class="linenr">16: </span><span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">etc...</span>
<span class="linenr">17: </span>
<span class="linenr">18: </span><span style="color: #56b4e9; font-weight: bold;">delete</span> [] a;
</pre>
</div>
<p>
Note that omitting the first <code>delete</code> statement will cause no error.
However, the memory allocated by the first <code>new</code> statement will not
be freed, and thus inaccessible. This is bad because the memory cannot
be allocated to other resources. You should generally try to avoid
manually memory management when possible, but a good tool for debugging
memory problems is called <a href="http://valgrind.org/">valgrind</a>.
</p>
</div>
</div>
</div>
</div>
<div id="outline-container-orgheadline34" class="outline-2">
<h2 id="orgheadline34"><span class="section-number-2">4</span> Control structures</h2>
<div class="outline-text-2" id="text-4">
</div><div id="outline-container-orgheadline10" class="outline-3">
<h3 id="orgheadline10"><span class="section-number-3">4.1</span> Conditionals</h3>
<div class="outline-text-3" id="text-4-1">
</div><div id="outline-container-orgheadline25" class="outline-4">
<h4 id="orgheadline25"><span class="section-number-4">4.1.1</span> Example: <code>if</code> / <code>else</code> and random number generation</h4>
<div class="outline-text-4" id="text-4-1-1">
<p>
Often a code block should only be executed if some condition is true.
Below, we generate a random number between 0 and 1; print the number; and,
print whether or not the number was greater than 0.5.
</p>
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr"> 1: </span><span style="color: #0072b2;">#include</span> <span style="color: #848ea9;"><iostream></span>
<span class="linenr"> 2: </span><span style="color: #0072b2;">#include</span> <span style="color: #848ea9;"><stdlib.h></span>
<span class="linenr"> 3: </span><span style="color: #0072b2;">#include</span> <span style="color: #848ea9;"><time.h></span>
<span class="linenr"> 4: </span>
<span class="linenr"> 5: </span><span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #d55e00;">main</span>() {
<span class="linenr"> 6: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Seed a random number generator.</span>
<span class="linenr"> 7: </span> srand(123);
<span class="linenr"> 8: </span>
<span class="linenr"> 9: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">rand() produces a random integer between 0 and RAND_MAX.</span>
<span class="linenr">10: </span> <span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #e69f00; font-weight: bold;">num</span> = rand() / ((<span style="color: #0072b2; font-weight: bold;">double</span>) RAND_MAX);
<span class="linenr">11: </span>
<span class="linenr">12: </span> <span style="color: #d55e00; font-weight: bold;">std</span>::cout << <span style="color: #848ea9;">"num: "</span> << num << <span style="color: #848ea9;">"\n"</span>;
<span class="linenr">13: </span>
<span class="linenr">14: </span> <span style="color: #56b4e9; font-weight: bold;">if</span> (num < 0.5) {
<span class="linenr">15: </span> <span style="color: #d55e00; font-weight: bold;">std</span>::cout << <span style="color: #848ea9;">"num was less than 0.5.\n"</span>;
<span class="linenr">16: </span> }
<span class="linenr">17: </span> <span style="color: #56b4e9; font-weight: bold;">else</span> {
<span class="linenr">18: </span> <span style="color: #d55e00; font-weight: bold;">std</span>::cout << <span style="color: #848ea9;">"num was greater than 0.5.\n"</span>;
<span class="linenr">19: </span> }
<span class="linenr">20: </span>
<span class="linenr">21: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Do it again.</span>
<span class="linenr">22: </span> num = rand() / ((<span style="color: #0072b2; font-weight: bold;">double</span>) RAND_MAX);
<span class="linenr">23: </span>
<span class="linenr">24: </span> <span style="color: #d55e00; font-weight: bold;">std</span>::cout << <span style="color: #848ea9;">"num: "</span> << num << <span style="color: #848ea9;">"\n"</span>;
<span class="linenr">25: </span>
<span class="linenr">26: </span> <span style="color: #56b4e9; font-weight: bold;">if</span> (num < 0.5) {
<span class="linenr">27: </span> <span style="color: #d55e00; font-weight: bold;">std</span>::cout << <span style="color: #848ea9;">"num was less than 0.5.\n"</span>;
<span class="linenr">28: </span> }
<span class="linenr">29: </span> <span style="color: #56b4e9; font-weight: bold;">else</span> {
<span class="linenr">30: </span> <span style="color: #d55e00; font-weight: bold;">std</span>::cout << <span style="color: #848ea9;">"num was greater than 0.5.\n"</span>;
<span class="linenr">31: </span> }
<span class="linenr">32: </span>
<span class="linenr">33: </span> <span style="color: #56b4e9; font-weight: bold;">return</span> 0;
<span class="linenr">34: </span>}
</pre>
</div>
<pre class="example">
num: 0.0600514
num was less than 0.5.
num: 0.788318
num was greater than 0.5.
</pre>
</div>
</div>
<div id="outline-container-orgheadline26" class="outline-4">
<h4 id="orgheadline26"><span class="section-number-4">4.1.2</span> Example: <code>if</code> / <code>else if</code> / <code>else</code></h4>
<div class="outline-text-4" id="text-4-1-2">
<p>
You can follow <code>else</code> immediate by another <code>if</code> to have multiple mutually-
exclusive blocks:
</p>
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr"> 1: </span><span style="color: #0072b2;">#include</span> <span style="color: #848ea9;"><iostream></span>
<span class="linenr"> 2: </span><span style="color: #0072b2;">#include</span> <span style="color: #848ea9;"><stdlib.h></span>
<span class="linenr"> 3: </span><span style="color: #0072b2;">#include</span> <span style="color: #848ea9;"><time.h></span>
<span class="linenr"> 4: </span>
<span class="linenr"> 5: </span><span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #d55e00;">main</span>() {
<span class="linenr"> 6: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Seed the random number generator based on the current time.</span>
<span class="linenr"> 7: </span> srand(time(<span style="color: #d55e00; font-weight: bold;">NULL</span>));
<span class="linenr"> 8: </span>
<span class="linenr"> 9: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">rand() produces a random integer between 0 and RAND_MAX.</span>
<span class="linenr">10: </span> <span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #e69f00; font-weight: bold;">num</span> = rand() / ((<span style="color: #0072b2; font-weight: bold;">double</span>) RAND_MAX);
<span class="linenr">11: </span>
<span class="linenr">12: </span> <span style="color: #d55e00; font-weight: bold;">std</span>::cout << <span style="color: #848ea9;">"num: "</span> << num << <span style="color: #848ea9;">"\n"</span>;
<span class="linenr">13: </span>
<span class="linenr">14: </span> <span style="color: #56b4e9; font-weight: bold;">if</span> (num >= 0.75) {
<span class="linenr">15: </span> <span style="color: #d55e00; font-weight: bold;">std</span>::cout << <span style="color: #848ea9;">"num was between 0.75 and 1.\n"</span>;
<span class="linenr">16: </span> }
<span class="linenr">17: </span> <span style="color: #56b4e9; font-weight: bold;">else</span> <span style="color: #56b4e9; font-weight: bold;">if</span> (num >= 0.5) {
<span class="linenr">18: </span> <span style="color: #d55e00; font-weight: bold;">std</span>::cout << <span style="color: #848ea9;">"num was between 0.5 and 0.75."</span>;
<span class="linenr">19: </span> }
<span class="linenr">20: </span> <span style="color: #56b4e9; font-weight: bold;">else</span> <span style="color: #56b4e9; font-weight: bold;">if</span> (num >= 0.25) {
<span class="linenr">21: </span> <span style="color: #d55e00; font-weight: bold;">std</span>::cout << <span style="color: #848ea9;">"num was between 0.25 and 0.5."</span>;
<span class="linenr">22: </span> }
<span class="linenr">23: </span> <span style="color: #56b4e9; font-weight: bold;">else</span> {
<span class="linenr">24: </span> <span style="color: #d55e00; font-weight: bold;">std</span>::cout << <span style="color: #848ea9;">"num was between 0 and 0.25"</span>;
<span class="linenr">25: </span> }
<span class="linenr">26: </span>
<span class="linenr">27: </span> <span style="color: #56b4e9; font-weight: bold;">return</span> 0;
<span class="linenr">28: </span>}
</pre>
</div>
<pre class="example">
num: 0.0456405
num was between 0 and 0.25
</pre>
<p>
The conditions are checked in the order that they're written. So, for example,
in the second condition, we don't need to specify <code>num >= 0.5 && num < 0.75</code>
because we know that this condition will only be checked if the previous
was false.
</p>
</div>
</div>
</div>
<div id="outline-container-orgheadline33" class="outline-3">
<h3 id="orgheadline33"><span class="section-number-3">4.2</span> Loops</h3>
<div class="outline-text-3" id="text-4-2">
<p>
We discuss two main structures for iterating – the <code>for</code> and <code>while</code> loops.
</p>
</div>
<div id="outline-container-orgheadline28" class="outline-4">
<h4 id="orgheadline28"><span class="section-number-4">4.2.1</span> The <code>for</code> loop</h4>
<div class="outline-text-4" id="text-4-2-1">
<p>
The <code>for</code> loop requires three specifications – the iteration variable
initialization, the termination condition, and the update rule. The body
of the loop follows these three specifications. Shown below, we declare
an array; assign to its components; and, print the current component to
the screen.
</p>
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr"> 1: </span><span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">length</span> = 11;
<span class="linenr"> 2: </span><span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #e69f00; font-weight: bold;">x</span>[length];
<span class="linenr"> 3: </span>
<span class="linenr"> 4: </span><span style="color: #56b4e9; font-weight: bold;">for</span>(<span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">i</span>=0; i < length; i++) {
<span class="linenr"> 5: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Assign to each array component.</span>
<span class="linenr"> 6: </span> x[i] = (<span style="color: #0072b2; font-weight: bold;">double</span>) i / (length - 1);
<span class="linenr"> 7: </span>
<span class="linenr"> 8: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Print the current component.</span>
<span class="linenr"> 9: </span> <span style="color: #d55e00; font-weight: bold;">std</span>::cout << <span style="color: #848ea9;">"x["</span> << i << <span style="color: #848ea9;">"] = "</span> << x[i] << <span style="color: #d55e00; font-weight: bold;">std</span>::endl;
<span class="linenr">10: </span>}
</pre>
</div>
<pre class="example">
x[0] = 0
x[1] = 0.1
x[2] = 0.2
x[3] = 0.3
x[4] = 0.4
x[5] = 0.5
x[6] = 0.6
x[7] = 0.7
x[8] = 0.8
x[9] = 0.9
x[10] = 1
</pre>
</div>
<div id="outline-container-orgheadline27" class="outline-5">
<h5 id="orgheadline27"><span class="section-number-5">4.2.1.1</span> Example: row-major matrix</h5>
<div class="outline-text-5" id="text-4-2-1-1">
<p>
You can nest loops, i.e., loops inside of loops, etc.
</p>
<p>
Below, is an example of a double loop for creating and accessing
matrix data stored in a flat array. The matrix data is stored in
<a href="https://en.wikipedia.org/wiki/Row-major-order">row-major order</a>. This means the first <code>n_cols</code> elements of the
array named, <code>matrix</code>, will contain the first row of the matrix,
the second <code>n_cols</code> elements of <code>matrix</code> will contain the second row, etc.
</p>
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr"> 1: </span><span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">n_rows</span> = 4;
<span class="linenr"> 2: </span><span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">n_cols</span> = 3;
<span class="linenr"> 3: </span>
<span class="linenr"> 4: </span><span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Row-major matrix array.</span>
<span class="linenr"> 5: </span><span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #e69f00; font-weight: bold;">matrix</span> [n_rows*n_cols];
<span class="linenr"> 6: </span>
<span class="linenr"> 7: </span><span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">temporary index.</span>
<span class="linenr"> 8: </span><span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">k</span>;
<span class="linenr"> 9: </span>
<span class="linenr">10: </span><span style="color: #56b4e9; font-weight: bold;">for</span>(<span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">i</span>=0; i < n_rows; i++) {
<span class="linenr">11: </span> <span style="color: #56b4e9; font-weight: bold;">for</span>(<span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">j</span>=0; j < n_cols; j++) {
<span class="linenr">12: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Convert the (i,j) matrix index to the "flat" row-major index.</span>
<span class="linenr">13: </span> k = i*n_cols + j;
<span class="linenr">14: </span>
<span class="linenr">15: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Assign a value of 1.0 to the diagonal,</span>
<span class="linenr">16: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">2 to the off-diagonal, and 0 otherwise.</span>
<span class="linenr">17: </span> <span style="color: #56b4e9; font-weight: bold;">if</span> (i == j) {
<span class="linenr">18: </span> matrix[k] = 1.0;
<span class="linenr">19: </span> }
<span class="linenr">20: </span> <span style="color: #56b4e9; font-weight: bold;">else</span> <span style="color: #56b4e9; font-weight: bold;">if</span> ((i == (j+1)) || (i == (j-1))){
<span class="linenr">21: </span> matrix[k] = 2.0;
<span class="linenr">22: </span> }
<span class="linenr">23: </span> <span style="color: #56b4e9; font-weight: bold;">else</span> {
<span class="linenr">24: </span> matrix[k] = 0.0;
<span class="linenr">25: </span> }
<span class="linenr">26: </span> }
<span class="linenr">27: </span>}
<span class="linenr">28: </span>
<span class="linenr">29: </span>
<span class="linenr">30: </span><span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Print the matrix elements.</span>
<span class="linenr">31: </span><span style="color: #56b4e9; font-weight: bold;">for</span>(<span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">i</span>=0; i < n_rows; i++) {
<span class="linenr">32: </span> <span style="color: #56b4e9; font-weight: bold;">for</span>(<span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">j</span>=0; j < n_cols; j++) {
<span class="linenr">33: </span> k = i*n_cols + j;
<span class="linenr">34: </span>
<span class="linenr">35: </span> <span style="color: #d55e00; font-weight: bold;">std</span>::cout << matrix[k];
<span class="linenr">36: </span> <span style="color: #56b4e9; font-weight: bold;">if</span> (j != (n_cols-1)) {
<span class="linenr">37: </span> <span style="color: #d55e00; font-weight: bold;">std</span>::cout << <span style="color: #848ea9;">", "</span>;
<span class="linenr">38: </span> }
<span class="linenr">39: </span> }
<span class="linenr">40: </span>
<span class="linenr">41: </span> <span style="color: #56b4e9; font-weight: bold;">if</span> (i != (n_rows-1)) {
<span class="linenr">42: </span> <span style="color: #d55e00; font-weight: bold;">std</span>::cout << <span style="color: #848ea9;">"\n"</span>;
<span class="linenr">43: </span> }
<span class="linenr">44: </span>}
</pre>
</div>
<pre class="example">
1, 2, 0
2, 1, 2
0, 2, 1
0, 0, 2
</pre>
</div>
</div>
</div>
<div id="outline-container-orgheadline31" class="outline-4">
<h4 id="orgheadline31"><span class="section-number-4">4.2.2</span> The <code>while</code> loop</h4>
<div class="outline-text-4" id="text-4-2-2">
<p>
A <code>while</code> loop iterates while a condition is <code>true</code>. Essentially, it is a <code>for</code>
loop without an update variable.
</p>
</div>
<div id="outline-container-orgheadline29" class="outline-5">
<h5 id="orgheadline29"><span class="section-number-5">4.2.2.1</span> Example: truncated sum</h5>
<div class="outline-text-5" id="text-4-2-2-1">
<p>
In the following example, we approximate the geometric series:
</p>
<p>
\[
1 = \sum_{n=1}^{\infty} \left(\frac{1}{2}\right)^n
\]
</p>
<p>
The loop exits when the absolute error,
</p>
<p>
\[
\text{absolute error} := 1-\sum_{n=1}^N \left(\frac{1}{2}\right)^n
\]
</p>
<p>
is less than some specified tolerance, <code>tol</code>.
</p>
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr"> 1: </span><span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #e69f00; font-weight: bold;">sum</span> = 0.0;
<span class="linenr"> 2: </span><span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #e69f00; font-weight: bold;">base</span> = 0.5;
<span class="linenr"> 3: </span><span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #e69f00; font-weight: bold;">pow</span> = base; <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">initialize to base^1</span>
<span class="linenr"> 4: </span><span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #e69f00; font-weight: bold;">tol</span> = 1e-4;
<span class="linenr"> 5: </span><span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">iter</span> = 1;
<span class="linenr"> 6: </span>
<span class="linenr"> 7: </span><span style="color: #56b4e9; font-weight: bold;">while</span>((1-sum) >= tol) {
<span class="linenr"> 8: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Add `pow` to `sum`.</span>
<span class="linenr"> 9: </span> sum += pow;
<span class="linenr">10: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Update `pow` by one power of `base`.</span>
<span class="linenr">11: </span> pow *= base;
<span class="linenr">12: </span>
<span class="linenr">13: </span> printf(<span style="color: #848ea9;">"Iter: %03d, Sum: %.5f, Abs Err: %.5f\n"</span>, iter, sum, 1-sum);
<span class="linenr">14: </span>
<span class="linenr">15: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Update the `iter` val by 1.</span>
<span class="linenr">16: </span> iter += 1;
<span class="linenr">17: </span>}
</pre>
</div>
<pre class="example">
Iter: 001, Sum: 0.50000, Abs Err: 0.50000
Iter: 002, Sum: 0.75000, Abs Err: 0.25000
Iter: 003, Sum: 0.87500, Abs Err: 0.12500
Iter: 004, Sum: 0.93750, Abs Err: 0.06250
Iter: 005, Sum: 0.96875, Abs Err: 0.03125
Iter: 006, Sum: 0.98438, Abs Err: 0.01562
Iter: 007, Sum: 0.99219, Abs Err: 0.00781
Iter: 008, Sum: 0.99609, Abs Err: 0.00391
Iter: 009, Sum: 0.99805, Abs Err: 0.00195
Iter: 010, Sum: 0.99902, Abs Err: 0.00098
Iter: 011, Sum: 0.99951, Abs Err: 0.00049
Iter: 012, Sum: 0.99976, Abs Err: 0.00024
Iter: 013, Sum: 0.99988, Abs Err: 0.00012
Iter: 014, Sum: 0.99994, Abs Err: 0.00006
</pre>
</div>
</div>
<div id="outline-container-orgheadline30" class="outline-5">
<h5 id="orgheadline30"><span class="section-number-5">4.2.2.2</span> Example: estimating machine epsilon</h5>
<div class="outline-text-5" id="text-4-2-2-2">
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr"> 1: </span><span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #e69f00; font-weight: bold;">eps</span> = 1;
<span class="linenr"> 2: </span><span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">count</span> = 1;
<span class="linenr"> 3: </span>
<span class="linenr"> 4: </span><span style="color: #56b4e9; font-weight: bold;">while</span>(1.0 + eps*0.5 > 1.0) {
<span class="linenr"> 5: </span> eps *= 0.5;
<span class="linenr"> 6: </span> count += 1;
<span class="linenr"> 7: </span>}
<span class="linenr"> 8: </span>
<span class="linenr"> 9: </span><span style="color: #d55e00; font-weight: bold;">std</span>::cout << eps << <span style="color: #848ea9;">", "</span> << <span style="color: #d55e00; font-weight: bold;">std</span>::<span style="color: #d55e00; font-weight: bold;">numeric_limits</span><<span style="color: #0072b2; font-weight: bold;">double</span>>::epsilon() << <span style="color: #848ea9;">"\n"</span>
<span class="linenr">10: </span> << count << <span style="color: #848ea9;">", "</span> << <span style="color: #d55e00; font-weight: bold;">std</span>::<span style="color: #d55e00; font-weight: bold;">numeric_limits</span><<span style="color: #0072b2; font-weight: bold;">double</span>>::digits;
</pre>
</div>
<pre class="example">
2.22045e-16, 2.22045e-16
53, 53
</pre>
</div>
</div>
</div>
<div id="outline-container-orgheadline32" class="outline-4">
<h4 id="orgheadline32"><span class="section-number-4">4.2.3</span> The <code>break</code> keyword</h4>
<div class="outline-text-4" id="text-4-2-3">
<p>
The <code>break</code> keyword provides a mechanism for exiting the direct parent loop
for which the <code>break</code> statement is placed. For example:
</p>
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr">1: </span><span style="color: #56b4e9; font-weight: bold;">for</span>(<span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">i</span>=0; i < 3; i++) {
<span class="linenr">2: </span> <span style="color: #56b4e9; font-weight: bold;">while</span>(<span style="color: #d55e00; font-weight: bold;">true</span>) {
<span class="linenr">3: </span> <span style="color: #d55e00; font-weight: bold;">std</span>::cout << <span style="color: #848ea9;">"Entering infinite loop number "</span> << (i+1) << <span style="color: #848ea9;">"\n"</span>;
<span class="linenr">4: </span> <span style="color: #56b4e9; font-weight: bold;">break</span>;
<span class="linenr">5: </span> }
<span class="linenr">6: </span> <span style="color: #d55e00; font-weight: bold;">std</span>::cout << <span style="color: #848ea9;">"We escaped the infinite loop!\n"</span>;
<span class="linenr">7: </span>}
</pre>
</div>
<pre class="example">
Entering infinite loop number 1
We escaped the infinite loop!
Entering infinite loop number 2
We escaped the infinite loop!
Entering infinite loop number 3
We escaped the infinite loop!
</pre>
<p>
The previous example is contrived, but there are situations, where you
might find the break statement within an infinite loop useful. Of course,
you should avoid this sort of thing if there is a more straight-forward
approach.
</p>
</div>
</div>
</div>
</div>
<div id="outline-container-orgheadline39" class="outline-2">
<h2 id="orgheadline39"><span class="section-number-2">5</span> Input / Output</h2>
<div class="outline-text-2" id="text-5">
<p>
We have already used the <code><iostream></code> library to print results to
the console. However, in many cases, we'd like to read in lots of
data from a file, pass option flags to the program from the command
line, or save the results of some computation to a file for further
analysis.
</p>
</div>
<div id="outline-container-orgheadline35" class="outline-3">
<h3 id="orgheadline35"><span class="section-number-3">5.1</span> Inputs to <code>main</code>: <code>argc</code> and <code>argv</code></h3>
<div class="outline-text-3" id="text-5-1">
<p>
The <code>main</code> function has two optional arguments which we have thus far omitted,
<code>argc</code> and <code>argv</code>. These arguments allow arguments to passed to the <code>main</code>
function when the program is run. This is how flags and other arguments are
passed to programs you use from the command line. The first, <code>argc</code>, is of
type, <code>int</code>, and stands for arg count. It gives the number of arguments
to the program. The arg count is always at least 1 because the program's
name is always the first argument. The second, <code>argv</code>, is a double pointer to
<code>char</code>. In essence, <code>argv</code> is an array of strings.
</p>
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr"> 1: </span><span style="color: #0072b2;">#include</span> <span style="color: #848ea9;"><iostream></span>
<span class="linenr"> 2: </span>
<span class="linenr"> 3: </span><span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #d55e00;">main</span>(<span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">argc</span>, <span style="color: #0072b2; font-weight: bold;">char</span> ** <span style="color: #e69f00; font-weight: bold;">argv</span>) {
<span class="linenr"> 4: </span> <span style="color: #d55e00; font-weight: bold;">std</span>::cout << <span style="color: #848ea9;">"argc = "</span> << argc << <span style="color: #848ea9;">"\n"</span>;
<span class="linenr"> 5: </span>
<span class="linenr"> 6: </span> <span style="color: #56b4e9; font-weight: bold;">for</span>(<span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">i</span>=0; i < argc; i++) {
<span class="linenr"> 7: </span> <span style="color: #d55e00; font-weight: bold;">std</span>::cout << <span style="color: #848ea9;">"argv["</span> << i << <span style="color: #848ea9;">"] = "</span> << argv[i] << <span style="color: #848ea9;">"\n"</span>;
<span class="linenr"> 8: </span> }
<span class="linenr"> 9: </span> <span style="color: #56b4e9; font-weight: bold;">return</span> 0;
<span class="linenr">10: </span>}
</pre>
</div>
<p>
Compile this program, and run, for example:
</p>
<div class="org-src-container">
<pre class="src src-bash">g++ main.cpp
./a.out hello 1 2 34
</pre>
</div>
<pre class="example">
argc = 5
argv[0] = ./a.out
argv[1] = hello
argv[2] = 1
argv[3] = 2
argv[4] = 34
</pre>
<p>
<code>argc</code> and <code>argv</code> are handy for setting up large experiments. You could, for
example, set up your main function so that different functions or parameters
or used based on the arguments of <code>arcv</code>. Then, you could set up a shell
script that loops through the desired arguments to be supplied to the main
function.
</p>
</div>
</div>
<div id="outline-container-orgheadline38" class="outline-3">
<h3 id="orgheadline38"><span class="section-number-3">5.2</span> Filestreams</h3>
<div class="outline-text-3" id="text-5-2">
<p>
File input and output is crucial for numerical experiments with lots of data.
In this section, we see how to read and write data to files.
</p>
</div>
<div id="outline-container-orgheadline36" class="outline-4">
<h4 id="orgheadline36"><span class="section-number-4">5.2.1</span> Reading data from a file</h4>
<div class="outline-text-4" id="text-5-2-1">
<p>
In general, how data is read in depends heavily on how the data is stored.
Nevertheless, we will give an example of reading in a vector stored in a
particular fashion. Suppose a text file exists in the directory,
<code>./data/vector.txt</code>, containing
</p>
<pre class="example">
1 2 3.14 4 5 6.28
</pre>
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr"> 1: </span><span style="color: #0072b2;">#include</span> <span style="color: #848ea9;"><iostream></span>
<span class="linenr"> 2: </span><span style="color: #0072b2;">#include</span> <span style="color: #848ea9;"><fstream></span>
<span class="linenr"> 3: </span>
<span class="linenr"> 4: </span><span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #d55e00;">main</span>() {
<span class="linenr"> 5: </span> <span style="color: #d55e00; font-weight: bold;">std</span>::<span style="color: #0072b2; font-weight: bold;">fstream</span> <span style="color: #e69f00; font-weight: bold;">fin</span>(<span style="color: #848ea9;">"./data/vector.txt"</span>, <span style="color: #d55e00; font-weight: bold;">std</span>::<span style="color: #d55e00; font-weight: bold;">ios_base</span>::in);
<span class="linenr"> 6: </span> <span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #e69f00; font-weight: bold;">vector</span>[6];
<span class="linenr"> 7: </span> <span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">i</span> = 0;
<span class="linenr"> 8: </span> <span style="color: #56b4e9; font-weight: bold;">while</span>(fin >> vector[i]) {
<span class="linenr"> 9: </span> <span style="color: #d55e00; font-weight: bold;">std</span>::cout << vector[i] << <span style="color: #848ea9;">" "</span>;
<span class="linenr">10: </span> i++;
<span class="linenr">11: </span> }
<span class="linenr">12: </span> <span style="color: #56b4e9; font-weight: bold;">return</span> 0;
<span class="linenr">13: </span>}
</pre>
</div>
<div class="org-src-container">
<pre class="src src-sh">g++ main.cpp && ./a.out
</pre>
</div>
<pre class="example">
1 2 3.14 4 5 6.28
</pre>
<p>
This simply prints the data in the file back out to the console. Note, however,
that the data is read into an array of type <code>double</code>, so it can be processed
numerically thereafter.
</p>
<p>
In this example dealt with simply stored data, and it was
assumed that the number of data entries was known beforehand. Parsing
data can become quite complicated depending on how it is stored, and
depending on the intended format of the data.
</p>
</div>
</div>
<div id="outline-container-orgheadline37" class="outline-4">
<h4 id="orgheadline37"><span class="section-number-4">5.2.2</span> Writing data to a file</h4>
<div class="outline-text-4" id="text-5-2-2">
<p>
Writing to a file is similar, using the <code><fstream></code> library.
</p>
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr"> 1: </span><span style="color: #0072b2;">#include</span> <span style="color: #848ea9;"><fstream></span>
<span class="linenr"> 2: </span><span style="color: #0072b2;">#include</span> <span style="color: #848ea9;"><cmath></span>
<span class="linenr"> 3: </span>
<span class="linenr"> 4: </span><span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #d55e00;">main</span>() {
<span class="linenr"> 5: </span> <span style="color: #d55e00; font-weight: bold;">std</span>::<span style="color: #0072b2; font-weight: bold;">fstream</span> <span style="color: #e69f00; font-weight: bold;">fout</span>(<span style="color: #848ea9;">"./data/new_shiny_data.txt"</span>, <span style="color: #d55e00; font-weight: bold;">std</span>::<span style="color: #d55e00; font-weight: bold;">ios_base</span>::out);
<span class="linenr"> 6: </span> <span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #e69f00; font-weight: bold;">x</span>;
<span class="linenr"> 7: </span>
<span class="linenr"> 8: </span> fout << <span style="color: #848ea9;">"x\tsin(x)\n"</span>;
<span class="linenr"> 9: </span>
<span class="linenr">10: </span> <span style="color: #56b4e9; font-weight: bold;">for</span>(<span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">i</span>=0; i < 11; i++) {
<span class="linenr">11: </span> x = i / 10.0;
<span class="linenr">12: </span> fout << x << <span style="color: #848ea9;">"\t"</span> << sin(x) << <span style="color: #848ea9;">"\n"</span>;
<span class="linenr">13: </span> }
<span class="linenr">14: </span>
<span class="linenr">15: </span> fout.close();
<span class="linenr">16: </span>
<span class="linenr">17: </span> <span style="color: #56b4e9; font-weight: bold;">return</span> 0;
<span class="linenr">18: </span>}
</pre>
</div>
<p>
This produces a file called <code>new_shiny_data.txt</code> in the folder, <code>data</code>,
containing:
</p>
<div class="org-src-container">
<pre class="src src-sh">x sin(x)
0 0
0.1 0.0998334
0.2 0.198669
0.3 0.29552
0.4 0.389418
0.5 0.479426
0.6 0.564642
0.7 0.644218
0.8 0.717356
0.9 0.783327
1 0.841471
</pre>
</div>
</div>
</div>
</div>
</div>
<div id="outline-container-orgheadline18" class="outline-2">
<h2 id="orgheadline18"><span class="section-number-2">6</span> Functions</h2>
<div class="outline-text-2" id="text-6">
<p>
So far, we've piled everything into the <code>main</code> function. When we have a block of
code used for a specific subtask, we can offload it to a function. This promotes
code which is separated based on the subtasks each block is intended to perform.
This, in turn, makes your code easier to debug and easier to understand.
</p>
</div>
<div id="outline-container-orgheadline41" class="outline-3">
<h3 id="orgheadline41"><span class="section-number-3">6.1</span> Writing a function</h3>
<div class="outline-text-3" id="text-6-1">
<p>
A function must be declared before use. Thus, a function usual consists of two
parts, a declaration and an implementation. You must declare the return type
of a function as well as the types of all the function's arguments. If the
function is defined in the same file as the <code>main</code> function, you should write
the declaration before <code>main</code> and the implementation after <code>main</code>.
</p>
</div>
<div id="outline-container-orgheadline40" class="outline-4">
<h4 id="orgheadline40"><span class="section-number-4">6.1.1</span> Example: <code>linspace</code>: generating a set of equally-spaced points</h4>
<div class="outline-text-4" id="text-6-1-1">
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr"> 1: </span><span style="color: #0072b2;">#include</span> <span style="color: #848ea9;"><iostream></span>
<span class="linenr"> 2: </span>
<span class="linenr"> 3: </span><span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">This is the function declaration.</span>
<span class="linenr"> 4: </span><span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">You should describe the functions arguments</span>
<span class="linenr"> 5: </span><span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">and what is returned by the function in comments</span>
<span class="linenr"> 6: </span><span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">near the declaration.</span>
<span class="linenr"> 7: </span><span style="color: #009e73; font-style: italic;">//</span>
<span class="linenr"> 8: </span><span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">`linspace` returns an array of doubles containing</span>
<span class="linenr"> 9: </span><span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">`n_points` entries which are equally-spaced, starting</span>
<span class="linenr">10: </span><span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">at `start` and ending at `stop`.</span>
<span class="linenr">11: </span><span style="color: #0072b2; font-weight: bold;">double</span> * <span style="color: #d55e00;">linspace</span>(<span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #e69f00; font-weight: bold;">start</span>, <span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #e69f00; font-weight: bold;">stop</span>, <span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">n_points</span>);
<span class="linenr">12: </span>
<span class="linenr">13: </span><span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">`void` is a function with no return type.</span>
<span class="linenr">14: </span><span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">`print_array` takes an array and prints it to std out.</span>
<span class="linenr">15: </span><span style="color: #0072b2; font-weight: bold;">void</span> <span style="color: #d55e00;">print_array</span>(<span style="color: #0072b2; font-weight: bold;">double</span> * <span style="color: #e69f00; font-weight: bold;">arr</span>, <span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">arr_len</span>);
<span class="linenr">16: </span>
<span class="linenr">17: </span><span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #d55e00;">main</span>() {
<span class="linenr">18: </span> <span style="color: #0072b2; font-weight: bold;">double</span> * <span style="color: #e69f00; font-weight: bold;">xs</span> = linspace(-1, 1, 5);
<span class="linenr">19: </span> print_array(xs, 5);
<span class="linenr">20: </span> <span style="color: #56b4e9; font-weight: bold;">delete</span> [] xs;
<span class="linenr">21: </span>
<span class="linenr">22: </span> <span style="color: #56b4e9; font-weight: bold;">return</span> 0;
<span class="linenr">23: </span>}
<span class="linenr">24: </span>
<span class="linenr">25: </span><span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Implementation of `linspace`.</span>
<span class="linenr">26: </span><span style="color: #0072b2; font-weight: bold;">double</span> * <span style="color: #d55e00;">linspace</span>(<span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #e69f00; font-weight: bold;">start</span>, <span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #e69f00; font-weight: bold;">stop</span>, <span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">n_points</span>) {
<span class="linenr">27: </span> <span style="color: #0072b2; font-weight: bold;">double</span> * <span style="color: #e69f00; font-weight: bold;">arr</span> = <span style="color: #56b4e9; font-weight: bold;">new</span> <span style="color: #0072b2; font-weight: bold;">double</span> [n_points];
<span class="linenr">28: </span> <span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #e69f00; font-weight: bold;">dx</span> = (stop-start) / (n_points-1.0);
<span class="linenr">29: </span>
<span class="linenr">30: </span> <span style="color: #56b4e9; font-weight: bold;">for</span>(<span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">i</span>=0; i < n_points; i++) {
<span class="linenr">31: </span> arr[i] = start + i*dx;
<span class="linenr">32: </span> }
<span class="linenr">33: </span>
<span class="linenr">34: </span> <span style="color: #56b4e9; font-weight: bold;">return</span> arr;
<span class="linenr">35: </span>}
<span class="linenr">36: </span>
<span class="linenr">37: </span><span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Implementation of `print_array`.</span>
<span class="linenr">38: </span><span style="color: #0072b2; font-weight: bold;">void</span> <span style="color: #d55e00;">print_array</span>(<span style="color: #0072b2; font-weight: bold;">double</span> * <span style="color: #e69f00; font-weight: bold;">arr</span>, <span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">arr_len</span>) {
<span class="linenr">39: </span> <span style="color: #56b4e9; font-weight: bold;">for</span>(<span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">i</span>=0; i < arr_len; i++) {
<span class="linenr">40: </span> <span style="color: #d55e00; font-weight: bold;">std</span>::cout << arr[i] << <span style="color: #848ea9;">"\n"</span>;
<span class="linenr">41: </span> }
<span class="linenr">42: </span>}
</pre>
</div>
<pre class="example">
-1
-0.5
0
0.5
1
</pre>
</div>
</div>
</div>
<div id="outline-container-orgheadline45" class="outline-3">
<h3 id="orgheadline45"><span class="section-number-3">6.2</span> Header and implementation files</h3>
<div class="outline-text-3" id="text-6-2">
<p>
The example in the previous section certainly made the <code>main</code> function
cleaner and simpler to understand, having only two function calls. However,
the file itself was still pretty messy. Thankfully, there is a way to modularize
further, by creating header and implementation files. Here is how we do it:
</p>
</div>
<div id="outline-container-orgheadline42" class="outline-4">
<h4 id="orgheadline42"><span class="section-number-4">6.2.1</span> The header file</h4>
<div class="outline-text-4" id="text-6-2-1">
<p>
Put the declarations from the into a header file, called <code>my_library.h</code>:
</p>
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr"> 1: </span><span style="color: #0072b2;">#if</span><span style="color: #0072b2;">n</span><span style="color: #0072b2;">def</span> MY_LIBRARY_H
<span class="linenr"> 2: </span><span style="color: #0072b2;">#define</span> <span style="color: #e69f00; font-weight: bold;">MY_LIBRARY_H</span>
<span class="linenr"> 3: </span>
<span class="linenr"> 4: </span><span style="color: #0072b2;">#include</span> <span style="color: #848ea9;"><iostream></span>
<span class="linenr"> 5: </span>
<span class="linenr"> 6: </span><span style="color: #56b4e9; font-weight: bold;">namespace</span> <span style="color: #d55e00; font-weight: bold;">my_namespace</span> {
<span class="linenr"> 7: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">`linspace` returns an array of doubles containing</span>
<span class="linenr"> 8: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">`n_points` entries which are equally-spaced, starting</span>
<span class="linenr"> 9: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">at `start` and ending at `stop`.</span>
<span class="linenr">10: </span> <span style="color: #0072b2; font-weight: bold;">double</span> * <span style="color: #d55e00;">linspace</span>(<span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #e69f00; font-weight: bold;">start</span>, <span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #e69f00; font-weight: bold;">stop</span>, <span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">n_points</span>);
<span class="linenr">11: </span>
<span class="linenr">12: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">`void` is a function with no return type.</span>
<span class="linenr">13: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">`print_array` takes an array and prints it to std out.</span>
<span class="linenr">14: </span> <span style="color: #0072b2; font-weight: bold;">void</span> <span style="color: #d55e00;">print_array</span>(<span style="color: #0072b2; font-weight: bold;">double</span> * <span style="color: #e69f00; font-weight: bold;">arr</span>, <span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">arr_len</span>);
<span class="linenr">15: </span>}
<span class="linenr">16: </span>
<span class="linenr">17: </span><span style="color: #0072b2;">#endif</span>
</pre>
</div>
<p>
Note the the function declarations are wrapped in conditional "macro"
statments, <code>#ifndef</code>, <code>#define</code>, and <code>#endif</code>. You can think of this
as protecting your library from being imported twice.
</p>
<p>
We have also introduced the notion of a <code>namespace</code> above. Namespaces
help to prevent naming clashes between separate libraries. When calling
a function from a particular namespace, you must write the namespace
followed by <code>::</code> and then the function name. This is why many standard
library functions like <code><iostream></code> begin with <code>std::</code>.
</p>
</div>
</div>
<div id="outline-container-orgheadline43" class="outline-4">
<h4 id="orgheadline43"><span class="section-number-4">6.2.2</span> The implementation file</h4>
<div class="outline-text-4" id="text-6-2-2">
<p>
Create a file called <code>my_library.cpp</code> containing the implementations as follows:
</p>
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr"> 1: </span><span style="color: #0072b2;">#include</span> <span style="color: #848ea9;">"my_library.h"</span>
<span class="linenr"> 2: </span>
<span class="linenr"> 3: </span><span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Implementation of `linspace`.</span>
<span class="linenr"> 4: </span><span style="color: #0072b2; font-weight: bold;">double</span> * <span style="color: #d55e00; font-weight: bold;">my_namespace</span>::<span style="color: #d55e00;">linspace</span>(<span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #e69f00; font-weight: bold;">start</span>, <span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #e69f00; font-weight: bold;">stop</span>, <span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">n_points</span>) {
<span class="linenr"> 5: </span> <span style="color: #0072b2; font-weight: bold;">double</span> * <span style="color: #e69f00; font-weight: bold;">arr</span> = <span style="color: #56b4e9; font-weight: bold;">new</span> <span style="color: #0072b2; font-weight: bold;">double</span> [n_points];
<span class="linenr"> 6: </span> <span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #e69f00; font-weight: bold;">dx</span> = (stop-start) / (n_points-1.0);
<span class="linenr"> 7: </span>
<span class="linenr"> 8: </span> <span style="color: #56b4e9; font-weight: bold;">for</span>(<span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">i</span>=0; i < n_points; i++) {
<span class="linenr"> 9: </span> arr[i] = start + i*dx;
<span class="linenr">10: </span> }
<span class="linenr">11: </span>
<span class="linenr">12: </span> <span style="color: #56b4e9; font-weight: bold;">return</span> arr;
<span class="linenr">13: </span>}
<span class="linenr">14: </span>
<span class="linenr">15: </span><span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Implementation of `print_array`.</span>
<span class="linenr">16: </span><span style="color: #0072b2; font-weight: bold;">void</span> <span style="color: #d55e00; font-weight: bold;">my_namespace</span>::<span style="color: #d55e00;">print_array</span>(<span style="color: #0072b2; font-weight: bold;">double</span> * <span style="color: #e69f00; font-weight: bold;">arr</span>, <span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">arr_len</span>) {
<span class="linenr">17: </span> <span style="color: #56b4e9; font-weight: bold;">for</span>(<span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">i</span>=0; i < arr_len; i++) {
<span class="linenr">18: </span> <span style="color: #d55e00; font-weight: bold;">std</span>::cout << arr[i] << <span style="color: #848ea9;">"\n"</span>;
<span class="linenr">19: </span> }
<span class="linenr">20: </span>}
</pre>
</div>
<p>
Note that we have to include the header file in quotations at the beginning,
and the names of the functions must be prepended by the namespace that we've
given in the header file.
</p>
</div>
</div>
<div id="outline-container-orgheadline44" class="outline-4">
<h4 id="orgheadline44"><span class="section-number-4">6.2.3</span> The file containing <code>main</code></h4>
<div class="outline-text-4" id="text-6-2-3">
<p>
Create a file with the main function, say <code>main.cpp</code>:
</p>
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr"> 1: </span><span style="color: #0072b2;">#include</span> <span style="color: #848ea9;"><iostream></span>
<span class="linenr"> 2: </span><span style="color: #0072b2;">#include</span> <span style="color: #848ea9;">"my_library.h"</span>
<span class="linenr"> 3: </span>
<span class="linenr"> 4: </span><span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #d55e00;">main</span>() {
<span class="linenr"> 5: </span> <span style="color: #0072b2; font-weight: bold;">double</span> * <span style="color: #e69f00; font-weight: bold;">xs</span> = <span style="color: #d55e00; font-weight: bold;">my_namespace</span>::linspace(-1,1,5);
<span class="linenr"> 6: </span> <span style="color: #d55e00; font-weight: bold;">my_namespace</span>::print_array(xs, 5);
<span class="linenr"> 7: </span> <span style="color: #56b4e9; font-weight: bold;">delete</span> [] xs;
<span class="linenr"> 8: </span>
<span class="linenr"> 9: </span> <span style="color: #56b4e9; font-weight: bold;">return</span> 0;
<span class="linenr">10: </span>}
</pre>
</div>
<p>
Now the main function is very nice and clean, but now we 3 separate files we
must compile into one executable. This is done as follows:
</p>
<div class="org-src-container">
<pre class="src src-bash"><span style="color: #009e73; font-style: italic;"># </span><span style="color: #009e73; font-style: italic;">Convert the library into an object file.</span>
g++ -c my_library.cpp
<span style="color: #009e73; font-style: italic;"># </span><span style="color: #009e73; font-style: italic;">Compile the main to an executable.</span>
g++ my_library.o main.cpp
<span style="color: #009e73; font-style: italic;"># </span><span style="color: #009e73; font-style: italic;">Run it.</span>
./a.out
</pre>
</div>
<pre class="example">
-1
-0.5
0
0.5
1
</pre>
<p>
If successful, you will see the same output <a href="#orgheadline41">as previously</a>.
</p>
</div>
</div>
</div>
<div id="outline-container-orgheadline48" class="outline-3">
<h3 id="orgheadline48"><span class="section-number-3">6.3</span> Function pointers</h3>
<div class="outline-text-3" id="text-6-3">
<p>
Pointers can be made to functions, and these function pointers can be used
as arguments to other functions. We'll look at two functions that accept a
function pointer as one of their arguments.
</p>
</div>
<div id="outline-container-orgheadline46" class="outline-4">
<h4 id="orgheadline46"><span class="section-number-4">6.3.1</span> Example: Newton's method for rootfinding</h4>
<div class="outline-text-4" id="text-6-3-1">
<p>
Suppose \(f: \mathbb{R} \to \mathbb{R}\), and we'd like to find a root of \(f\).
Newton's method is an iterative method for finding roots, which, starting
from some initial guess, \(x_0\), iterates:
</p>
<p>
\[
x_{n+1} \leftarrow x_n - \frac{f(x_n)}{f'(x_n)}
\]
</p>
<p>
For simplicity, we'll dump everything into the file containing <code>main</code>, but
you could imagine a libary with many methods for finding roots, which would
contain Newton's method.
</p>
<p>
Let's consider \(f(x) = x^2 - 2\).
</p>
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr"> 1: </span><span style="color: #0072b2;">#include</span> <span style="color: #848ea9;"><cmath></span>
<span class="linenr"> 2: </span><span style="color: #0072b2;">#include</span> <span style="color: #848ea9;"><iostream></span>
<span class="linenr"> 3: </span>
<span class="linenr"> 4: </span><span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">The function to find the root of.</span>
<span class="linenr"> 5: </span><span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #d55e00;">func</span>(<span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #e69f00; font-weight: bold;">x</span>);
<span class="linenr"> 6: </span><span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Its derivative.</span>
<span class="linenr"> 7: </span><span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #d55e00;">dfunc</span>(<span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #e69f00; font-weight: bold;">x</span>);
<span class="linenr"> 8: </span>
<span class="linenr"> 9: </span><span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Find the root of `f` using Newton's method,</span>
<span class="linenr">10: </span><span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">starting from `x0` until |f(x)| < `tol` or `max_iters`</span>
<span class="linenr">11: </span><span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">is reached.</span>
<span class="linenr">12: </span><span style="color: #009e73; font-style: italic;">//</span>
<span class="linenr">13: </span><span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Note the first and second arguments are function pointers.</span>
<span class="linenr">14: </span><span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #d55e00;">newton_root</span>(<span style="color: #0072b2; font-weight: bold;">double</span> (*<span style="color: #e69f00; font-weight: bold;">f</span>)(<span style="color: #0072b2; font-weight: bold;">double</span>), <span style="color: #0072b2; font-weight: bold;">double</span> (*<span style="color: #e69f00; font-weight: bold;">df</span>)(<span style="color: #0072b2; font-weight: bold;">double</span>), <span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #e69f00; font-weight: bold;">x0</span>,
<span class="linenr">15: </span> <span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #e69f00; font-weight: bold;">tol</span>, <span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">max_iters</span>, <span style="color: #0072b2; font-weight: bold;">bool</span> <span style="color: #e69f00; font-weight: bold;">print_iters</span>);
<span class="linenr">16: </span>
<span class="linenr">17: </span><span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #d55e00;">main</span>() {
<span class="linenr">18: </span> <span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #e69f00; font-weight: bold;">x</span> = newton_root(&func, &dfunc, 1.0, 1e-6, 1000, <span style="color: #d55e00; font-weight: bold;">true</span>);
<span class="linenr">19: </span>
<span class="linenr">20: </span> <span style="color: #56b4e9; font-weight: bold;">return</span> 0;
<span class="linenr">21: </span>}
<span class="linenr">22: </span>
<span class="linenr">23: </span><span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #d55e00;">func</span>( <span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #e69f00; font-weight: bold;">x</span>) { <span style="color: #56b4e9; font-weight: bold;">return</span> x*x - 2; }
<span class="linenr">24: </span><span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #d55e00;">dfunc</span>(<span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #e69f00; font-weight: bold;">x</span>) { <span style="color: #56b4e9; font-weight: bold;">return</span> 2*x; }
<span class="linenr">25: </span>
<span class="linenr">26: </span><span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #d55e00;">newton_root</span>(<span style="color: #0072b2; font-weight: bold;">double</span> (*<span style="color: #e69f00; font-weight: bold;">f</span>)(<span style="color: #0072b2; font-weight: bold;">double</span>), <span style="color: #0072b2; font-weight: bold;">double</span> (*<span style="color: #e69f00; font-weight: bold;">df</span>)(<span style="color: #0072b2; font-weight: bold;">double</span>), <span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #e69f00; font-weight: bold;">x0</span>,
<span class="linenr">27: </span> <span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #e69f00; font-weight: bold;">tol</span>, <span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">max_iters</span>, <span style="color: #0072b2; font-weight: bold;">bool</span> <span style="color: #e69f00; font-weight: bold;">print_iters</span>) {
<span class="linenr">28: </span> <span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #e69f00; font-weight: bold;">x</span> = x0;
<span class="linenr">29: </span> <span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">iter</span> = 0;
<span class="linenr">30: </span>
<span class="linenr">31: </span> <span style="color: #56b4e9; font-weight: bold;">while</span> (<span style="color: #d55e00; font-weight: bold;">std</span>::abs(f(x)) > tol && iter < max_iters) {
<span class="linenr">32: </span> <span style="color: #56b4e9; font-weight: bold;">if</span> (print_iters) {
<span class="linenr">33: </span> <span style="color: #d55e00; font-weight: bold;">std</span>::cout << <span style="color: #848ea9;">"f("</span> << x << <span style="color: #848ea9;">") = "</span> << f(x) << <span style="color: #848ea9;">"\n"</span>;
<span class="linenr">34: </span> }
<span class="linenr">35: </span>
<span class="linenr">36: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Newton's method update.</span>
<span class="linenr">37: </span> x -= f(x) / df(x);
<span class="linenr">38: </span> iter++;
<span class="linenr">39: </span> }
<span class="linenr">40: </span>
<span class="linenr">41: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">One last print if necessary.</span>
<span class="linenr">42: </span> <span style="color: #56b4e9; font-weight: bold;">if</span> (print_iters) {
<span class="linenr">43: </span> <span style="color: #d55e00; font-weight: bold;">std</span>::cout << <span style="color: #848ea9;">"f("</span> << x << <span style="color: #848ea9;">") = "</span> << f(x) << <span style="color: #848ea9;">"\n"</span>;
<span class="linenr">44: </span> }
<span class="linenr">45: </span>
<span class="linenr">46: </span> <span style="color: #56b4e9; font-weight: bold;">return</span> x;
<span class="linenr">47: </span>}
</pre>
</div>
<pre class="example">
f(1) = -1
f(1.5) = 0.25
f(1.41667) = 0.00694444
f(1.41422) = 6.0073e-06
f(1.41421) = 4.51061e-12
</pre>
</div>
</div>
<div id="outline-container-orgheadline47" class="outline-4">
<h4 id="orgheadline47"><span class="section-number-4">6.3.2</span> Example: The midpoint rule for definite integrals</h4>
<div class="outline-text-4" id="text-6-3-2">
<p>
The midpoint rule is a numerical integration method which approximates
the definite integral of a specified function over a specified interval
using a specified number of subintervals where on each subinterval, the
area under the curve is approximated by a rectangle whose width is the
width of the subinterval and whose height is the height of the function
at the midpoint between the points defining the end points of the subinterval.
</p>
<p>
Specifically, if \(n\) equally-sized subintervals are used on \([a,b]\), then
the midpoint rule approximation, \(M_n\), to the definite integral of \(f(x)\)
on \([a,b]\) is:
</p>
<p>
\[
\int_a^b f(x) \; dx \approx \sum_{i=1}^n f\left( \frac{x_{i-1}+x_i}{2} \right) \Delta x =: M_n
\]
</p>
<p>
where \(\Delta x = \frac{b-a}{n}\), and \(x_i = a + i \cdot \Delta x, \;\; i=0, 1, \ldots, n\).
</p>
<p>
Let's consider \(f(x) = \frac{1}{x}\) on \([1, e]\).
</p>
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr"> 1: </span><span style="color: #0072b2;">#include</span> <span style="color: #848ea9;"><iostream></span>
<span class="linenr"> 2: </span><span style="color: #0072b2;">#include</span> <span style="color: #848ea9;"><cmath></span>
<span class="linenr"> 3: </span>
<span class="linenr"> 4: </span><span style="color: #56b4e9; font-weight: bold;">const</span> <span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #e69f00; font-weight: bold;">E</span> = <span style="color: #d55e00; font-weight: bold;">std</span>::exp(1.0);
<span class="linenr"> 5: </span>
<span class="linenr"> 6: </span><span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">The function to be integrated.</span>
<span class="linenr"> 7: </span><span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #d55e00;">func</span>(<span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #e69f00; font-weight: bold;">x</span>);
<span class="linenr"> 8: </span>
<span class="linenr"> 9: </span><span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Compute the midpoint rule approximation to</span>
<span class="linenr">10: </span><span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">the definite integral of `f` from `a` to `b`</span>
<span class="linenr">11: </span><span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">using `n` subintervals.</span>
<span class="linenr">12: </span><span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #d55e00;">midpoint_rule</span>(<span style="color: #0072b2; font-weight: bold;">double</span> (*<span style="color: #e69f00; font-weight: bold;">f</span>)(<span style="color: #0072b2; font-weight: bold;">double</span>), <span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #e69f00; font-weight: bold;">a</span>, <span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #e69f00; font-weight: bold;">b</span>, <span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">n</span>);
<span class="linenr">13: </span>
<span class="linenr">14: </span>
<span class="linenr">15: </span><span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #d55e00;">main</span>() {
<span class="linenr">16: </span> <span style="color: #56b4e9; font-weight: bold;">for</span>(<span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">n</span>=2; n <= 20; n += 2) {
<span class="linenr">17: </span> <span style="color: #d55e00; font-weight: bold;">std</span>::cout << <span style="color: #848ea9;">"n = "</span> << n << <span style="color: #848ea9;">", "</span>
<span class="linenr">18: </span> << <span style="color: #848ea9;">"M_n = "</span> << midpoint_rule(&func, 1, E, n) << <span style="color: #848ea9;">"\n"</span>;
<span class="linenr">19: </span> }
<span class="linenr">20: </span>
<span class="linenr">21: </span> <span style="color: #56b4e9; font-weight: bold;">return</span> 0;
<span class="linenr">22: </span>}
<span class="linenr">23: </span>
<span class="linenr">24: </span><span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #d55e00;">func</span>(<span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #e69f00; font-weight: bold;">x</span>) { <span style="color: #56b4e9; font-weight: bold;">return</span> 1.0 / x; }
<span class="linenr">25: </span>
<span class="linenr">26: </span><span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #d55e00;">midpoint_rule</span>(<span style="color: #0072b2; font-weight: bold;">double</span> (*<span style="color: #e69f00; font-weight: bold;">f</span>)(<span style="color: #0072b2; font-weight: bold;">double</span>), <span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #e69f00; font-weight: bold;">a</span>, <span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #e69f00; font-weight: bold;">b</span>, <span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">n</span>) {
<span class="linenr">27: </span> <span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #e69f00; font-weight: bold;">xi</span>;
<span class="linenr">28: </span> <span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #e69f00; font-weight: bold;">xi_prev</span> = a;
<span class="linenr">29: </span> <span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #e69f00; font-weight: bold;">dx</span> = (b-a) / n;
<span class="linenr">30: </span> <span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #e69f00; font-weight: bold;">sum</span>;
<span class="linenr">31: </span>
<span class="linenr">32: </span> <span style="color: #56b4e9; font-weight: bold;">for</span>(<span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">i</span>=1; i <= n; i++) {
<span class="linenr">33: </span> xi = a + i*dx;
<span class="linenr">34: </span> sum += f(0.5*(xi_prev + xi));
<span class="linenr">35: </span> xi_prev = xi;
<span class="linenr">36: </span> }
<span class="linenr">37: </span>
<span class="linenr">38: </span> <span style="color: #56b4e9; font-weight: bold;">return</span> sum*dx;
<span class="linenr">39: </span>}
</pre>
</div>
<pre class="example">
n = 2, M_n = 0.97636
n = 4, M_n = 0.993575
n = 6, M_n = 0.997091
n = 8, M_n = 0.998353
n = 10, M_n = 0.998942
n = 12, M_n = 0.999264
n = 14, M_n = 0.999459
n = 16, M_n = 0.999585
n = 18, M_n = 0.999672
n = 20, M_n = 0.999734
</pre>
</div>
</div>
</div>
</div>
<div id="outline-container-orgheadline9" class="outline-2">
<h2 id="orgheadline9"><span class="section-number-2">7</span> Object-oriented programming</h2>
<div class="outline-text-2" id="text-7">
<p>
New data types can be created by writing a new <code>class</code>. A <code>class</code> has
state variables and functions that act on the state variables. An instance
of a <code>class</code> is called an <b>object</b>. Let's write a <code>vector</code> class that
improves upon the default <code>double</code> array.
</p>
</div>
<div id="outline-container-orgheadline59" class="outline-3">
<h3 id="orgheadline59"><span class="section-number-3">7.1</span> Example: a vector <code>class</code></h3>
<div class="outline-text-3" id="text-7-1">
</div><div id="outline-container-orgheadline49" class="outline-4">
<h4 id="orgheadline49"><span class="section-number-4">7.1.1</span> The header file</h4>
<div class="outline-text-4" id="text-7-1-1">
<p>
Create the header file, <code>vector.h</code>:
</p>
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr"> 1: </span><span style="color: #0072b2;">#if</span><span style="color: #0072b2;">n</span><span style="color: #0072b2;">def</span> VECTOR_H
<span class="linenr"> 2: </span><span style="color: #0072b2;">#define</span> <span style="color: #e69f00; font-weight: bold;">VECTOR_H</span>
<span class="linenr"> 3: </span>
<span class="linenr"> 4: </span><span style="color: #56b4e9; font-weight: bold;">namespace</span> <span style="color: #d55e00; font-weight: bold;">vec</span> {
<span class="linenr"> 5: </span> <span style="color: #56b4e9; font-weight: bold;">class</span> <span style="color: #0072b2; font-weight: bold;">vector</span> {
<span class="linenr"> 6: </span> <span style="color: #56b4e9; font-weight: bold;">public</span>:
<span class="linenr"> 7: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Constructor. This function is called when the object is created.</span>
<span class="linenr"> 8: </span> <span style="color: #d55e00;">vector</span>(<span style="color: #0072b2; font-weight: bold;">unsigned</span> <span style="color: #e69f00; font-weight: bold;">len</span>);
<span class="linenr"> 9: </span>
<span class="linenr">10: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Destructor. This function is called when the object is destroyed.</span>
<span class="linenr">11: </span> ~<span style="color: #d55e00;">vector</span>();
<span class="linenr">12: </span>
<span class="linenr">13: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">length accessor.</span>
<span class="linenr">14: </span> <span style="color: #0072b2; font-weight: bold;">unsigned</span> <span style="color: #d55e00;">len</span>() <span style="color: #56b4e9; font-weight: bold;">const</span>;
<span class="linenr">15: </span>
<span class="linenr">16: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">data accessor.</span>
<span class="linenr">17: </span> <span style="color: #0072b2; font-weight: bold;">double</span> & <span style="color: #d55e00;">element</span>(<span style="color: #0072b2; font-weight: bold;">unsigned</span> <span style="color: #e69f00; font-weight: bold;">i</span>) <span style="color: #56b4e9; font-weight: bold;">const</span>;
<span class="linenr">18: </span>
<span class="linenr">19: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Simple print function.</span>
<span class="linenr">20: </span> <span style="color: #0072b2; font-weight: bold;">void</span> <span style="color: #d55e00;">print</span>() <span style="color: #56b4e9; font-weight: bold;">const</span>;
<span class="linenr">21: </span>
<span class="linenr">22: </span> <span style="color: #56b4e9; font-weight: bold;">private</span>:
<span class="linenr">23: </span> <span style="color: #0072b2; font-weight: bold;">unsigned</span> <span style="color: #e69f00; font-weight: bold;">length</span>;
<span class="linenr">24: </span> <span style="color: #0072b2; font-weight: bold;">double</span> * <span style="color: #e69f00; font-weight: bold;">data</span>;
<span class="linenr">25: </span> <span style="color: #0072b2; font-weight: bold;">void</span> <span style="color: #d55e00;">check_index</span>(<span style="color: #0072b2; font-weight: bold;">unsigned</span> <span style="color: #e69f00; font-weight: bold;">i</span>) <span style="color: #56b4e9; font-weight: bold;">const</span>;
<span class="linenr">26: </span> };
<span class="linenr">27: </span>}
<span class="linenr">28: </span><span style="color: #0072b2;">#endif</span>
</pre>
</div>
<p>
First note the macro guards, <code>#ifndef</code>, <code>#define</code>, and <code>#endif</code>, as well as the
namespace, <code>vec</code>, wrapping the <code>class</code> declaration. Macro guards and namespaces
were previously discussed when we initially introduced
<a href="#orgheadline45">how to write header and implementation files</a>.
</p>
<p>
Now, within the namespace, we've declared a class, <code>vector</code>, which
contains <code>public</code> and <code>private</code> variables and function declarations. Private
functions and variables may only be accessed through the public methods. This
means if you created an instance of the class, <code>vector</code>, you would not be able
to access the private variable directly. You could only call the <b>public
member-functions</b>, which, in turn, may manipulate the
<b>private member-variables</b>, or call the <b>private member-functions</b>.
Splitting variables and functions into public and private helps to ensure
that other libraries and programs use your class as intended.
</p>
<p>
Thus far, this class has 5 public member-functions, 2 private member-variables,
and 1 private member-function. The first two member functions are special,
the <b>constructor</b> and <b>destructor</b>, respectively. The constructor is called
explicitly when you declare a new instance of this class, while the destructor
is usually called implicitly when the object is deleted or when it goes out of
scope.
</p>
<p>
Notice that the method for accessing elements of <code>vector</code> is called <code>element</code>
and its return type is a <b>reference</b> (<a href="#orgheadline19">discussed previously</a>). This allows us
to use this function on both the left side of assignment operators, i.e., to
assign values to <code>vector</code> components, and on the right side of assignments,
i.e., to treat it as the value of the component.
</p>
<p>
Finally, notice that some member function declarations end with the keyword,
<code>const</code>. Functions with such a signature are not allowed to modify member
variables, and they are also not allowed to call other non <code>const</code> member
functions.
</p>
</div>
</div>
<div id="outline-container-orgheadline50" class="outline-4">
<h4 id="orgheadline50"><span class="section-number-4">7.1.2</span> The implementation file</h4>
<div class="outline-text-4" id="text-7-1-2">
<p>
Create the implementation file, <code>vector.cpp</code>:
</p>
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr"> 1: </span><span style="color: #0072b2;">#include</span> <span style="color: #848ea9;"><iostream></span>
<span class="linenr"> 2: </span><span style="color: #0072b2;">#include</span> <span style="color: #848ea9;"><cstdlib></span>
<span class="linenr"> 3: </span><span style="color: #0072b2;">#include</span> <span style="color: #848ea9;">"vector.h"</span>
<span class="linenr"> 4: </span>
<span class="linenr"> 5: </span><span style="color: #56b4e9; font-weight: bold;">namespace</span> <span style="color: #d55e00; font-weight: bold;">vec</span> {
<span class="linenr"> 6: </span> <span style="color: #d55e00; font-weight: bold;">vector</span>::<span style="color: #d55e00;">vector</span>(<span style="color: #0072b2; font-weight: bold;">unsigned</span> <span style="color: #e69f00; font-weight: bold;">len</span>) {
<span class="linenr"> 7: </span> <span style="color: #56b4e9; font-weight: bold;">this</span>->length = len;
<span class="linenr"> 8: </span> <span style="color: #56b4e9; font-weight: bold;">this</span>->data = <span style="color: #56b4e9; font-weight: bold;">new</span> <span style="color: #0072b2; font-weight: bold;">double</span>[len];
<span class="linenr"> 9: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Initialize data to zeros.</span>
<span class="linenr">10: </span> <span style="color: #56b4e9; font-weight: bold;">for</span>(<span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">i</span>=0; i < <span style="color: #56b4e9; font-weight: bold;">this</span>->len(); i++) { <span style="color: #56b4e9; font-weight: bold;">this</span>->data[i] = 0.0; }
<span class="linenr">11: </span> }
<span class="linenr">12: </span>
<span class="linenr">13: </span> <span style="color: #d55e00; font-weight: bold;">vector</span>::~<span style="color: #d55e00;">vector</span>() {
<span class="linenr">14: </span> <span style="color: #56b4e9; font-weight: bold;">delete</span> [] <span style="color: #56b4e9; font-weight: bold;">this</span>->data;
<span class="linenr">15: </span> }
<span class="linenr">16: </span>
<span class="linenr">17: </span> <span style="color: #0072b2; font-weight: bold;">unsigned</span> <span style="color: #d55e00; font-weight: bold;">vector</span>::<span style="color: #d55e00;">len</span>() <span style="color: #56b4e9; font-weight: bold;">const</span> {
<span class="linenr">18: </span> <span style="color: #56b4e9; font-weight: bold;">return</span> <span style="color: #56b4e9; font-weight: bold;">this</span>->length;
<span class="linenr">19: </span> }
<span class="linenr">20: </span>
<span class="linenr">21: </span> <span style="color: #0072b2; font-weight: bold;">double</span> & <span style="color: #d55e00; font-weight: bold;">vector</span>::<span style="color: #d55e00;">element</span>(<span style="color: #0072b2; font-weight: bold;">unsigned</span> <span style="color: #e69f00; font-weight: bold;">i</span>) <span style="color: #56b4e9; font-weight: bold;">const</span> {
<span class="linenr">22: </span><span style="color: #0072b2;"> #if</span><span style="color: #0072b2;">n</span><span style="color: #0072b2;">def</span> NDEBUG
<span class="linenr">23: </span> check_index(i);
<span class="linenr">24: </span><span style="color: #0072b2;"> #endif</span>
<span class="linenr">25: </span> <span style="color: #56b4e9; font-weight: bold;">return</span> <span style="color: #56b4e9; font-weight: bold;">this</span>->data[i];
<span class="linenr">26: </span> }
<span class="linenr">27: </span>
<span class="linenr">28: </span> <span style="color: #0072b2; font-weight: bold;">void</span> <span style="color: #d55e00; font-weight: bold;">vector</span>::<span style="color: #d55e00;">print</span>() <span style="color: #56b4e9; font-weight: bold;">const</span> {
<span class="linenr">29: </span> <span style="color: #56b4e9; font-weight: bold;">for</span>(<span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">i</span>=0; i < <span style="color: #56b4e9; font-weight: bold;">this</span>->len(); i++) {
<span class="linenr">30: </span> <span style="color: #d55e00; font-weight: bold;">std</span>::cout << <span style="color: #56b4e9; font-weight: bold;">this</span>->data[i] << <span style="color: #848ea9;">'\n'</span>;
<span class="linenr">31: </span> }
<span class="linenr">32: </span> }
<span class="linenr">33: </span>
<span class="linenr">34: </span> <span style="color: #0072b2; font-weight: bold;">void</span> <span style="color: #d55e00; font-weight: bold;">vector</span>::<span style="color: #d55e00;">check_index</span>(<span style="color: #0072b2; font-weight: bold;">unsigned</span> <span style="color: #e69f00; font-weight: bold;">i</span>) <span style="color: #56b4e9; font-weight: bold;">const</span> {
<span class="linenr">35: </span> <span style="color: #56b4e9; font-weight: bold;">if</span> (i < 0 || i >= <span style="color: #56b4e9; font-weight: bold;">this</span>->length) {
<span class="linenr">36: </span> <span style="color: #d55e00; font-weight: bold;">std</span>::cerr << <span style="color: #848ea9;">"ERROR: index, "</span> << i << <span style="color: #848ea9;">", is out-of-bounds.\n"</span>
<span class="linenr">37: </span> << <span style="color: #848ea9;">"(valid indices are 0-"</span> << (<span style="color: #56b4e9; font-weight: bold;">this</span>->length-1) << <span style="color: #848ea9;">")\n"</span>;
<span class="linenr">38: </span> exit(1);
<span class="linenr">39: </span> }
<span class="linenr">40: </span> }
<span class="linenr">41: </span>}
</pre>
</div>
<p>
Note that we again wrap the implementations in the same namespace as wrapped
by the class declaration. Also observe how each member-function is prepended
by <code>vector::</code>.
</p>
<p>
The keyword, <code>this</code>, is a pointer to the calling object. Writing, <code>this-></code>,
is equivalent to <code>(*this).</code>, and in fact, can be used for any pointer. Thus,
<code>this->length</code> is equivalent to <code>(*this).length</code>.
</p>
<p>
Observe how the private member function, <code>check_index</code>, is used in the
public <code>element</code> accessor function. If this library is compiled with the
flag, <code>-DNEDUBG</code>, then the check function will not be called. You could
read this flag as "define no debug". Thus, when this flag is present, the
debug function <code>check_index</code> is called whenever the element accessor is
called. The <code>check_index</code> function simply checks if the provided index is
out-of-bounds for the vector. If it is, an informative message is printed,
and the program terminates prematurely by calling <code>exit(1)</code>. Such assertions
with informative messages are a good practice, and will likely save you lots
of headaches in the future.
</p>
</div>
</div>
<div id="outline-container-orgheadline51" class="outline-4">
<h4 id="orgheadline51"><span class="section-number-4">7.1.3</span> Example usage</h4>
<div class="outline-text-4" id="text-7-1-3">
<p>
Ok. Let's see some example usage, by creating a <code>main.cpp</code>, containing:
</p>
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr"> 1: </span><span style="color: #0072b2;">#include</span> <span style="color: #848ea9;"><iostream></span>
<span class="linenr"> 2: </span><span style="color: #0072b2;">#include</span> <span style="color: #848ea9;">"vector.h"</span>
<span class="linenr"> 3: </span>
<span class="linenr"> 4: </span><span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #d55e00;">main</span>() {
<span class="linenr"> 5: </span> <span style="color: #d55e00; font-weight: bold;">vec</span>::<span style="color: #0072b2; font-weight: bold;">vector</span> <span style="color: #e69f00; font-weight: bold;">v</span>(5);
<span class="linenr"> 6: </span>
<span class="linenr"> 7: </span> <span style="color: #d55e00; font-weight: bold;">std</span>::cout << <span style="color: #848ea9;">"`v` has length = "</span> << v.len() << <span style="color: #848ea9;">"\n"</span>;
<span class="linenr"> 8: </span>
<span class="linenr"> 9: </span> v.element(0) = -1.27;
<span class="linenr">10: </span> v.element(3) = 3.1;
<span class="linenr">11: </span>
<span class="linenr">12: </span> v.print();
<span class="linenr">13: </span>
<span class="linenr">14: </span> v.element(5) = 1234.0;
<span class="linenr">15: </span>
<span class="linenr">16: </span> <span style="color: #56b4e9; font-weight: bold;">return</span> 0;
<span class="linenr">17: </span>}
</pre>
</div>
<p>
Let's first compile with our <code>check_index</code> debugger function in place:
</p>
<div class="org-src-container">
<pre class="src src-bash">g++ -c vector.cpp
g++ vector.o main.cpp
./a.out
</pre>
</div>
<p>
If successful, you should see:
</p>
<pre class="example">
`v` has length = 5
-1.27
0
0
3.1
0
ERROR: index, 5, is out-of-bounds.
(valid indices are 0-4)
</pre>
<p>
Now let's run without <code>check_index</code>:
</p>
<div class="org-src-container">
<pre class="src src-bash">g++ -DNDEBUG -c vector.cpp
g++ vector.o main.cpp
./a.out
</pre>
</div>
<p>
Upon running, you will likely see some extensive list of errors when
the element beyond the array's length is attempted to be accessed. Again,
by liberally sprinkling these sorts of assertions through your code, you
will (sometimes) find debugging much easier. After you're fairly certain
that your code is working, you can simply compile with <code>-DNEDUBG</code>.
</p>
</div>
</div>
<div id="outline-container-orgheadline54" class="outline-4">
<h4 id="orgheadline54"><span class="section-number-4">7.1.4</span> Operator overloading</h4>
<div class="outline-text-4" id="text-7-1-4">
<p>
The <code>v.element(i)</code> accessor is a bit clunky. We can replace this with the
more natural, <code>v[i]</code>, by <b>overloading</b> the <code>[]</code> operator. Indeed,
<a href="http://en.cppreference.com/w/cpp/language/operators">we can overload many</a> of the normal C++ operators, e.g. <code>+</code>, <code>-</code>, <code>=</code>, etc.
Some of these operators require more careful consideration when implementing
class that utilizes dynamic memory allocation, such as our <code>vector</code> class.
</p>
</div>
<div id="outline-container-orgheadline52" class="outline-5">
<h5 id="orgheadline52"><span class="section-number-5">7.1.4.1</span> Overloading <code>operator[]</code></h5>
<div class="outline-text-5" id="text-7-1-4-1">
<p>
In the header file, replace the <code>element</code> function declaration with:
</p>
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr">1: </span><span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">data accessor.</span>
<span class="linenr">2: </span><span style="color: #0072b2; font-weight: bold;">double</span> & <span style="color: #56b4e9; font-weight: bold;">operator</span><span style="color: #d55e00;">[]</span>(<span style="color: #0072b2; font-weight: bold;">unsigned</span> <span style="color: #e69f00; font-weight: bold;">i</span>) <span style="color: #56b4e9; font-weight: bold;">const</span>;
</pre>
</div>
<p>
and in the implementation file, replace the <code>element</code> implementation with:
</p>
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr">1: </span><span style="color: #0072b2; font-weight: bold;">double</span> & <span style="color: #d55e00; font-weight: bold;">vector</span>::<span style="color: #56b4e9; font-weight: bold;">operator</span><span style="color: #d55e00;">[]</span>(<span style="color: #0072b2; font-weight: bold;">unsigned</span> <span style="color: #e69f00; font-weight: bold;">i</span>) <span style="color: #56b4e9; font-weight: bold;">const</span> {
<span class="linenr">2: </span><span style="color: #0072b2;"> #if</span><span style="color: #0072b2;">n</span><span style="color: #0072b2;">def</span> NDEBUG
<span class="linenr">3: </span> check_index(i);
<span class="linenr">4: </span><span style="color: #0072b2;"> #endif</span>
<span class="linenr">5: </span> <span style="color: #56b4e9; font-weight: bold;">return</span> <span style="color: #56b4e9; font-weight: bold;">this</span>->data[i];
<span class="linenr">6: </span>}
</pre>
</div>
<p>
Just think of <code>operator[]</code> as the new name of the function, <code>element</code>. We
can now use the overloaded operator just like how we would use the <code>[]</code> for
operator for accessing array components, expect now our overloaded operator
function is called instead.
</p>
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr">1: </span>v[0] = -1.27;
<span class="linenr">2: </span>v[3] = 3.1;
</pre>
</div>
<p>
Compile and run:
</p>
<div class="org-src-container">
<pre class="src src-bash">g++ -c vector.cpp && g++ vector.o main.cpp && ./a.out
</pre>
</div>
<p>
and you should see:
</p>
<pre class="example">
-1.27
0
0
3.1
0
</pre>
<p>
just like before.
</p>
</div>
</div>
<div id="outline-container-orgheadline53" class="outline-5">
<h5 id="orgheadline53"><span class="section-number-5">7.1.4.2</span> Overloading <code>operator=</code></h5>
<div class="outline-text-5" id="text-7-1-4-2">
<p>
Let's overload the <code>=</code> operator so we can assign one vector to another.
We'll write in a way such that the vector on the left hand side is
overwritten by the one on the right.
</p>
<p>
Let's add a declaration to the header file,
</p>
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr">1: </span><span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">assignment operator.</span>
<span class="linenr">2: </span><span style="color: #0072b2; font-weight: bold;">vector</span> & <span style="color: #56b4e9; font-weight: bold;">operator</span><span style="color: #d55e00;">=</span>(<span style="color: #56b4e9; font-weight: bold;">const</span> <span style="color: #0072b2; font-weight: bold;">vector</span> & <span style="color: #e69f00; font-weight: bold;">src</span>);
</pre>
</div>
<p>
and let's add to the implementation file,
</p>
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr"> 1: </span><span style="color: #0072b2; font-weight: bold;">vector</span> & <span style="color: #d55e00; font-weight: bold;">vector</span>::<span style="color: #56b4e9; font-weight: bold;">operator</span><span style="color: #d55e00;">=</span>(<span style="color: #56b4e9; font-weight: bold;">const</span> <span style="color: #0072b2; font-weight: bold;">vector</span> & <span style="color: #e69f00; font-weight: bold;">src</span>) {
<span class="linenr"> 2: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Delete the old data.</span>
<span class="linenr"> 3: </span> <span style="color: #56b4e9; font-weight: bold;">delete</span> [] <span style="color: #56b4e9; font-weight: bold;">this</span>->data;
<span class="linenr"> 4: </span>
<span class="linenr"> 5: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Initialize the new data.</span>
<span class="linenr"> 6: </span> <span style="color: #56b4e9; font-weight: bold;">this</span>->length = src.len();
<span class="linenr"> 7: </span> <span style="color: #56b4e9; font-weight: bold;">this</span>->data = <span style="color: #56b4e9; font-weight: bold;">new</span> <span style="color: #0072b2; font-weight: bold;">double</span>[<span style="color: #56b4e9; font-weight: bold;">this</span>->len()];
<span class="linenr"> 8: </span>
<span class="linenr"> 9: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Copy over the new data.</span>
<span class="linenr">10: </span> <span style="color: #56b4e9; font-weight: bold;">for</span>(<span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">i</span>=0; i < <span style="color: #56b4e9; font-weight: bold;">this</span>->len(); i++) {
<span class="linenr">11: </span> <span style="color: #56b4e9; font-weight: bold;">this</span>->data[i] = src[i];
<span class="linenr">12: </span> }
<span class="linenr">13: </span>
<span class="linenr">14: </span> <span style="color: #56b4e9; font-weight: bold;">return</span> *<span style="color: #56b4e9; font-weight: bold;">this</span>;
<span class="linenr">15: </span>}
</pre>
</div>
<p>
Now, let's assume the <code>vector</code> instance, <code>v</code>, from above is still defined, and
we'll create a new vector:
</p>
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr">1: </span><span style="color: #d55e00; font-weight: bold;">vec</span>::<span style="color: #0072b2; font-weight: bold;">vector</span> <span style="color: #d55e00;">w</span>(14);
<span class="linenr">2: </span>w = v;
<span class="linenr">3: </span>w.print();
</pre>
</div>
<p>
This should print,
</p>
<pre class="example">
-1.27
0
0
3.1
0
</pre>
<p>
Notice that <code>w</code> is intially defined to be of length <code>14</code>, but this is
overwritten, and its new length is the length of <code>v</code>. Also note that all of
<code>w</code>'s old data is deleted.
</p>
</div>
</div>
</div>
<div id="outline-container-orgheadline55" class="outline-4">
<h4 id="orgheadline55"><span class="section-number-4">7.1.5</span> The copy constructor</h4>
<div class="outline-text-4" id="text-7-1-5">
<p>
It may be tempting at this point to attempt to initialize <code>w</code> from <code>v</code>
directly:
</p>
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr">1: </span><span style="color: #d55e00; font-weight: bold;">vec</span>::<span style="color: #0072b2; font-weight: bold;">vector</span> <span style="color: #e69f00; font-weight: bold;">w</span> = v;
</pre>
</div>
<p>
If you attempt this currently, you will see all sorts of errors. This is
because this type of intialization does not call the assignment operator.
It calls the <b>copy constructor</b>. The assignment operator is only called
when the object has already been intialized. Writing the previous line of
code is essentially equivalent to
</p>
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr">1: </span><span style="color: #d55e00; font-weight: bold;">vec</span>::<span style="color: #0072b2; font-weight: bold;">vector</span> <span style="color: #d55e00;">w</span>(v);
</pre>
</div>
<p>
In other words, the constructor is called with the existing vector, <code>v</code>, as
the argument, but we have not written a constructor yet with such a call
signature.
</p>
<p>
The constructor can be overloaded, i.e., we can write multiple versions of the
constructor function, and the one that matches the correct call signature will
be used. This function overloading behavior actually applies to all functions
in C++.
</p>
<p>
Let's add the copy constructor declaration to the header file:
</p>
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr">1: </span><span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Copy constructor.</span>
<span class="linenr">2: </span><span style="color: #d55e00;">vector</span>(<span style="color: #56b4e9; font-weight: bold;">const</span> <span style="color: #0072b2; font-weight: bold;">vector</span> & <span style="color: #e69f00; font-weight: bold;">src</span>);
</pre>
</div>
<p>
and let's add its implementation:
</p>
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr">1: </span><span style="color: #d55e00; font-weight: bold;">vector</span>::<span style="color: #d55e00;">vector</span>(<span style="color: #56b4e9; font-weight: bold;">const</span> <span style="color: #0072b2; font-weight: bold;">vector</span> & <span style="color: #e69f00; font-weight: bold;">src</span>) {
<span class="linenr">2: </span> <span style="color: #56b4e9; font-weight: bold;">this</span>->length = src.len();
<span class="linenr">3: </span> <span style="color: #56b4e9; font-weight: bold;">this</span>->data = <span style="color: #56b4e9; font-weight: bold;">new</span> <span style="color: #0072b2; font-weight: bold;">double</span>[<span style="color: #56b4e9; font-weight: bold;">this</span>->len()];
<span class="linenr">4: </span>
<span class="linenr">5: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Copy over the data.</span>
<span class="linenr">6: </span> <span style="color: #56b4e9; font-weight: bold;">for</span>(<span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">i</span>=0; i < <span style="color: #56b4e9; font-weight: bold;">this</span>->len(); i++) {
<span class="linenr">7: </span> <span style="color: #56b4e9; font-weight: bold;">this</span>->data[i] = src[i];
<span class="linenr">8: </span> }
<span class="linenr">9: </span>}
</pre>
</div>
<p>
Now we compile and run something like:
</p>
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr">1: </span><span style="color: #d55e00; font-weight: bold;">vec</span>::<span style="color: #0072b2; font-weight: bold;">vector</span> <span style="color: #e69f00; font-weight: bold;">w</span> = v;
<span class="linenr">2: </span>w.print();
</pre>
</div>
<p>
we will see:
</p>
<pre class="example">
-1.27
0
0
3.1
0
</pre>
</div>
</div>
<div id="outline-container-orgheadline58" class="outline-4">
<h4 id="orgheadline58"><span class="section-number-4">7.1.6</span> Friend functions</h4>
<div class="outline-text-4" id="text-7-1-6">
<p>
Non-member functions may access private member variables and private member
functions of by labeling them as <code>friend</code>. This is useful in situations where
it is not clear that function should be "called on" an object, i.e.
<code>object.method(params)</code>. Friend functions should be declared in the class
declaration, but their implementation is not prepended with <code>class::</code>, which
is necessary for member functions.
</p>
</div>
<div id="outline-container-orgheadline56" class="outline-5">
<h5 id="orgheadline56"><span class="section-number-5">7.1.6.1</span> Overloading <code>operator*</code></h5>
<div class="outline-text-5" id="text-7-1-6-1">
<p>
As an example, we'll overload <code>operator*</code> to implement scalar multiplication.
</p>
<p>
Because scalar multiplication should commute, let's add the following
two declarations to our header file:
</p>
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr">1: </span><span style="color: #56b4e9; font-weight: bold;">friend</span> <span style="color: #0072b2; font-weight: bold;">vector</span> <span style="color: #56b4e9; font-weight: bold;">operator</span><span style="color: #d55e00;">*</span>(<span style="color: #56b4e9; font-weight: bold;">const</span> <span style="color: #0072b2; font-weight: bold;">vector</span> & <span style="color: #e69f00; font-weight: bold;">v</span>, <span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #e69f00; font-weight: bold;">s</span>);
<span class="linenr">2: </span><span style="color: #56b4e9; font-weight: bold;">friend</span> <span style="color: #0072b2; font-weight: bold;">vector</span> <span style="color: #56b4e9; font-weight: bold;">operator</span><span style="color: #d55e00;">*</span>(<span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #e69f00; font-weight: bold;">s</span>, <span style="color: #56b4e9; font-weight: bold;">const</span> <span style="color: #0072b2; font-weight: bold;">vector</span> & <span style="color: #e69f00; font-weight: bold;">v</span>);
</pre>
</div>
<p>
Let's implement the first by adding to the declaration file:
</p>
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr">1: </span><span style="color: #0072b2; font-weight: bold;">vector</span> <span style="color: #56b4e9; font-weight: bold;">operator</span><span style="color: #d55e00;">*</span>(<span style="color: #56b4e9; font-weight: bold;">const</span> <span style="color: #0072b2; font-weight: bold;">vector</span> & <span style="color: #e69f00; font-weight: bold;">v</span>, <span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #e69f00; font-weight: bold;">s</span>) {
<span class="linenr">2: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Copy v to start.</span>
<span class="linenr">3: </span> <span style="color: #0072b2; font-weight: bold;">vector</span> <span style="color: #e69f00; font-weight: bold;">result</span> = v;
<span class="linenr">4: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Then multiply all entries by scalar, s.</span>
<span class="linenr">5: </span> <span style="color: #56b4e9; font-weight: bold;">for</span>(<span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">i</span>=0; i < v.len(); i++) {
<span class="linenr">6: </span> result[i] *= s;
<span class="linenr">7: </span> }
<span class="linenr">8: </span> <span style="color: #56b4e9; font-weight: bold;">return</span> result;
<span class="linenr">9: </span>}
</pre>
</div>
<p>
Observe how we didn't prepend <code>vector::</code> before <code>operator*</code> because
these are <code>friend</code> functions.
</p>
<p>
Now, we can use the first implementation to achieve the second:
</p>
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr">1: </span><span style="color: #0072b2; font-weight: bold;">vector</span> <span style="color: #56b4e9; font-weight: bold;">operator</span><span style="color: #d55e00;">*</span>(<span style="color: #0072b2; font-weight: bold;">double</span> <span style="color: #e69f00; font-weight: bold;">s</span>, <span style="color: #56b4e9; font-weight: bold;">const</span> <span style="color: #0072b2; font-weight: bold;">vector</span> & <span style="color: #e69f00; font-weight: bold;">v</span>) {
<span class="linenr">2: </span> <span style="color: #56b4e9; font-weight: bold;">return</span> v*s;
<span class="linenr">3: </span>}
</pre>
</div>
<p>
And let's try it out:
</p>
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr"> 1: </span><span style="color: #0072b2;">#include</span> <span style="color: #848ea9;"><iostream></span>
<span class="linenr"> 2: </span><span style="color: #0072b2;">#include</span> <span style="color: #848ea9;">"vector.h"</span>
<span class="linenr"> 3: </span>
<span class="linenr"> 4: </span><span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #d55e00;">main</span>() {
<span class="linenr"> 5: </span> <span style="color: #d55e00; font-weight: bold;">vec</span>::<span style="color: #0072b2; font-weight: bold;">vector</span> <span style="color: #e69f00; font-weight: bold;">v</span>(5);
<span class="linenr"> 6: </span>
<span class="linenr"> 7: </span> v[0] = -1.27;
<span class="linenr"> 8: </span> v[3] = 3.1;
<span class="linenr"> 9: </span>
<span class="linenr">10: </span> <span style="color: #d55e00; font-weight: bold;">vec</span>::<span style="color: #0072b2; font-weight: bold;">vector</span> <span style="color: #e69f00; font-weight: bold;">w</span> = 2*v;
<span class="linenr">11: </span>
<span class="linenr">12: </span> w.print();
<span class="linenr">13: </span>
<span class="linenr">14: </span> <span style="color: #d55e00; font-weight: bold;">std</span>::cout << <span style="color: #848ea9;">"\n"</span>;
<span class="linenr">15: </span>
<span class="linenr">16: </span> w = w*0.5;
<span class="linenr">17: </span>
<span class="linenr">18: </span> w.print();
<span class="linenr">19: </span>
<span class="linenr">20: </span> <span style="color: #56b4e9; font-weight: bold;">return</span> 0;
<span class="linenr">21: </span>}
</pre>
</div>
<p>
prints:
</p>
<pre class="example">
-2.54
0
0
6.2
0
-1.27
0
0
3.1
0
</pre>
<p>
Other binary arithmetic operators could also be implements as <code>friend</code>
functions, e.g., vector addition and subtraction and component-wise
multiplication and division. Component-wise multiplication would overload
<code>operator*</code> for a third time but would accept two vectors as function
arguments.
</p>
</div>
</div>
<div id="outline-container-orgheadline57" class="outline-5">
<h5 id="orgheadline57"><span class="section-number-5">7.1.6.2</span> Overloading <code>operator<<</code></h5>
<div class="outline-text-5" id="text-7-1-6-2">
<p>
We may overload <code>operator<<</code> to send things to the output stream. This
is a more C++ way to print than our current <code>print</code> function. We overload
this operator by adding the following <code>friend</code> declaration to the header:
</p>
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr">1: </span><span style="color: #56b4e9; font-weight: bold;">friend</span> <span style="color: #d55e00; font-weight: bold;">std</span>::<span style="color: #0072b2; font-weight: bold;">ostream</span> & <span style="color: #56b4e9; font-weight: bold;">operator</span><span style="color: #d55e00;"><<</span>(<span style="color: #d55e00; font-weight: bold;">std</span>::<span style="color: #0072b2; font-weight: bold;">ostream</span> & <span style="color: #e69f00; font-weight: bold;">outs</span>, <span style="color: #56b4e9; font-weight: bold;">const</span> <span style="color: #0072b2; font-weight: bold;">vector</span> & <span style="color: #e69f00; font-weight: bold;">v</span>);
</pre>
</div>
<p>
Note that the <code>ostream</code> object belongs to the <code>std</code> namespace. Next, we
add to the implementation file:
</p>
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr">1: </span><span style="color: #d55e00; font-weight: bold;">std</span>::<span style="color: #0072b2; font-weight: bold;">ostream</span> & <span style="color: #56b4e9; font-weight: bold;">operator</span><span style="color: #d55e00;"><<</span>(<span style="color: #d55e00; font-weight: bold;">std</span>::<span style="color: #0072b2; font-weight: bold;">ostream</span> & <span style="color: #e69f00; font-weight: bold;">outs</span>, <span style="color: #56b4e9; font-weight: bold;">const</span> <span style="color: #0072b2; font-weight: bold;">vector</span> & <span style="color: #e69f00; font-weight: bold;">v</span>) {
<span class="linenr">2: </span> <span style="color: #56b4e9; font-weight: bold;">for</span>(<span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">i</span>=0; i < v.len(); i++) {
<span class="linenr">3: </span> outs << v.data[i] << <span style="color: #848ea9;">"\n"</span>;
<span class="linenr">4: </span> }
</pre>
</div>
<p>
So now we stream the vector to <code>std::cout</code> just as we do for printing
numbers and strings to the screen:
</p>
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr"> 1: </span><span style="color: #0072b2;">#include</span> <span style="color: #848ea9;"><iostream></span>
<span class="linenr"> 2: </span><span style="color: #0072b2;">#include</span> <span style="color: #848ea9;">"vector.h"</span>
<span class="linenr"> 3: </span>
<span class="linenr"> 4: </span><span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #d55e00;">main</span>() {
<span class="linenr"> 5: </span> <span style="color: #d55e00; font-weight: bold;">vec</span>::<span style="color: #0072b2; font-weight: bold;">vector</span> <span style="color: #e69f00; font-weight: bold;">v</span>(5);
<span class="linenr"> 6: </span>
<span class="linenr"> 7: </span> v[0] = -1.27;
<span class="linenr"> 8: </span> v[3] = 3.1;
<span class="linenr"> 9: </span>
<span class="linenr">10: </span> <span style="color: #d55e00; font-weight: bold;">std</span>::cout << v;
<span class="linenr">11: </span>
<span class="linenr">12: </span> <span style="color: #56b4e9; font-weight: bold;">return</span> 0;
<span class="linenr">13: </span>}
</pre>
</div>
<p>
which prints
</p>
<pre class="example">
-1.27
0
0
3.1
0
</pre>
</div>
</div>
</div>
</div>
<div id="outline-container-orgheadline61" class="outline-3">
<h3 id="orgheadline61"><span class="section-number-3">7.2</span> Templating: a matrix class</h3>
<div class="outline-text-3" id="text-7-2">
<p>
Template classes allow you to generalize your classes. We will introduce the
concept of templating by creating a templated matrix class. Below is a bare-bones
template class for a matrix type. Note that the implementations for templates
must all go in the header file; they cannot be split into seperate
implementation files.
</p>
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr"> 1: </span><span style="color: #0072b2;">#if</span><span style="color: #0072b2;">n</span><span style="color: #0072b2;">def</span> MATRIX_H
<span class="linenr"> 2: </span><span style="color: #0072b2;">#define</span> <span style="color: #e69f00; font-weight: bold;">MATRIX_H</span>
<span class="linenr"> 3: </span>
<span class="linenr"> 4: </span><span style="color: #0072b2;">#include</span> <span style="color: #848ea9;"><iostream></span>
<span class="linenr"> 5: </span><span style="color: #0072b2;">#include</span> <span style="color: #848ea9;"><cstdlib></span>
<span class="linenr"> 6: </span>
<span class="linenr"> 7: </span><span style="color: #56b4e9; font-weight: bold;">namespace</span> <span style="color: #d55e00; font-weight: bold;">mtx</span> {
<span class="linenr"> 8: </span> <span style="color: #56b4e9; font-weight: bold;">template</span><<span style="color: #56b4e9; font-weight: bold;">class</span> <span style="color: #0072b2; font-weight: bold;">T</span>>
<span class="linenr"> 9: </span> <span style="color: #56b4e9; font-weight: bold;">class</span> <span style="color: #0072b2; font-weight: bold;">matrix</span> {
<span class="linenr">10: </span> <span style="color: #56b4e9; font-weight: bold;">public</span>:
<span class="linenr">11: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Default empty constructor.</span>
<span class="linenr">12: </span> <span style="color: #d55e00;">matrix</span>() {}
<span class="linenr">13: </span>
<span class="linenr">14: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Constructor from matrix dimensions.</span>
<span class="linenr">15: </span> <span style="color: #d55e00;">matrix</span>(<span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">m</span>, <span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">n</span>) {
<span class="linenr">16: </span> <span style="color: #56b4e9; font-weight: bold;">this</span>->m = m; <span style="color: #56b4e9; font-weight: bold;">this</span>->n = n;
<span class="linenr">17: </span> <span style="color: #56b4e9; font-weight: bold;">this</span>->data = <span style="color: #56b4e9; font-weight: bold;">new</span> <span style="color: #0072b2; font-weight: bold;">T</span>[m*n];
<span class="linenr">18: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Initialize to zeros.</span>
<span class="linenr">19: </span> <span style="color: #56b4e9; font-weight: bold;">for</span>(<span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">i</span>=0; i < m*n; i++) <span style="color: #56b4e9; font-weight: bold;">this</span>->data[i] = T(0);
<span class="linenr">20: </span> }
<span class="linenr">21: </span>
<span class="linenr">22: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Copy constructor.</span>
<span class="linenr">23: </span> <span style="color: #d55e00;">matrix</span>(<span style="color: #56b4e9; font-weight: bold;">const</span> <span style="color: #0072b2; font-weight: bold;">matrix</span> & <span style="color: #e69f00; font-weight: bold;">src</span>) : matrix(src.m, src.n) {
<span class="linenr">24: </span> <span style="color: #56b4e9; font-weight: bold;">for</span>(<span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">i</span>=0; i < m; i++) {
<span class="linenr">25: </span> <span style="color: #56b4e9; font-weight: bold;">for</span>(<span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">j</span>=0; j < n; j++) {
<span class="linenr">26: </span> (*<span style="color: #56b4e9; font-weight: bold;">this</span>)(i,j) = src(i,j);
<span class="linenr">27: </span> }
<span class="linenr">28: </span> }
<span class="linenr">29: </span> }
<span class="linenr">30: </span>
<span class="linenr">31: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Destructor.</span>
<span class="linenr">32: </span> ~<span style="color: #d55e00;">matrix</span>() { <span style="color: #56b4e9; font-weight: bold;">delete</span> [] <span style="color: #56b4e9; font-weight: bold;">this</span>->data; }
<span class="linenr">33: </span>
<span class="linenr">34: </span> <span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #d55e00;">n_rows</span>() <span style="color: #56b4e9; font-weight: bold;">const</span> { <span style="color: #56b4e9; font-weight: bold;">return</span> <span style="color: #56b4e9; font-weight: bold;">this</span>->m; }
<span class="linenr">35: </span> <span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #d55e00;">n_cols</span>() <span style="color: #56b4e9; font-weight: bold;">const</span> { <span style="color: #56b4e9; font-weight: bold;">return</span> <span style="color: #56b4e9; font-weight: bold;">this</span>->n; }
<span class="linenr">36: </span>
<span class="linenr">37: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Element accessor.</span>
<span class="linenr">38: </span> <span style="color: #0072b2; font-weight: bold;">T</span> & <span style="color: #56b4e9; font-weight: bold;">operator</span><span style="color: #d55e00;">()</span>(<span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">i</span>, <span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">j</span>) <span style="color: #56b4e9; font-weight: bold;">const</span> {
<span class="linenr">39: </span><span style="color: #0072b2;"> #if</span><span style="color: #0072b2;">n</span><span style="color: #0072b2;">def</span> NDEBUG
<span class="linenr">40: </span> check_indices(i,j);
<span class="linenr">41: </span><span style="color: #0072b2;"> #endif</span>
<span class="linenr">42: </span>
<span class="linenr">43: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Data stored in row-major.</span>
<span class="linenr">44: </span> <span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">k</span> = i*<span style="color: #56b4e9; font-weight: bold;">this</span>->n + j;
<span class="linenr">45: </span>
<span class="linenr">46: </span> <span style="color: #56b4e9; font-weight: bold;">return</span> <span style="color: #56b4e9; font-weight: bold;">this</span>->data[k];
<span class="linenr">47: </span> }
<span class="linenr">48: </span> <span style="color: #56b4e9; font-weight: bold;">protected</span>:
<span class="linenr">49: </span> <span style="color: #0072b2; font-weight: bold;">T</span> * <span style="color: #e69f00; font-weight: bold;">data</span>;
<span class="linenr">50: </span> <span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">m</span>; <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">n_rows</span>
<span class="linenr">51: </span> <span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">n</span>; <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">n_cols</span>
<span class="linenr">52: </span> <span style="color: #0072b2; font-weight: bold;">void</span> <span style="color: #d55e00;">check_indices</span>(<span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">i</span>, <span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">j</span>) <span style="color: #56b4e9; font-weight: bold;">const</span> {
<span class="linenr">53: </span> <span style="color: #56b4e9; font-weight: bold;">if</span> (i < 0 || i >= <span style="color: #56b4e9; font-weight: bold;">this</span>->m) {
<span class="linenr">54: </span> <span style="color: #d55e00; font-weight: bold;">std</span>::cerr << <span style="color: #848ea9;">"Invalid row index, "</span> << i << <span style="color: #848ea9;">".\n"</span>
<span class="linenr">55: </span> << <span style="color: #848ea9;">"Valid row indices are: 0-"</span> << (<span style="color: #56b4e9; font-weight: bold;">this</span>->m-1)
<span class="linenr">56: </span> << <span style="color: #848ea9;">".\n"</span>;
<span class="linenr">57: </span> exit(1);
<span class="linenr">58: </span> }
<span class="linenr">59: </span> <span style="color: #56b4e9; font-weight: bold;">if</span> (j < 0 || j >= <span style="color: #56b4e9; font-weight: bold;">this</span>->n) {
<span class="linenr">60: </span> <span style="color: #d55e00; font-weight: bold;">std</span>::cerr << <span style="color: #848ea9;">"Invalid column index, "</span> << j << <span style="color: #848ea9;">".\n"</span>
<span class="linenr">61: </span> << <span style="color: #848ea9;">"Valid column indices are: 0-"</span> << (<span style="color: #56b4e9; font-weight: bold;">this</span>->n-1)
<span class="linenr">62: </span> << <span style="color: #848ea9;">".\n"</span>;
<span class="linenr">63: </span> exit(1);
<span class="linenr">64: </span> }
<span class="linenr">65: </span> }
<span class="linenr">66: </span> };
<span class="linenr">67: </span>}
<span class="linenr">68: </span>
<span class="linenr">69: </span><span style="color: #0072b2;">#endif</span>
</pre>
</div>
<p>
Observe how the class declaration begins with the statement, <code>template<class T></code>.
When new instances of the class, <code>matrix</code>, are instantiated, the type, <code>T</code>, must
be specified. Notice that the private variable now holds a pointer to type, <code>T</code>,
and the element accessor returns a reference to to type, <code>T</code>. Compare this to
our previous <code>vector</code> class where each vector instance held type, <code>double</code>.
</p>
<p>
Also notice that the copy constructor uses something called an argument list.
We've used it to call the initial constructor before proceeding with the copying
of components from the <code>matrix</code> instance, <code>src</code>. To use this type of syntax (i.e.,
initializer lists) you may have to compiel with the flag <code>-std=c++11</code>, e.g.,
<code>g++ -std=c++11 main.cpp</code>.
</p>
<p>
Finally, observe that rather the keyword, <code>private</code>, we have used the keyword
<code>protected</code>. This will allow inherited classes to use the private member variables
and functions. We will talk about inheritance later, but for now, it essentially
the same as marking these things as <code>private</code>.
</p>
<p>
Here is some example usage:
</p>
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr"> 1: </span><span style="color: #0072b2;">#include</span> <span style="color: #848ea9;"><iostream></span>
<span class="linenr"> 2: </span><span style="color: #0072b2;">#include</span> <span style="color: #848ea9;">"matrix.h"</span>
<span class="linenr"> 3: </span>
<span class="linenr"> 4: </span><span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #d55e00;">main</span>() {
<span class="linenr"> 5: </span> <span style="color: #d55e00; font-weight: bold;">mtx</span>::<span style="color: #0072b2; font-weight: bold;">matrix</span><<span style="color: #0072b2; font-weight: bold;">float</span>> <span style="color: #e69f00; font-weight: bold;">A</span>(10, 10);
<span class="linenr"> 6: </span> <span style="color: #d55e00; font-weight: bold;">mtx</span>::<span style="color: #0072b2; font-weight: bold;">matrix</span><<span style="color: #0072b2; font-weight: bold;">double</span>> <span style="color: #e69f00; font-weight: bold;">B</span>(10, 12);
<span class="linenr"> 7: </span>
<span class="linenr"> 8: </span> A(0,0) = 3.14;
<span class="linenr"> 9: </span> B(2,4) = -2*A(0,0);
<span class="linenr">10: </span>
<span class="linenr">11: </span> <span style="color: #d55e00; font-weight: bold;">std</span>::cout << A(0,0) << <span style="color: #848ea9;">"\n"</span>;
<span class="linenr">12: </span> <span style="color: #d55e00; font-weight: bold;">std</span>::cout << B(2,4) << <span style="color: #848ea9;">"\n"</span>;
<span class="linenr">13: </span>
<span class="linenr">14: </span> <span style="color: #d55e00; font-weight: bold;">mtx</span>::<span style="color: #0072b2; font-weight: bold;">matrix</span><<span style="color: #0072b2; font-weight: bold;">double</span>> <span style="color: #e69f00; font-weight: bold;">C</span> = B;
<span class="linenr">15: </span> <span style="color: #d55e00; font-weight: bold;">mtx</span>::<span style="color: #0072b2; font-weight: bold;">matrix</span><<span style="color: #0072b2; font-weight: bold;">bool</span>> <span style="color: #e69f00; font-weight: bold;">D</span>(3,3);
<span class="linenr">16: </span>
<span class="linenr">17: </span> <span style="color: #d55e00; font-weight: bold;">std</span>::cout << C(2,4) << <span style="color: #848ea9;">"\n"</span>;
<span class="linenr">18: </span> <span style="color: #d55e00; font-weight: bold;">std</span>::cout << D(0,0) << <span style="color: #848ea9;">"\n"</span>;
<span class="linenr">19: </span>
<span class="linenr">20: </span> C(0,12);
<span class="linenr">21: </span>
<span class="linenr">22: </span> <span style="color: #56b4e9; font-weight: bold;">return</span> 0;
<span class="linenr">23: </span>}
</pre>
</div>
<p>
Outputs:
</p>
<pre class="example">
3.14
-6.28
-6.28
0
Invalid column index, 12.
Valid column indices are: 0-11.
</pre>
</div>
<div id="outline-container-orgheadline60" class="outline-4">
<h4 id="orgheadline60"><span class="section-number-4">7.2.1</span> Overloading <code>operator*</code></h4>
<div class="outline-text-4" id="text-7-2-1">
<p>
Let us overload <code>operator*</code> to implement matrix multiplication. It would also
make sense to overload <code>operator*</code> with other call signatures to implement scalar
multiplication or vector broadcasting, but for now we will just stick with
matrix multiplication.
</p>
<p>
Add the following two non-friend, non-member functions. They should be defined
just after the class definition, but they should remain in the body of namespace,
<code>mtx</code>.
</p>
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr"> 1: </span><span style="color: #56b4e9; font-weight: bold;">template</span><<span style="color: #56b4e9; font-weight: bold;">class</span> <span style="color: #0072b2; font-weight: bold;">T</span>>
<span class="linenr"> 2: </span><span style="color: #0072b2; font-weight: bold;">matrix</span><<span style="color: #0072b2; font-weight: bold;">T</span>> <span style="color: #56b4e9; font-weight: bold;">operator</span><span style="color: #d55e00;">*</span>(<span style="color: #56b4e9; font-weight: bold;">const</span> <span style="color: #0072b2; font-weight: bold;">matrix</span><<span style="color: #0072b2; font-weight: bold;">T</span>> & <span style="color: #e69f00; font-weight: bold;">L</span>, <span style="color: #56b4e9; font-weight: bold;">const</span> <span style="color: #0072b2; font-weight: bold;">matrix</span><<span style="color: #0072b2; font-weight: bold;">T</span>> & <span style="color: #e69f00; font-weight: bold;">R</span>) {
<span class="linenr"> 3: </span><span style="color: #0072b2;"> #if</span><span style="color: #0072b2;">n</span><span style="color: #0072b2;">def</span> NDEBUG
<span class="linenr"> 4: </span> check_sizes_for_matmul(L, R);
<span class="linenr"> 5: </span><span style="color: #0072b2;"> #endif</span>
<span class="linenr"> 6: </span>
<span class="linenr"> 7: </span> <span style="color: #0072b2; font-weight: bold;">matrix</span><<span style="color: #0072b2; font-weight: bold;">T</span>> <span style="color: #e69f00; font-weight: bold;">P</span>(L.n_rows(), R.n_cols());
<span class="linenr"> 8: </span> <span style="color: #56b4e9; font-weight: bold;">for</span>(<span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">i</span>=0; i < L.n_rows(); i++) {
<span class="linenr"> 9: </span> <span style="color: #56b4e9; font-weight: bold;">for</span>(<span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">j</span>=0; j < R.n_cols(); j++) {
<span class="linenr">10: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Inner product between row `i` of matrix, `L`</span>
<span class="linenr">11: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">and column `j` of matrix, `R`.</span>
<span class="linenr">12: </span> <span style="color: #56b4e9; font-weight: bold;">for</span>(<span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">k</span>=0; k < L.n_cols(); k++) {
<span class="linenr">13: </span> P(i,j) += L(i,k)*R(k,j);
<span class="linenr">14: </span> }
<span class="linenr">15: </span> }
<span class="linenr">16: </span> }
<span class="linenr">17: </span>
<span class="linenr">18: </span> <span style="color: #56b4e9; font-weight: bold;">return</span> P;
<span class="linenr">19: </span>}
<span class="linenr">20: </span>
<span class="linenr">21: </span><span style="color: #56b4e9; font-weight: bold;">template</span><<span style="color: #56b4e9; font-weight: bold;">class</span> <span style="color: #0072b2; font-weight: bold;">T</span>>
<span class="linenr">22: </span><span style="color: #0072b2; font-weight: bold;">void</span> <span style="color: #d55e00;">check_sizes_for_matmul</span>(<span style="color: #56b4e9; font-weight: bold;">const</span> <span style="color: #0072b2; font-weight: bold;">matrix</span><<span style="color: #0072b2; font-weight: bold;">T</span>> & <span style="color: #e69f00; font-weight: bold;">L</span>, <span style="color: #56b4e9; font-weight: bold;">const</span> <span style="color: #0072b2; font-weight: bold;">matrix</span><<span style="color: #0072b2; font-weight: bold;">T</span>> & <span style="color: #e69f00; font-weight: bold;">R</span>) {
<span class="linenr">23: </span> <span style="color: #56b4e9; font-weight: bold;">if</span> (L.n_cols() != R.n_rows()) {
<span class="linenr">24: </span> <span style="color: #d55e00; font-weight: bold;">std</span>::cerr << <span style="color: #848ea9;">"Size mismatch for matrix multiplication.\n"</span>
<span class="linenr">25: </span> << <span style="color: #848ea9;">"Left matrix has "</span> << L.n_cols() << <span style="color: #848ea9;">" cols, but\n"</span>
<span class="linenr">26: </span> << <span style="color: #848ea9;">"right matrix has "</span> << R.n_rows() << <span style="color: #848ea9;">" rows.\n"</span>;
<span class="linenr">27: </span> exit(1);
<span class="linenr">28: </span> }
<span class="linenr">29: </span>}
</pre>
</div>
<p>
Note that because we've chosen to implement these functions as non-member
and non-friend, we must prefix the function definitions with
<code>template<class T></code>. This isn't necessary when the functions were declared
inside of the class body (i.e., member or friend functions) since the class
declaration already begins with <code>template<class T></code>.
</p>
<p>
We've added a simple matrix multiplcation function above and a debugger
functions to ensure that if two matrices are multiplied, that it is well-
defined to do so. Otherwise, we print an error message and abort, as with
the element accessor function. We've also wrapped the debugger function
with compiler macros, so that it can be left out if we choose to do so.
</p>
<p>
Here is some example usage:
</p>
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr"> 1: </span><span style="color: #0072b2;">#include</span> <span style="color: #848ea9;"><iostream></span>
<span class="linenr"> 2: </span><span style="color: #0072b2;">#include</span> <span style="color: #848ea9;">"matrix.h"</span>
<span class="linenr"> 3: </span>
<span class="linenr"> 4: </span><span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #d55e00;">main</span>() {
<span class="linenr"> 5: </span> <span style="color: #d55e00; font-weight: bold;">mtx</span>::<span style="color: #0072b2; font-weight: bold;">matrix</span><<span style="color: #0072b2; font-weight: bold;">double</span>> <span style="color: #e69f00; font-weight: bold;">A</span>(2,2);
<span class="linenr"> 6: </span> <span style="color: #d55e00; font-weight: bold;">mtx</span>::<span style="color: #0072b2; font-weight: bold;">matrix</span><<span style="color: #0072b2; font-weight: bold;">double</span>> <span style="color: #e69f00; font-weight: bold;">B</span>(2,3);
<span class="linenr"> 7: </span>
<span class="linenr"> 8: </span> A(0,0) = A(0,1) = A(1,0) = 1; A(1,1) = -1;
<span class="linenr"> 9: </span>
<span class="linenr">10: </span> B(0,0) = B(1,0) = 1;
<span class="linenr">11: </span> B(0,1) = B(1,1) = -1;
<span class="linenr">12: </span> B(0,2) = B(1,2) = 3;
<span class="linenr">13: </span>
<span class="linenr">14: </span>
<span class="linenr">15: </span> <span style="color: #d55e00; font-weight: bold;">mtx</span>::<span style="color: #0072b2; font-weight: bold;">matrix</span><<span style="color: #0072b2; font-weight: bold;">double</span>> <span style="color: #e69f00; font-weight: bold;">C</span> = A*B;
<span class="linenr">16: </span>
<span class="linenr">17: </span> <span style="color: #56b4e9; font-weight: bold;">for</span>(<span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">i</span>=0; i < C.n_rows(); i++) {
<span class="linenr">18: </span> <span style="color: #56b4e9; font-weight: bold;">for</span>(<span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">j</span>=0; j < C.n_cols(); j++) {
<span class="linenr">19: </span> <span style="color: #d55e00; font-weight: bold;">std</span>::cout << C(i,j);
<span class="linenr">20: </span> <span style="color: #56b4e9; font-weight: bold;">if</span> (j < C.n_cols()-1)
<span class="linenr">21: </span> <span style="color: #d55e00; font-weight: bold;">std</span>::cout << <span style="color: #848ea9;">", "</span>;
<span class="linenr">22: </span> }
<span class="linenr">23: </span> <span style="color: #d55e00; font-weight: bold;">std</span>::cout << <span style="color: #848ea9;">"\n"</span>;
<span class="linenr">24: </span> }
<span class="linenr">25: </span>
<span class="linenr">26: </span> <span style="color: #0072b2; font-weight: bold;">B</span>*<span style="color: #e69f00; font-weight: bold;">A</span>;
<span class="linenr">27: </span>
<span class="linenr">28: </span> <span style="color: #56b4e9; font-weight: bold;">return</span> 0;
<span class="linenr">29: </span>}
</pre>
</div>
<p>
Output:
</p>
<pre class="example">
2, -2, 6
0, 0, 0
Size mismatch for matrix multiplication.
Left matrix has 3 cols, but
right matrix has 2 rows.
</pre>
</div>
</div>
</div>
<div id="outline-container-orgheadline62" class="outline-3">
<h3 id="orgheadline62"><span class="section-number-3">7.3</span> Inheritance</h3>
<div class="outline-text-3" id="text-7-3">
<p>
Often classes have a natural, hierarchical structure.
For example, a row or column vector could be seen as natural
"subclass" of the matrix class. So, supposing we had only
written the matrix class, it would be nice if we could write
a row or column vector class that inherited many of the properties
of the parent matrix class, but with additional properties,
unique to the vector classes. This is what inheritance may allow
us to do.
</p>
<p>
Let's create another couple of vector classes, called <code>rowvec</code> and
<code>colvec</code> in a file called <code>vectors.h</code>:
</p>
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr"> 1: </span><span style="color: #0072b2;">#if</span><span style="color: #0072b2;">n</span><span style="color: #0072b2;">def</span> VECTORS_H
<span class="linenr"> 2: </span><span style="color: #0072b2;">#define</span> <span style="color: #e69f00; font-weight: bold;">VECTORS_H</span>
<span class="linenr"> 3: </span>
<span class="linenr"> 4: </span><span style="color: #0072b2;">#include</span> <span style="color: #848ea9;">"matrix.h"</span>
<span class="linenr"> 5: </span>
<span class="linenr"> 6: </span><span style="color: #56b4e9; font-weight: bold;">namespace</span> <span style="color: #d55e00; font-weight: bold;">vec</span> {
<span class="linenr"> 7: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Row vector class inherits from the matrix class.</span>
<span class="linenr"> 8: </span> <span style="color: #56b4e9; font-weight: bold;">template</span><<span style="color: #56b4e9; font-weight: bold;">class</span> <span style="color: #0072b2; font-weight: bold;">T</span>>
<span class="linenr"> 9: </span> <span style="color: #56b4e9; font-weight: bold;">class</span> <span style="color: #0072b2; font-weight: bold;">rowvec</span> : <span style="color: #56b4e9; font-weight: bold;">public</span> <span style="color: #d55e00; font-weight: bold;">mtx</span>::<span style="color: #0072b2; font-weight: bold;">matrix</span><<span style="color: #0072b2; font-weight: bold;">T</span>> {
<span class="linenr">10: </span> <span style="color: #56b4e9; font-weight: bold;">public</span>:
<span class="linenr">11: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Constructor.</span>
<span class="linenr">12: </span> <span style="color: #d55e00;">rowvec</span>(<span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">n</span>) : <span style="color: #d55e00; font-weight: bold;">mtx</span>::<span style="color: #0072b2; font-weight: bold;">matrix</span><<span style="color: #0072b2; font-weight: bold;">T</span>>(1, n) {}
<span class="linenr">13: </span>
<span class="linenr">14: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">A length function.</span>
<span class="linenr">15: </span> <span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #d55e00;">len</span>() <span style="color: #56b4e9; font-weight: bold;">const</span> { <span style="color: #56b4e9; font-weight: bold;">return</span> <span style="color: #56b4e9; font-weight: bold;">this</span>->n_cols(); }
<span class="linenr">16: </span>
<span class="linenr">17: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Accessors.</span>
<span class="linenr">18: </span> <span style="color: #0072b2; font-weight: bold;">T</span> & <span style="color: #56b4e9; font-weight: bold;">operator</span><span style="color: #d55e00;">()</span>(<span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">i</span>) <span style="color: #56b4e9; font-weight: bold;">const</span> { <span style="color: #56b4e9; font-weight: bold;">return</span> <span style="color: #d55e00; font-weight: bold;">mtx</span>::<span style="color: #d55e00; font-weight: bold;">matrix</span><<span style="color: #0072b2; font-weight: bold;">T</span>>::<span style="color: #56b4e9; font-weight: bold;">operator</span>()(0, i); }
<span class="linenr">19: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Add [] as a possibility, too.</span>
<span class="linenr">20: </span> <span style="color: #0072b2; font-weight: bold;">T</span> & <span style="color: #56b4e9; font-weight: bold;">operator</span><span style="color: #d55e00;">[]</span>(<span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">i</span>) <span style="color: #56b4e9; font-weight: bold;">const</span> { <span style="color: #56b4e9; font-weight: bold;">return</span> (*<span style="color: #56b4e9; font-weight: bold;">this</span>)(i); }
<span class="linenr">21: </span> };
<span class="linenr">22: </span>
<span class="linenr">23: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Column vector class also inherits from the matrix class.</span>
<span class="linenr">24: </span> <span style="color: #56b4e9; font-weight: bold;">template</span><<span style="color: #56b4e9; font-weight: bold;">class</span> <span style="color: #0072b2; font-weight: bold;">T</span>>
<span class="linenr">25: </span> <span style="color: #56b4e9; font-weight: bold;">class</span> <span style="color: #0072b2; font-weight: bold;">colvec</span> : <span style="color: #56b4e9; font-weight: bold;">public</span> <span style="color: #d55e00; font-weight: bold;">mtx</span>::<span style="color: #0072b2; font-weight: bold;">matrix</span><<span style="color: #0072b2; font-weight: bold;">T</span>> {
<span class="linenr">26: </span> <span style="color: #56b4e9; font-weight: bold;">public</span>:
<span class="linenr">27: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Constructor.</span>
<span class="linenr">28: </span> <span style="color: #d55e00;">colvec</span>(<span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">n</span>) : <span style="color: #d55e00; font-weight: bold;">mtx</span>::<span style="color: #0072b2; font-weight: bold;">matrix</span><<span style="color: #0072b2; font-weight: bold;">T</span>>(n, 1) {}
<span class="linenr">29: </span>
<span class="linenr">30: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">A length function.</span>
<span class="linenr">31: </span> <span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #d55e00;">len</span>() <span style="color: #56b4e9; font-weight: bold;">const</span> { <span style="color: #56b4e9; font-weight: bold;">return</span> <span style="color: #56b4e9; font-weight: bold;">this</span>->n_rows(); }
<span class="linenr">32: </span>
<span class="linenr">33: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Accessor.</span>
<span class="linenr">34: </span> <span style="color: #0072b2; font-weight: bold;">T</span> & <span style="color: #56b4e9; font-weight: bold;">operator</span><span style="color: #d55e00;">()</span>(<span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">i</span>) <span style="color: #56b4e9; font-weight: bold;">const</span> { <span style="color: #56b4e9; font-weight: bold;">return</span> <span style="color: #d55e00; font-weight: bold;">mtx</span>::<span style="color: #d55e00; font-weight: bold;">matrix</span><<span style="color: #0072b2; font-weight: bold;">T</span>>::<span style="color: #56b4e9; font-weight: bold;">operator</span>()(i, 0); }
<span class="linenr">35: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Add [] as a possibility, too.</span>
<span class="linenr">36: </span> <span style="color: #0072b2; font-weight: bold;">T</span> & <span style="color: #56b4e9; font-weight: bold;">operator</span><span style="color: #d55e00;">[]</span>(<span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">i</span>) <span style="color: #56b4e9; font-weight: bold;">const</span> { <span style="color: #56b4e9; font-weight: bold;">return</span> (*<span style="color: #56b4e9; font-weight: bold;">this</span>)(i); }
<span class="linenr">37: </span> };
<span class="linenr">38: </span>}
<span class="linenr">39: </span>
<span class="linenr">40: </span><span style="color: #0072b2;">#endif</span>
</pre>
</div>
<p>
We have defined two classes in the <code>vec</code> namespace, <code>rowvec</code> and <code>colvec</code>.
Notice that the declaration of the <code>rowvec</code> class is slightly augmented,
</p>
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr">1: </span><span style="color: #56b4e9; font-weight: bold;">class</span> <span style="color: #0072b2; font-weight: bold;">rowvec</span> : <span style="color: #56b4e9; font-weight: bold;">public</span> <span style="color: #d55e00; font-weight: bold;">mtx</span>::<span style="color: #0072b2; font-weight: bold;">matrix</span><T> {
<span class="linenr">2: </span><span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">...</span>
<span class="linenr">3: </span>};
</pre>
</div>
<p>
This says that the <code>rowvec</code> class inherits all the methods and
variables defined for the <code>matrix</code> class. The keyword <code>protected</code>,
rather than <code>private</code> in the <code>matrix</code> class allows derived classes
access to these variables and functions. The keyword, <code>public</code>, in
the class declaration above signals that every <code>rowvec<T></code> instance
may be cast as a <code>matrix<T></code> instance. This means that functions (such
as our <code>operator*</code> in the <code>mtx</code> namespace) which expect <code>matrix</code> types
in argument, can also now accept <code>rowvec</code> types: <code>rowvec</code> instances
will be cast as <code>matrix</code> instances when possible.
</p>
<p>
Next, observe that the constructor simply calls the parent class
constructor, with an argument of <code>1</code> for the number of rows. We also
add a <code>len()</code> function which returns `n<sub>cols</sub>()`, calling the parent
class function. <code>len()</code> is more natural for a vector class.
</p>
<p>
Finally, <code>operator()</code> is overwritten so that only a single argument is
necessary. In the body, we simply call the parent class method. We also
add <code>operator[]</code>, which duplicates the functionality of <code>operator()</code>.
</p>
<p>
Notice that any other methods that exist for the parent class, <code>matrix</code>,
have been inherited. For example, it is not necessary to write a new
destructor function since this already exists for the parent and the
functionality is the same.
</p>
<p>
In the same, <code>vec</code>, namespace we have declared a <code>colvec</code> class which
is analogous to the <code>rowvec</code> class except with the dimensions swapped.
</p>
<p>
Let's check out an example:
</p>
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr"> 1: </span><span style="color: #0072b2;">#include</span> <span style="color: #848ea9;"><iostream></span>
<span class="linenr"> 2: </span><span style="color: #0072b2;">#include</span> <span style="color: #848ea9;">"matrix.h"</span>
<span class="linenr"> 3: </span><span style="color: #0072b2;">#include</span> <span style="color: #848ea9;">"vectors.h"</span>
<span class="linenr"> 4: </span>
<span class="linenr"> 5: </span><span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #d55e00;">main</span>() {
<span class="linenr"> 6: </span> <span style="color: #d55e00; font-weight: bold;">vec</span>::<span style="color: #0072b2; font-weight: bold;">rowvec</span><<span style="color: #0072b2; font-weight: bold;">float</span>> <span style="color: #e69f00; font-weight: bold;">x</span>(5);
<span class="linenr"> 7: </span> <span style="color: #d55e00; font-weight: bold;">vec</span>::<span style="color: #0072b2; font-weight: bold;">colvec</span><<span style="color: #0072b2; font-weight: bold;">float</span>> <span style="color: #e69f00; font-weight: bold;">y</span>(5);
<span class="linenr"> 8: </span>
<span class="linenr"> 9: </span> <span style="color: #56b4e9; font-weight: bold;">for</span>(<span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">i</span>=0; i < x.len(); i++) {
<span class="linenr">10: </span> <span style="color: #0072b2; font-weight: bold;">x</span>(<span style="color: #e69f00; font-weight: bold;">i</span>) = i;
<span class="linenr">11: </span> y[i] = -i; <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">elements may be accessed with either operator.</span>
<span class="linenr">12: </span> }
<span class="linenr">13: </span>
<span class="linenr">14: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">`rowvec` and `colvec` inherit from `matrix`,</span>
<span class="linenr">15: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">so `operator*` is well-defined if the shapes</span>
<span class="linenr">16: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">align, which always do for `rowvec` * `colvec`.</span>
<span class="linenr">17: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">However, the result is a `matrix` object.</span>
<span class="linenr">18: </span> <span style="color: #d55e00; font-weight: bold;">mtx</span>::<span style="color: #0072b2; font-weight: bold;">matrix</span><<span style="color: #0072b2; font-weight: bold;">float</span>> <span style="color: #e69f00; font-weight: bold;">a</span> = x*y;
<span class="linenr">19: </span>
<span class="linenr">20: </span> <span style="color: #d55e00; font-weight: bold;">std</span>::cout << a.n_rows() << <span style="color: #848ea9;">", "</span> << a.n_cols()
<span class="linenr">21: </span> << <span style="color: #848ea9;">" ... "</span> << a(0,0) << <span style="color: #848ea9;">"\n"</span>;
<span class="linenr">22: </span>
<span class="linenr">23: </span> <span style="color: #56b4e9; font-weight: bold;">return</span> 0;
<span class="linenr">24: </span>}
</pre>
</div>
<p>
This prints:
</p>
<pre class="example">
1, 1 ... -30
</pre>
<p>
So <code>operator*</code> works by casting both the <code>rowvec</code> and <code>colvec</code> arguments
as <code>matrix</code> types. The dimensions of these matrices match for matrix
multiplication and the resulting 1x1 matrix is returned. Let's overwrite
this behavior to return a scalar instead of a matrix when the operation,
<code>rowvec*colvec</code> is performed. Add the inside the <code>vec</code> namespace but outside
the class declarations:
</p>
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr"> 1: </span><span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Overwrite the matrix inherited `operator*`</span>
<span class="linenr"> 2: </span><span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">so that a scalar is returned when a `rowvec` and</span>
<span class="linenr"> 3: </span><span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">`colvec` are multiplied.</span>
<span class="linenr"> 4: </span><span style="color: #56b4e9; font-weight: bold;">template</span><<span style="color: #56b4e9; font-weight: bold;">class</span> <span style="color: #0072b2; font-weight: bold;">T</span>>
<span class="linenr"> 5: </span><span style="color: #0072b2; font-weight: bold;">T</span> <span style="color: #56b4e9; font-weight: bold;">operator</span><span style="color: #d55e00;">*</span>(<span style="color: #56b4e9; font-weight: bold;">const</span> <span style="color: #0072b2; font-weight: bold;">rowvec</span><<span style="color: #0072b2; font-weight: bold;">T</span>> & <span style="color: #e69f00; font-weight: bold;">x</span>, <span style="color: #56b4e9; font-weight: bold;">const</span> <span style="color: #0072b2; font-weight: bold;">colvec</span><<span style="color: #0072b2; font-weight: bold;">T</span>> & <span style="color: #e69f00; font-weight: bold;">y</span>) {
<span class="linenr"> 6: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Use the inherited matrix operator defined in the `mtx` namespace.</span>
<span class="linenr"> 7: </span> <span style="color: #d55e00; font-weight: bold;">mtx</span>::<span style="color: #0072b2; font-weight: bold;">matrix</span><<span style="color: #0072b2; font-weight: bold;">T</span>> <span style="color: #e69f00; font-weight: bold;">a</span> = <span style="color: #d55e00; font-weight: bold;">mtx</span>::<span style="color: #56b4e9; font-weight: bold;">operator</span>*(x,y);
<span class="linenr"> 8: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">Grab the only entry.</span>
<span class="linenr"> 9: </span> <span style="color: #0072b2; font-weight: bold;">T</span> <span style="color: #e69f00; font-weight: bold;">z</span> = a(0,0);
<span class="linenr">10: </span> <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">And return it.</span>
<span class="linenr">11: </span> <span style="color: #56b4e9; font-weight: bold;">return</span> z;
<span class="linenr">12: </span>}
</pre>
</div>
<p>
Observe that we can call the <code>operator*</code> function in the <code>mtx</code> namespace,
just like any other function. Now, the product between a <code>rowvec</code> and a <code>colvec</code>
is more natural:
</p>
<div class="org-src-container">
<pre class="src src-cpp"><span class="linenr"> 1: </span><span style="color: #0072b2;">#include</span> <span style="color: #848ea9;"><iostream></span>
<span class="linenr"> 2: </span><span style="color: #0072b2;">#include</span> <span style="color: #848ea9;">"matrix.h"</span>
<span class="linenr"> 3: </span><span style="color: #0072b2;">#include</span> <span style="color: #848ea9;">"vectors.h"</span>
<span class="linenr"> 4: </span>
<span class="linenr"> 5: </span><span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #d55e00;">main</span>() {
<span class="linenr"> 6: </span> <span style="color: #d55e00; font-weight: bold;">vec</span>::<span style="color: #0072b2; font-weight: bold;">rowvec</span><<span style="color: #0072b2; font-weight: bold;">float</span>> <span style="color: #e69f00; font-weight: bold;">x</span>(5);
<span class="linenr"> 7: </span> <span style="color: #d55e00; font-weight: bold;">vec</span>::<span style="color: #0072b2; font-weight: bold;">colvec</span><<span style="color: #0072b2; font-weight: bold;">float</span>> <span style="color: #e69f00; font-weight: bold;">y</span>(5);
<span class="linenr"> 8: </span>
<span class="linenr"> 9: </span> <span style="color: #56b4e9; font-weight: bold;">for</span>(<span style="color: #0072b2; font-weight: bold;">int</span> <span style="color: #e69f00; font-weight: bold;">i</span>=0; i < x.len(); i++) {
<span class="linenr">10: </span> <span style="color: #0072b2; font-weight: bold;">x</span>(<span style="color: #e69f00; font-weight: bold;">i</span>) = i;
<span class="linenr">11: </span> y[i] = -i; <span style="color: #009e73; font-style: italic;">// </span><span style="color: #009e73; font-style: italic;">elements may be accessed with either operator.</span>
<span class="linenr">12: </span> }
<span class="linenr">13: </span>
<span class="linenr">14: </span> <span style="color: #0072b2; font-weight: bold;">float</span> <span style="color: #e69f00; font-weight: bold;">a</span> = x*y;
<span class="linenr">15: </span> <span style="color: #d55e00; font-weight: bold;">std</span>::cout << a << <span style="color: #848ea9;">"\n"</span>;
<span class="linenr">16: </span>
<span class="linenr">17: </span> <span style="color: #56b4e9; font-weight: bold;">return</span> 0;
<span class="linenr">18: </span>}
</pre>
</div>
<p>
Output:
</p>
<pre class="example">
-30
</pre>
</div>
</div>
</div>
</div>
</div>
</body>
</html>
|
{
"redpajama_set_name": "RedPajamaGithub"
}
| 4,081 |
[
128000,
1340,
6591,
2373,
429,
16,
13,
15,
1,
11418,
429,
4867,
12,
23,
87090,
13853,
15822,
5385,
32516,
67996,
54,
18,
34,
322,
43119,
65952,
220,
16,
13,
15,
53981,
322,
965,
702,
78032,
1129,
2185,
1444,
18,
2726,
53427,
35232,
16,
76714,
35232,
16,
12,
6765,
61240,
891,
14063,
26015,
429,
1277,
1129,
2185,
1444,
18,
2726,
14,
2550,
24,
35232,
1,
8859,
429,
268,
1,
8562,
25,
5317,
429,
268,
891,
16747,
397,
6499,
220,
679,
21,
12,
605,
12,
605,
3206,
220,
868,
25,
2166,
3743,
18816,
220,
1795,
23793,
429,
2831,
11038,
1,
2262,
429,
1342,
14056,
46117,
23060,
12,
23,
1,
2662,
18816,
220,
836,
429,
18774,
1,
2262,
429,
3175,
20297,
9531,
11,
2926,
13230,
28,
16,
1,
2662,
28436,
29
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
1340,
6591,
2373,
429,
16,
13,
15,
1,
11418,
429,
4867,
12,
23,
87090,
13853,
15822,
5385,
32516,
67996,
54,
18,
34,
322,
43119,
65952,
220,
16,
13,
15,
53981,
322,
965,
702,
78032,
1129,
2185,
1444,
18,
2726,
53427,
35232,
16,
76714,
35232,
16,
12,
6765,
61240,
891,
14063,
26015,
429,
1277,
1129,
2185,
1444,
18,
2726,
14,
2550,
24,
35232,
1,
8859,
429,
268,
1,
8562,
25,
5317,
429,
268,
891,
16747,
397,
6499,
220,
679,
21,
12,
605,
12,
605,
3206,
220,
868,
25,
2166,
3743,
18816,
220,
1795,
23793,
429,
2831,
11038,
1,
2262,
429,
1342,
14056,
46117,
23060,
12,
23,
1,
2662,
18816,
220,
836,
429,
18774,
1,
2262,
429,
3175,
20297,
9531,
11,
2926,
13230,
28,
16,
1,
2662,
28436,
29,
-100
] |
African Entertainment better watch out! Performance, passion and expression have found their embodiment in the unbridled personality of Gina Castel.
Gina Castel is an actor, on-air personality, model, and fashion designer. Hailing from the Cape Coast, Ghana, Gina is a graduate of Mass Communication from Bayero University, Kano. A vivacious, mulit-talented persona, Gina works across a multitude of artistic and entertainment platforms, bringing her 'no-holds-barred' exuberance to each project.
A queen of the stage and a master of the screen, Gina Castel is a sight to behold when executing her acting skills. Her indepth performances, and often surprising dedication to the creation of her characters, have led to Gina being recognized as one of the next top acts to be on the look out for.
Gina often works alongside major figures in Nollywood, having to her record titles such as Covert Operations, a flick by Alex Mouth (of Mnet's Tinsel) where she played a blind woman alongside Patience Nzokwu, Olu Jacobs, Yemi Black; When Is It Enough written and directed by Daniel Ademinokan and Stella Damasus, where she played the female lead; others include Private Lies, Wife, Story Story (a BBC Radio production), Zozanatu, Affiong and a lot of other titles.
As a model Gina is known to rock a signature afro cut which tops off a graceful aura that glides across runways. And when she's not on the runway, Gina sketches and designs fashion wears.
One of Gina's biggest passions is being a Co-Presenter for the lifestyle, faith-based radio program THR3E, which airs on Praise World Radio and is powered by In His Steps magazine, and has been featured on Impact Detroit magazine.
One of the most fun things about this talented lady is her ability to adeptly imitate an impressive array of accents – Hausa, Igbo, Yoruba, British, Indian, American, etc. If you ever run into Gina and can cajole her into doing some accents for you, it'll probably be your most fun experience for the day!
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 2,388 |
[
128000,
32,
48893,
23334,
2731,
3821,
704,
0,
21304,
11,
11939,
323,
7645,
617,
1766,
872,
49497,
304,
279,
653,
16621,
839,
17743,
315,
77312,
11514,
301,
627,
38,
2259,
11514,
301,
374,
459,
12360,
11,
389,
38635,
17743,
11,
1646,
11,
323,
11401,
15034,
13,
473,
14612,
505,
279,
29715,
16377,
11,
48668,
11,
77312,
374,
264,
19560,
315,
9346,
31966,
505,
86520,
78,
3907,
11,
735,
5770,
13,
362,
18434,
19995,
11,
15975,
275,
2442,
278,
16243,
29055,
11,
77312,
4375,
4028,
264,
49068,
315,
32692,
323,
16924,
15771,
11,
12967,
1077,
364,
2201,
2902,
18938,
16150,
1171,
6,
506,
30970,
685,
311,
1855,
2447,
627,
32,
29006,
315,
279,
6566,
323,
264,
7491,
315,
279,
4264,
11,
77312,
11514,
301,
374,
264,
14254,
311,
57215
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
32,
48893,
23334,
2731,
3821,
704,
0,
21304,
11,
11939,
323,
7645,
617,
1766,
872,
49497,
304,
279,
653,
16621,
839,
17743,
315,
77312,
11514,
301,
627,
38,
2259,
11514,
301,
374,
459,
12360,
11,
389,
38635,
17743,
11,
1646,
11,
323,
11401,
15034,
13,
473,
14612,
505,
279,
29715,
16377,
11,
48668,
11,
77312,
374,
264,
19560,
315,
9346,
31966,
505,
86520,
78,
3907,
11,
735,
5770,
13,
362,
18434,
19995,
11,
15975,
275,
2442,
278,
16243,
29055,
11,
77312,
4375,
4028,
264,
49068,
315,
32692,
323,
16924,
15771,
11,
12967,
1077,
364,
2201,
2902,
18938,
16150,
1171,
6,
506,
30970,
685,
311,
1855,
2447,
627,
32,
29006,
315,
279,
6566,
323,
264,
7491,
315,
279,
4264,
11,
77312,
11514,
301,
374,
264,
14254,
311,
57215,
-100
] |
This is something I do with my students in my winter survival classes and quite frequently in my backyard when the snow is coming down. It is -1 degrees today in Flagstaff so it seemed like a good time to practice. It's one thing to start a survival fire under controlled conditions during a camping trip or when your body is in good shape but quite another when you are dealing with hypothermia and the reduced motor-skills that accompany it.
1. Place your hands in the snow for 2 minutes then 2) grab your different firemaking tools and try to ignite your tinder. Simple as that. 99% of the time the spark rod is my tool of choice as it requires gross motor skills unlike a lighter or matches and works when wet. Good luck trying to get a lighter going when you're all thumbs. Tinder is in the top pics. For my friends in Phoenix, get a bowl of ice cubes!
Since the majority of hypothermia cases happen in 50 degree F weather around the world and since most lost dayhikers are injured and hypothermic, it pays to build your skills and accompanying trail gear around the fact that you may have reduced motor-skills.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 2,968 |
[
128000,
2028,
374,
2555,
358,
656,
449,
856,
4236,
304,
856,
12688,
20237,
6989,
323,
5115,
14134,
304,
856,
36760,
994,
279,
12056,
374,
5108,
1523,
13,
1102,
374,
482,
16,
12628,
3432,
304,
23497,
28221,
779,
433,
9508,
1093,
264,
1695,
892,
311,
6725,
13,
1102,
596,
832,
3245,
311,
1212,
264,
20237,
4027,
1234,
14400,
4787,
2391,
264,
33873,
8577,
477,
994,
701,
2547,
374,
304,
1695,
6211,
719,
5115,
2500,
994,
499,
527,
14892,
449,
9950,
1605,
91299,
323,
279,
11293,
9048,
46407,
3385,
430,
19780,
433,
627,
16,
13,
11004,
701,
6206,
304,
279,
12056,
369,
220,
17,
4520,
1243,
220,
17,
8,
11894,
701,
2204,
4027,
28936,
7526,
323,
1456,
311,
90536,
701,
90265,
13,
9170,
439,
430,
13,
220,
1484,
4,
315
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
2028,
374,
2555,
358,
656,
449,
856,
4236,
304,
856,
12688,
20237,
6989,
323,
5115,
14134,
304,
856,
36760,
994,
279,
12056,
374,
5108,
1523,
13,
1102,
374,
482,
16,
12628,
3432,
304,
23497,
28221,
779,
433,
9508,
1093,
264,
1695,
892,
311,
6725,
13,
1102,
596,
832,
3245,
311,
1212,
264,
20237,
4027,
1234,
14400,
4787,
2391,
264,
33873,
8577,
477,
994,
701,
2547,
374,
304,
1695,
6211,
719,
5115,
2500,
994,
499,
527,
14892,
449,
9950,
1605,
91299,
323,
279,
11293,
9048,
46407,
3385,
430,
19780,
433,
627,
16,
13,
11004,
701,
6206,
304,
279,
12056,
369,
220,
17,
4520,
1243,
220,
17,
8,
11894,
701,
2204,
4027,
28936,
7526,
323,
1456,
311,
90536,
701,
90265,
13,
9170,
439,
430,
13,
220,
1484,
4,
315,
-100
] |
Q: NHibernate Linq // FirstOrDefault with children objects // Possible? I have an Employee.
Employee has a single (parent) Department. 1:N, Department to Employee.
Employee has N EmployeeToJobTitleMatchLink(s).
Employee has N number of ParkingArea(s).
When using NHibernate and Linq, and "FirstOrDefault"..
EmployeeNHEntity firstOrDefaultEmp = session.Query<EmployeeNHEntity>()
.Fetch(e => e.MyEmployeeToJobTitleMatchLinks)
.Fetch(e => e.ParentDepartment)
.Fetch(e => e.MyParkingAreas)
.Where(e => e.EmployeeUUID == employeeUuid)
.FirstOrDefault();
I get this sql. (the most important part is the "Top 1".
exec sp_executesql N'select TOP (1) employeenh0_.EmployeeUUID as Employee1_1_0_, myemployee1_.LinkSurrogateUUID as LinkSurr1_3_1_, department2_.DepartmentUUID as Departme1_0_2_, parkingare4_.ParkingAreaUUID as ParkingA1_5_3_, employeenh0_.MyVersionColumn as MyVersio2_1_0_, employeenh0_.SSN as SSN1_0_, employeenh0_.LastName as LastName1_0_, employeenh0_.FirstName as FirstName1_0_, employeenh0_.CreateDate as CreateDate1_0_, employeenh0_.HireDate as HireDate1_0_, employeenh0_.ParentDepartmentUUID as ParentDe8_1_0_, myemployee1_.PriorityRank as Priority2_3_1_, myemployee1_.JobStartedOnDate as JobStart3_3_1_, myemployee1_.TheEmployeeUUID as TheEmplo4_3_1_, myemployee1_.TheJobTitleUUID as TheJobTi5_3_1_, myemployee1_.TheEmployeeUUID as TheEmplo4_0__, myemployee1_.LinkSurrogateUUID as LinkSurr1_0__, department2_.MyVersionColumn as MyVersio2_0_2_, department2_.DepartmentName as Departme3_0_2_, department2_.CreateDate as CreateDate0_2_, parkingare4_.ParkingAreaName as ParkingA2_5_3_, parkingare4_.CreateDate as CreateDate5_3_, myparkinga3_.AbcEmployeeUUID as AbcEmplo1_1__, myparkinga3_.AbcParkingAreaUUID as AbcParki2_1__ from dbo.Employee employeenh0_ left outer join dbo.EmployeeToJobTitleMatchLink myemployee1_ on employeenh0_.EmployeeUUID=myemployee1_.TheEmployeeUUID left outer join dbo.Department department2_ on employeenh0_.ParentDepartmentUUID=department2_.DepartmentUUID left outer join EmployeeToParkingAreaLink myparkinga3_ on employeenh0_.EmployeeUUID=myparkinga3_.AbcEmployeeUUID left outer join dbo.ParkingArea parkingare4_ on myparkinga3_.AbcParkingAreaUUID=parkingare4_.ParkingAreaUUID where employeenh0_.EmployeeUUID=@p0',N'@p0 uniqueidentifier',@p0='64409648-15C5-4B67-BCE2-A265012E0926'
The issue is that I get 1 row (top 1). While I have 1 Employee, I (should) have many rows in the resultset.
Now, if I use the below, I get the desired result. (A single employee with the children objects populated)....if there is a match on the surrogate-key (EmployeeUUID), there is one item in the collection.
IList<EmployeeNHEntity> emps = session.Query<EmployeeNHEntity>()
.Fetch(e => e.MyEmployeeToJobTitleMatchLinks)
.Fetch(e => e.ParentDepartment)
.Fetch(e => e.MyParkingAreas)
.Where(e => e.EmployeeUUID == employeeUuid)
.ToList();
EmployeeNHEntity emp = null;
if(null!=emps)
{
if (emps.Count > 0)
{
emp = emps[0];
}
}
which generates this sql
exec sp_executesql N'select employeenh0_.EmployeeUUID as Employee1_1_0_, myemployee1_.LinkSurrogateUUID as LinkSurr1_3_1_, department2_.DepartmentUUID as Departme1_0_2_, parkingare4_.ParkingAreaUUID as ParkingA1_5_3_, employeenh0_.MyVersionColumn as MyVersio2_1_0_, employeenh0_.SSN as SSN1_0_, employeenh0_.LastName as LastName1_0_, employeenh0_.FirstName as FirstName1_0_, employeenh0_.CreateDate as CreateDate1_0_, employeenh0_.HireDate as HireDate1_0_, employeenh0_.ParentDepartmentUUID as ParentDe8_1_0_, myemployee1_.PriorityRank as Priority2_3_1_, myemployee1_.JobStartedOnDate as JobStart3_3_1_, myemployee1_.TheEmployeeUUID as TheEmplo4_3_1_, myemployee1_.TheJobTitleUUID as TheJobTi5_3_1_, myemployee1_.TheEmployeeUUID as TheEmplo4_0__, myemployee1_.LinkSurrogateUUID as LinkSurr1_0__, department2_.MyVersionColumn as MyVersio2_0_2_, department2_.DepartmentName as Departme3_0_2_, department2_.CreateDate as CreateDate0_2_, parkingare4_.ParkingAreaName as ParkingA2_5_3_, parkingare4_.CreateDate as CreateDate5_3_, myparkinga3_.AbcEmployeeUUID as AbcEmplo1_1__, myparkinga3_.AbcParkingAreaUUID as AbcParki2_1__ from dbo.Employee employeenh0_ left outer join dbo.EmployeeToJobTitleMatchLink myemployee1_ on employeenh0_.EmployeeUUID=myemployee1_.TheEmployeeUUID left outer join dbo.Department department2_ on employeenh0_.ParentDepartmentUUID=department2_.DepartmentUUID left outer join EmployeeToParkingAreaLink myparkinga3_ on employeenh0_.EmployeeUUID=myparkinga3_.AbcEmployeeUUID left outer join dbo.ParkingArea parkingare4_ on myparkinga3_.AbcParkingAreaUUID=parkingare4_.ParkingAreaUUID where employeenh0_.EmployeeUUID=@p0',N'@p0 uniqueidentifier',@p0='64409648-15C5-4B67-BCE2-A265012E0926'
The [0] indexer trick is ok. But just isn't as clean as I'd like my Linq code to be.
Any suggestions now LINQ syntax to get Top 1 Employee, but with all the children objects?
A: EmployeeNHEntity firstOrDefaultEmp = session.Query<EmployeeNHEntity>()
.Fetch(e => e.MyEmployeeToJobTitleMatchLinks)
.Fetch(e => e.ParentDepartment)
.Fetch(e => e.MyParkingAreas)
.Where(e => e.EmployeeUUID == employeeUuid)
.AsEnumerable() <-- note this line
.FirstOrDefault();
|
{
"redpajama_set_name": "RedPajamaStackExchange"
}
| 7,670 |
[
128000,
48,
25,
35931,
19348,
445,
6009,
443,
5629,
14522,
449,
2911,
6302,
443,
37207,
30,
358,
617,
459,
17275,
627,
17415,
706,
264,
3254,
320,
3850,
8,
6011,
13,
220,
16,
38794,
11,
6011,
311,
17275,
627,
17415,
706,
452,
17275,
1271,
12524,
3936,
8495,
4026,
1161,
4390,
17415,
706,
452,
1396,
315,
41282,
8900,
1161,
4390,
4599,
1701,
35931,
19348,
323,
445,
6009,
11,
323,
330,
27939,
1,
35047,
310,
17275,
45,
1837,
90358,
1176,
14522,
29831,
284,
3882,
16060,
59065,
45,
1837,
90358,
19485,
394,
662,
21373,
2069,
591,
384,
28152,
17415,
1271,
12524,
3936,
8495,
25056,
340,
394,
662,
21373,
2069,
591,
384,
30085,
27725,
340,
394,
662,
21373,
2069,
591,
384,
28152,
95416,
72337,
340,
394,
662,
9241,
2069,
591,
384,
52655,
25786
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
48,
25,
35931,
19348,
445,
6009,
443,
5629,
14522,
449,
2911,
6302,
443,
37207,
30,
358,
617,
459,
17275,
627,
17415,
706,
264,
3254,
320,
3850,
8,
6011,
13,
220,
16,
38794,
11,
6011,
311,
17275,
627,
17415,
706,
452,
17275,
1271,
12524,
3936,
8495,
4026,
1161,
4390,
17415,
706,
452,
1396,
315,
41282,
8900,
1161,
4390,
4599,
1701,
35931,
19348,
323,
445,
6009,
11,
323,
330,
27939,
1,
35047,
310,
17275,
45,
1837,
90358,
1176,
14522,
29831,
284,
3882,
16060,
59065,
45,
1837,
90358,
19485,
394,
662,
21373,
2069,
591,
384,
28152,
17415,
1271,
12524,
3936,
8495,
25056,
340,
394,
662,
21373,
2069,
591,
384,
30085,
27725,
340,
394,
662,
21373,
2069,
591,
384,
28152,
95416,
72337,
340,
394,
662,
9241,
2069,
591,
384,
52655,
25786,
-100
] |
Sept. 24, 2017 / 6:36 PM
Jose Bautista, offense carry Toronto Blue Jays over New York Yankees
Larry Millson, The Sports Xchange
Toronto Blue Jays Jose Bautista strikes out during the eight inning of a game against the Cleveland Indians at Progressive Field in Cleveland, Ohio on July 23, 2017. Photo by Aaron Josefczyk/UPI | License Photo
TORONTO -- Right fielder Jose Bautista might have played his last home game with the Toronto Blue Jays on Sunday.
If Bautista did, he did it right and so did the Blue Jays and their fans.
Receiving ovations with each at-bat, Bautista had two singles, a walk and an RBI to help the Blue Jays defeat the New York Yankees 9-5 in the rubber match of the three-game series.
"It was an exciting, the last home game and the focus on Jose," Blue Jays manager John Gibbons said. "He helped rebuild the team, he was the face of the franchise for a number of years."
RELATED Greg Bird's homer helps New York Yankees clinch postseason spot
Bautista was supported by Russell Martin's two doubles and three RBIs, and Teoscar Hernandez, who homered for the third straight game.
The Blue Jays (73-83) finish their season with six road games.
Aaron Judge hit his league-leading 47th and 48th homers and had three RBIs for the Yankees (86-67).
RELATED Todd Frazier: New York Yankees 3B falls for hidden ball trick vs. Toronto Blue Jays
Bautista took right field alone just before the game started as his teammates applauded from the dugout and the crowd gave a standing ovation. He was unaware that his teammates would not follow him on the field right away.
"We just decided that it was probably better if he just took the field alone to give him the credit that he deserved," center fielder Kevin Pillar said. "He's been a huge part of this franchise for a long time. Obviously with his future being uncertain we wanted him to go out in Jose fashion."
With one out in the ninth, Ezequiel Carrera replaced Bautista in right field. Bautista left the field to a standing ovation, hugged some teammates along the way and then returned from the dugout for a curtain call.
RELATED Ryan Goins, Toronto Blue Jays ride power surge in rout of New York Yankees
"I don't think he was aware of (the plan to substitute him in the ninth)," Gibbons said. "I know he appreciated it, I'm sure. He got a great ovation. The fans took care of him all day long as they should have."
"A lot of good emotions," Bautista said. "It's good to be recognized and it's good to feel the love and I appreciate everything that happened today. It is a possibility that I may not be back. Those decisions will be made and we'll see what happens."
"He's meant a lot to this city, he's meant a lot to this organization," Yankees manager Joe Girardi said. "You could see that he was emotional coming off the field. I thought it was nice what they did."
Blue Jays starter Marcus Stroman (13-8) allowed five hits, four walks and three runs while striking out two in 5 2/3 innings to earn his second straight win.
Yankees starter Jaime Garcia (5-10) allowed four hits, three walks and five runs in 2 1/3 innings. The left-hander struck out four.
The Yankees have clinched a wild-card spot, but their chances of winning the American League East are fading. "It's getting fairly large with seven games to go," Girardi said. "Obviously we're going to keep playing but I think you have to start to think ahead a little bit, too. It's frustrating because we've played really well the last six weeks.'
The Yankees are on a 16-7 roll.
"If it happens it happens, if it doesn't then we'll figure out the wild-card thing," Yankees designated hitter Chase Headley said. "You couldn't really ask for us to play any better than we have at this month. Unfortunately, Boston got hot at the same time."
Hernandez led off the bottom of the first with his fifth homer of the season and the first of his career leading off a game.
The Blue Jays scored again in the second on Darwin Barney's sacrifice fly that scored Pillar, who doubled and stole third.
Justin Smoak doubled with one out in the third, and Bautista and Kendrys Morales walked to load the bases. Jonathan Holder, who replaced Garcia, got a foul out to third before yielding Martin's bases-clearing hit to right center.
The Yankees scored in the fourth. Headley and Judge walked before Didi Gregorius hit an RBI single.
The Blue Jays scored four runs in the fourth against Bryan Mitchell. Josh Donaldson and Bautista had RBI singles and Morales had aa two-run single with Bautista being thrown out at home in a close play as he tried to score from first on the hit.
Judge led off the sixth with a homer. Stroman retired the next two batters on grounders before giving up a double to Jacoby Ellsbury and walking Todd Frazier.
Matt Dermody replaced Stroman and gave up an RBI double to Greg Bird before Ryan Tepera ended the inning on Austin Romine's grounder to third.
Tepera allowed a one-out single to Headley and Judge's second homer of the game and was replaced by Aaron Loup, who pitched around a single with two strikeouts to end the inning.
Toronto's Tom Koehler pitched a perfect eighth. Roberto Osuna struck out the side in the ninth.
NOTES: Toronto RF Jose Bautista singled to right on the first pitch to him in the first inning to end a 0-for-18 drought. ... Blue Jays LHP Brett Anderson (3-4, 7.15 ERA) will oppose Red Sox LHP Drew Pomeranz (16-5, 3.15) Monday in the opener of a three-game series at Boston. ... Yankees LHP CC Sabathia (12-5, 3.81) will face Kansas City Royals RHP Jake Junis (8-2, 4.05) Monday at Yankee Stadium. It is a makeup game for a postponement on May 25. ... The crowd of 47,394 at the Rogers Centre brought the total season attendance to 3,203,886.
Ezequiel Carrera
Roberto Osuna
Manager Joe Girardi
Manager John Gibbons
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 7,203 |
[
128000,
25669,
13,
220,
1187,
11,
220,
679,
22,
611,
220,
21,
25,
1927,
5975,
198,
46107,
426,
2784,
9265,
11,
17193,
6920,
14974,
8868,
66835,
927,
1561,
4356,
45383,
198,
89595,
8384,
942,
11,
578,
13482,
1630,
3455,
198,
72437,
8868,
66835,
11097,
426,
2784,
9265,
23170,
704,
2391,
279,
8223,
46386,
315,
264,
1847,
2403,
279,
24372,
30507,
520,
52870,
8771,
304,
24372,
11,
14689,
389,
5887,
220,
1419,
11,
220,
679,
22,
13,
11064,
555,
26757,
11097,
8480,
4341,
74,
63110,
1932,
765,
1914,
11064,
198,
50758,
715,
5319,
1198,
10291,
2115,
261,
11097,
426,
2784,
9265,
2643,
617,
6476,
813,
1566,
2162,
1847,
449,
279,
14974,
8868,
66835,
389,
7418,
627,
2746,
426,
2784,
9265,
1550,
11,
568,
1550,
433,
1314,
323,
779,
1550
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
25669,
13,
220,
1187,
11,
220,
679,
22,
611,
220,
21,
25,
1927,
5975,
198,
46107,
426,
2784,
9265,
11,
17193,
6920,
14974,
8868,
66835,
927,
1561,
4356,
45383,
198,
89595,
8384,
942,
11,
578,
13482,
1630,
3455,
198,
72437,
8868,
66835,
11097,
426,
2784,
9265,
23170,
704,
2391,
279,
8223,
46386,
315,
264,
1847,
2403,
279,
24372,
30507,
520,
52870,
8771,
304,
24372,
11,
14689,
389,
5887,
220,
1419,
11,
220,
679,
22,
13,
11064,
555,
26757,
11097,
8480,
4341,
74,
63110,
1932,
765,
1914,
11064,
198,
50758,
715,
5319,
1198,
10291,
2115,
261,
11097,
426,
2784,
9265,
2643,
617,
6476,
813,
1566,
2162,
1847,
449,
279,
14974,
8868,
66835,
389,
7418,
627,
2746,
426,
2784,
9265,
1550,
11,
568,
1550,
433,
1314,
323,
779,
1550,
-100
] |
CC = gcc
CXX = g++
DEFINES = -DQT_NO_DEBUG -DQT_GUI_LIB -DQT_CORE_LIB -DQT_SHARED
CFLAGS = -m64 -pipe -O2 -Wall -W -D_REENTRANT $(DEFINES)
CXXFLAGS = -m64 -pipe -O2 -Wall -W -D_REENTRANT $(DEFINES)
INCPATH = -I/usr/share/qt4/mkspecs/linux-g++-64 -I. -I/usr/include/qt4/QtCore -I/usr/include/qt4/QtGui -I/usr/include/qt4 -I. -I.
LINK = g++
LFLAGS = -m64 -Wl,-O1
LIBS = $(SUBLIBS) -L/usr/lib/x86_64-linux-gnu -lQtGui -lQtCore -lpthread
AR = ar cqs
RANLIB =
QMAKE = /usr/lib/x86_64-linux-gnu/qt4/bin/qmake
TAR = tar -cf
COMPRESS = gzip -9f
COPY = cp -f
SED = sed
COPY_FILE = $(COPY)
COPY_DIR = $(COPY) -r
STRIP = strip
INSTALL_FILE = install -m 644 -p
INSTALL_DIR = $(COPY_DIR)
INSTALL_PROGRAM = install -m 755 -p
DEL_FILE = rm -f
SYMLINK = ln -f -s
DEL_DIR = rmdir
MOVE = mv -f
CHK_DIR_EXISTS= test -d
MKDIR = mkdir -p
####### Output directory
OBJECTS_DIR = ./
####### Files
SOURCES = loadWindow.cpp \
main.cpp \
portfolio.cpp \
portfolioManager.cpp \
property.cpp \
tenant.cpp moc_loadWindow.cpp \
moc_portfolioManager.cpp
OBJECTS = loadWindow.o \
main.o \
portfolio.o \
portfolioManager.o \
property.o \
tenant.o \
moc_loadWindow.o \
moc_portfolioManager.o
DIST = /usr/share/qt4/mkspecs/common/unix.conf \
/usr/share/qt4/mkspecs/common/linux.conf \
/usr/share/qt4/mkspecs/common/gcc-base.conf \
/usr/share/qt4/mkspecs/common/gcc-base-unix.conf \
/usr/share/qt4/mkspecs/common/g++-base.conf \
/usr/share/qt4/mkspecs/common/g++-unix.conf \
/usr/share/qt4/mkspecs/qconfig.pri \
/usr/share/qt4/mkspecs/modules/qt_phonon.pri \
/usr/share/qt4/mkspecs/features/qt_functions.prf \
/usr/share/qt4/mkspecs/features/qt_config.prf \
/usr/share/qt4/mkspecs/features/exclusive_builds.prf \
/usr/share/qt4/mkspecs/features/default_pre.prf \
/usr/share/qt4/mkspecs/features/release.prf \
/usr/share/qt4/mkspecs/features/default_post.prf \
/usr/share/qt4/mkspecs/features/shared.prf \
/usr/share/qt4/mkspecs/features/unix/gdb_dwarf_index.prf \
/usr/share/qt4/mkspecs/features/warn_on.prf \
/usr/share/qt4/mkspecs/features/qt.prf \
/usr/share/qt4/mkspecs/features/unix/thread.prf \
/usr/share/qt4/mkspecs/features/moc.prf \
/usr/share/qt4/mkspecs/features/resources.prf \
/usr/share/qt4/mkspecs/features/uic.prf \
/usr/share/qt4/mkspecs/features/yacc.prf \
/usr/share/qt4/mkspecs/features/lex.prf \
/usr/share/qt4/mkspecs/features/include_source_dir.prf \
long.pro
QMAKE_TARGET = long
DESTDIR =
TARGET = long
first: all
####### Implicit rules
.SUFFIXES: .o .c .cpp .cc .cxx .C
.cpp.o:
$(CXX) -c $(CXXFLAGS) $(INCPATH) -o "$@" "$<"
.cc.o:
$(CXX) -c $(CXXFLAGS) $(INCPATH) -o "$@" "$<"
.cxx.o:
$(CXX) -c $(CXXFLAGS) $(INCPATH) -o "$@" "$<"
.C.o:
$(CXX) -c $(CXXFLAGS) $(INCPATH) -o "$@" "$<"
.c.o:
$(CC) -c $(CFLAGS) $(INCPATH) -o "$@" "$<"
####### Build rules
all: Makefile $(TARGET)
$(TARGET): $(OBJECTS)
$(LINK) $(LFLAGS) -o $(TARGET) $(OBJECTS) $(OBJCOMP) $(LIBS)
Makefile: long.pro /usr/share/qt4/mkspecs/linux-g++-64/qmake.conf /usr/share/qt4/mkspecs/common/unix.conf \
/usr/share/qt4/mkspecs/common/linux.conf \
/usr/share/qt4/mkspecs/common/gcc-base.conf \
/usr/share/qt4/mkspecs/common/gcc-base-unix.conf \
/usr/share/qt4/mkspecs/common/g++-base.conf \
/usr/share/qt4/mkspecs/common/g++-unix.conf \
/usr/share/qt4/mkspecs/qconfig.pri \
/usr/share/qt4/mkspecs/modules/qt_phonon.pri \
/usr/share/qt4/mkspecs/features/qt_functions.prf \
/usr/share/qt4/mkspecs/features/qt_config.prf \
/usr/share/qt4/mkspecs/features/exclusive_builds.prf \
/usr/share/qt4/mkspecs/features/default_pre.prf \
/usr/share/qt4/mkspecs/features/release.prf \
/usr/share/qt4/mkspecs/features/default_post.prf \
/usr/share/qt4/mkspecs/features/shared.prf \
/usr/share/qt4/mkspecs/features/unix/gdb_dwarf_index.prf \
/usr/share/qt4/mkspecs/features/warn_on.prf \
/usr/share/qt4/mkspecs/features/qt.prf \
/usr/share/qt4/mkspecs/features/unix/thread.prf \
/usr/share/qt4/mkspecs/features/moc.prf \
/usr/share/qt4/mkspecs/features/resources.prf \
/usr/share/qt4/mkspecs/features/uic.prf \
/usr/share/qt4/mkspecs/features/yacc.prf \
/usr/share/qt4/mkspecs/features/lex.prf \
/usr/share/qt4/mkspecs/features/include_source_dir.prf \
/usr/lib/x86_64-linux-gnu/libQtGui.prl \
/usr/lib/x86_64-linux-gnu/libQtCore.prl
$(QMAKE) -o Makefile long.pro
/usr/share/qt4/mkspecs/common/unix.conf:
/usr/share/qt4/mkspecs/common/linux.conf:
/usr/share/qt4/mkspecs/common/gcc-base.conf:
/usr/share/qt4/mkspecs/common/gcc-base-unix.conf:
/usr/share/qt4/mkspecs/common/g++-base.conf:
/usr/share/qt4/mkspecs/common/g++-unix.conf:
/usr/share/qt4/mkspecs/qconfig.pri:
/usr/share/qt4/mkspecs/modules/qt_phonon.pri:
/usr/share/qt4/mkspecs/features/qt_functions.prf:
/usr/share/qt4/mkspecs/features/qt_config.prf:
/usr/share/qt4/mkspecs/features/exclusive_builds.prf:
/usr/share/qt4/mkspecs/features/default_pre.prf:
/usr/share/qt4/mkspecs/features/release.prf:
/usr/share/qt4/mkspecs/features/default_post.prf:
/usr/share/qt4/mkspecs/features/shared.prf:
/usr/share/qt4/mkspecs/features/unix/gdb_dwarf_index.prf:
/usr/share/qt4/mkspecs/features/warn_on.prf:
/usr/share/qt4/mkspecs/features/qt.prf:
/usr/share/qt4/mkspecs/features/unix/thread.prf:
/usr/share/qt4/mkspecs/features/moc.prf:
/usr/share/qt4/mkspecs/features/resources.prf:
/usr/share/qt4/mkspecs/features/uic.prf:
/usr/share/qt4/mkspecs/features/yacc.prf:
/usr/share/qt4/mkspecs/features/lex.prf:
/usr/share/qt4/mkspecs/features/include_source_dir.prf:
/usr/lib/x86_64-linux-gnu/libQtGui.prl:
/usr/lib/x86_64-linux-gnu/libQtCore.prl:
qmake: FORCE
@$(QMAKE) -o Makefile long.pro
dist:
@$(CHK_DIR_EXISTS) .tmp/long1.0.0 || $(MKDIR) .tmp/long1.0.0
$(COPY_FILE) --parents $(SOURCES) $(DIST) .tmp/long1.0.0/ && $(COPY_FILE) --parents loadWindow.h multiProperty.h portfolio.h portfolioManager.h property.h tenant.h .tmp/long1.0.0/ && $(COPY_FILE) --parents loadWindow.cpp main.cpp portfolio.cpp portfolioManager.cpp property.cpp tenant.cpp .tmp/long1.0.0/ && (cd `dirname .tmp/long1.0.0` && $(TAR) long1.0.0.tar long1.0.0 && $(COMPRESS) long1.0.0.tar) && $(MOVE) `dirname .tmp/long1.0.0`/long1.0.0.tar.gz . && $(DEL_FILE) -r .tmp/long1.0.0
clean:compiler_clean
-$(DEL_FILE) $(OBJECTS)
-$(DEL_FILE) *~ core *.core
####### Sub-libraries
distclean: clean
-$(DEL_FILE) $(TARGET)
-$(DEL_FILE) Makefile
check: first
mocclean: compiler_moc_header_clean compiler_moc_source_clean
mocables: compiler_moc_header_make_all compiler_moc_source_make_all
compiler_moc_header_make_all: moc_loadWindow.cpp moc_portfolioManager.cpp
compiler_moc_header_clean:
-$(DEL_FILE) moc_loadWindow.cpp moc_portfolioManager.cpp
moc_loadWindow.cpp: portfolioManager.h \
portfolio.h \
tenant.h \
property.h \
loadWindow.h
/usr/lib/x86_64-linux-gnu/qt4/bin/moc $(DEFINES) $(INCPATH) loadWindow.h -o moc_loadWindow.cpp
moc_portfolioManager.cpp: portfolio.h \
tenant.h \
property.h \
portfolioManager.h
/usr/lib/x86_64-linux-gnu/qt4/bin/moc $(DEFINES) $(INCPATH) portfolioManager.h -o moc_portfolioManager.cpp
compiler_rcc_make_all:
compiler_rcc_clean:
compiler_image_collection_make_all: qmake_image_collection.cpp
compiler_image_collection_clean:
-$(DEL_FILE) qmake_image_collection.cpp
compiler_moc_source_make_all:
compiler_moc_source_clean:
compiler_uic_make_all:
compiler_uic_clean:
compiler_yacc_decl_make_all:
compiler_yacc_decl_clean:
compiler_yacc_impl_make_all:
compiler_yacc_impl_clean:
compiler_lex_make_all:
compiler_lex_clean:
compiler_clean: compiler_moc_header_clean
####### Compile
loadWindow.o: loadWindow.cpp loadWindow.h \
portfolioManager.h \
portfolio.h \
tenant.h \
property.h
$(CXX) -c $(CXXFLAGS) $(INCPATH) -o loadWindow.o loadWindow.cpp
main.o: main.cpp loadWindow.h \
portfolioManager.h \
portfolio.h \
tenant.h \
property.h
$(CXX) -c $(CXXFLAGS) $(INCPATH) -o main.o main.cpp
portfolio.o: portfolio.cpp portfolio.h \
tenant.h \
property.h
$(CXX) -c $(CXXFLAGS) $(INCPATH) -o portfolio.o portfolio.cpp
portfolioManager.o: portfolioManager.cpp portfolioManager.h \
portfolio.h \
tenant.h \
property.h
$(CXX) -c $(CXXFLAGS) $(INCPATH) -o portfolioManager.o portfolioManager.cpp
property.o: property.cpp property.h
$(CXX) -c $(CXXFLAGS) $(INCPATH) -o property.o property.cpp
tenant.o: tenant.cpp tenant.h
$(CXX) -c $(CXXFLAGS) $(INCPATH) -o tenant.o tenant.cpp
moc_loadWindow.o: moc_loadWindow.cpp
$(CXX) -c $(CXXFLAGS) $(INCPATH) -o moc_loadWindow.o moc_loadWindow.cpp
moc_portfolioManager.o: moc_portfolioManager.cpp
$(CXX) -c $(CXXFLAGS) $(INCPATH) -o moc_portfolioManager.o moc_portfolioManager.cpp
####### Install
install: FORCE
uninstall: FORCE
FORCE:
|
{
"redpajama_set_name": "RedPajamaGithub"
}
| 5,603 |
[
128000,
3791,
310,
284,
52017,
198,
34,
6277,
1881,
284,
342,
23792,
13963,
34708,
996,
284,
482,
35,
44778,
9278,
11386,
482,
35,
44778,
62405,
27299,
482,
35,
44778,
32410,
27299,
482,
35,
44778,
55818,
198,
34,
33848,
286,
284,
482,
76,
1227,
482,
14086,
482,
46,
17,
482,
33697,
482,
54,
482,
35,
2241,
1863,
49,
2891,
5035,
13963,
34708,
340,
34,
6277,
33848,
415,
284,
482,
76,
1227,
482,
14086,
482,
46,
17,
482,
33697,
482,
54,
482,
35,
2241,
1863,
49,
2891,
5035,
13963,
34708,
340,
691,
7269,
4932,
996,
284,
482,
40,
55438,
35014,
91076,
19,
3262,
2857,
1007,
82,
64944,
2427,
1044,
12,
1227,
482,
40,
13,
482,
40,
55438,
27491,
91076,
19,
14,
76302,
482,
40,
55438,
27491,
91076,
19,
14,
19377
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
3791,
310,
284,
52017,
198,
34,
6277,
1881,
284,
342,
23792,
13963,
34708,
996,
284,
482,
35,
44778,
9278,
11386,
482,
35,
44778,
62405,
27299,
482,
35,
44778,
32410,
27299,
482,
35,
44778,
55818,
198,
34,
33848,
286,
284,
482,
76,
1227,
482,
14086,
482,
46,
17,
482,
33697,
482,
54,
482,
35,
2241,
1863,
49,
2891,
5035,
13963,
34708,
340,
34,
6277,
33848,
415,
284,
482,
76,
1227,
482,
14086,
482,
46,
17,
482,
33697,
482,
54,
482,
35,
2241,
1863,
49,
2891,
5035,
13963,
34708,
340,
691,
7269,
4932,
996,
284,
482,
40,
55438,
35014,
91076,
19,
3262,
2857,
1007,
82,
64944,
2427,
1044,
12,
1227,
482,
40,
13,
482,
40,
55438,
27491,
91076,
19,
14,
76302,
482,
40,
55438,
27491,
91076,
19,
14,
19377,
-100
] |
Do you know that all guests can try wearing hanbok which is traditional clothing in Korea in Leo guest house for free?!!
Today! Our beautiful Russian guests tried wearing it and took lots of pictures!
Look at these! Aren't they lovely?
You also can see our new make-up room! which Leo made it all!!
Come and enjoy it with us!!
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 2,760 |
[
128000,
5519,
499,
1440,
430,
682,
15051,
649,
1456,
12512,
19538,
65,
564,
902,
374,
8776,
17895,
304,
12126,
304,
37848,
8810,
3838,
369,
1949,
30,
51447,
15724,
0,
5751,
6366,
8690,
15051,
6818,
12512,
433,
323,
3952,
10283,
315,
9364,
4999,
10596,
520,
1521,
0,
74604,
956,
814,
17104,
5380,
2675,
1101,
649,
1518,
1057,
502,
1304,
5352,
3130,
0,
902,
37848,
1903,
433,
682,
51447,
29951,
323,
4774,
433,
449,
603,
3001,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
] |
[
5519,
499,
1440,
430,
682,
15051,
649,
1456,
12512,
19538,
65,
564,
902,
374,
8776,
17895,
304,
12126,
304,
37848,
8810,
3838,
369,
1949,
30,
51447,
15724,
0,
5751,
6366,
8690,
15051,
6818,
12512,
433,
323,
3952,
10283,
315,
9364,
4999,
10596,
520,
1521,
0,
74604,
956,
814,
17104,
5380,
2675,
1101,
649,
1518,
1057,
502,
1304,
5352,
3130,
0,
902,
37848,
1903,
433,
682,
51447,
29951,
323,
4774,
433,
449,
603,
3001,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100
] |
New teams are always trying to figure out which of the basic Agile practices they can drop with impunity. One that Agile coaches tend to stress as critical is co-location, but information to back this up has been largely anecdotal. On the ScrumDevelopment list recently, Nicholas Cancelliere launched an interesting conversation by posting a link to this article [pdf], about a study comparing productivity gains in a "war room" style collaborative workspace versus a more traditional corporate cube-farm.
The four authors, from University of Michigan and Rutgers University, reported, in a combination of case study and empirical evaluation, on research conducted at a Fortune 50 automobile company. They ran a field study of six co-located teams, tracking their activity, attitudes, use of technology and productivity. They compared their results with metrics collected on teams at the same company doing software development in the traditional office arrangement. Since co-location does not work for every size of group, all projects were scoped to an estimated 4 months of work for a team of 6-8 people. The writers admit that their study is "confounded" in that two variables were introduced: teams were co-located and scope was "timeboxed," which are two separate Agile strategies, though usually practiced together. Most of the focus in this investigation was on the communication patterns of the individuals on the team.
So, what did they find? Teams in these "war rooms" showed a doubling of productivity. Factors identified as significant included: teams had easy access to each other for both coordination of their work and for learning, and the work artifacts they posted on the walls remained visible to all.
There needs to be provision for privacy, for calls about that disturbing rash, for example. Open rooms need to be inviting: windows and air and such are good for groups, not just high-status individuals.
It just has to be as Ron says, "percieved as a privilege", rather than "percieved as a cheap alternative to cubicle walls"... When the server room and the call center have more elbow room than the development environment, it is not the kind of colocation that will benefit the company.
I take your main point as being that with sufficient creativity, people can screw up even the best ideas. I fully agree.
I agree that any change this personal comes with both perceived and real loss that requires team ownership of the change and understanding of the benefits.
My findings, taken from survey feedback, show similar results to this study with respect to productivity, but also have shown an increase in software quality as well... Our findings also showed that balancing individual's personal space with team space was critical. While a team-only space works for a short term, integrating personal space was required for maintaining for a longer period.
We're wondering whether it makes sense to separate the teams physically. The different teams will probably be working on the same codebase - multiple teams drawing from the same backlog. I would love to hear suggestions about this; it seems tricky to decouple the teams entirely.
Various colleagues contributed ideas on this one... this list is a great place for both newbies and experienced practitioners to ask questions about applying Scrum in the real world.
In one case we moved a team 2 floors from their other related teams to create them a co-located space. While it increased the internal team's productivity and quality, it decreased the intra-team communication and cross-team dependencies were missed. Thus, it was determined that both inter- and intra-team co-location is important.
An interesting side note: the article writers quote a 1977 study which found that "A distance of 30 meters is equivalent to being truly remote" (Allen, T. J., 1977, see the article for the citation) The researchers themselves deduced that "if we are to truly support remote teams, we should provide constant awareness and easy transitions in and out of spontaneous meetings." Jeff Sutherland's SirsiDynix case study supports this, showing that even teams with significant off-shore components can achieve results comparable to a co-located team, if conscious and adequate compensation is made for the loss of high-bandwidth (face-to-face) communication.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 7,831 |
[
128000,
3648,
7411,
527,
2744,
4560,
311,
7216,
704,
902,
315,
279,
6913,
83284,
12659,
814,
649,
6068,
449,
98770,
13,
3861,
430,
83284,
24909,
8541,
311,
8631,
439,
9200,
374,
1080,
61970,
11,
719,
2038,
311,
1203,
420,
709,
706,
1027,
14090,
50750,
2426,
13,
1952,
279,
2522,
10952,
40519,
1160,
6051,
11,
40796,
356,
14575,
19380,
11887,
459,
7185,
10652,
555,
17437,
264,
2723,
311,
420,
4652,
510,
12091,
1145,
922,
264,
4007,
27393,
26206,
20192,
304,
264,
330,
11710,
3130,
1,
1742,
40806,
28614,
19579,
264,
810,
8776,
13166,
24671,
2269,
2227,
627,
791,
3116,
12283,
11,
505,
3907,
315,
14972,
323,
72091,
3907,
11,
5068,
11,
304,
264,
10824,
315,
1162,
4007,
323,
46763,
16865,
11,
389,
3495,
13375,
520,
264,
46331,
220,
1135
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
3648,
7411,
527,
2744,
4560,
311,
7216,
704,
902,
315,
279,
6913,
83284,
12659,
814,
649,
6068,
449,
98770,
13,
3861,
430,
83284,
24909,
8541,
311,
8631,
439,
9200,
374,
1080,
61970,
11,
719,
2038,
311,
1203,
420,
709,
706,
1027,
14090,
50750,
2426,
13,
1952,
279,
2522,
10952,
40519,
1160,
6051,
11,
40796,
356,
14575,
19380,
11887,
459,
7185,
10652,
555,
17437,
264,
2723,
311,
420,
4652,
510,
12091,
1145,
922,
264,
4007,
27393,
26206,
20192,
304,
264,
330,
11710,
3130,
1,
1742,
40806,
28614,
19579,
264,
810,
8776,
13166,
24671,
2269,
2227,
627,
791,
3116,
12283,
11,
505,
3907,
315,
14972,
323,
72091,
3907,
11,
5068,
11,
304,
264,
10824,
315,
1162,
4007,
323,
46763,
16865,
11,
389,
3495,
13375,
520,
264,
46331,
220,
1135,
-100
] |
Technicians are no longer alone in needing a multi-brand diagnostic tool to dialogue with electronic control units. Diagnostic tools are rapidly becoming indispensable to technicians working in specific areas of vehicle maintenance, such as tyre fitters, fast-fit service stations.
Until today, no diagnostic tool really fulfilled the needs of vehicle specialists of this kind. If they wanted one, they had to buy a costly and sophisticated model intended for the all-round mechanic, though they would never need most of the functions it provided.
Now, TEXA has filled this gap in the market with AXONE S. The AXONE S is the first product of its kind: a true multi-utility diagnostic tool.
AXONE S is equipped with a generous 5 inch capacitive LCD-TFT touchscreen. This type of display offers numerous benefits, including high sensitivity for a rapid response to even the lightest touch, and excellent luminosity.
AXONE S runs on a Cortex-A8 800 Mhz processor with a 2D and 3D graphic accelerator. AXONE S comes complete with the robust Navigator nano S vehicle interface.
The AXONE S uses the Jelly Bean 4.2.2 version of the Android™ operating system. Development for Android™ has made TEXA's celebrated IDC4a diagnostic software easier than ever to use, thanks to the operating system's highly intuitive gestures.
IDC4a for Android™ is ideal for the purposes for which AXONE S is intended, i.e. a multi-utility tool for daily use focused on a specific set of operations that need to be completed conveniently and quickly.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 6,443 |
[
128000,
29356,
9818,
527,
912,
5129,
7636,
304,
33921,
264,
7447,
34220,
15439,
5507,
311,
21976,
449,
14683,
2585,
8316,
13,
51088,
7526,
527,
19019,
10671,
64284,
311,
53459,
3318,
304,
3230,
5789,
315,
7458,
13709,
11,
1778,
439,
74028,
282,
29163,
11,
5043,
50660,
2532,
17789,
627,
25503,
3432,
11,
912,
15439,
5507,
2216,
41834,
279,
3966,
315,
7458,
35416,
315,
420,
3169,
13,
1442,
814,
4934,
832,
11,
814,
1047,
311,
3780,
264,
34348,
323,
27877,
1646,
10825,
369,
279,
682,
21691,
45239,
11,
3582,
814,
1053,
2646,
1205,
1455,
315,
279,
5865,
433,
3984,
627,
7184,
11,
77324,
32,
706,
10409,
420,
13225,
304,
279,
3157,
449,
41811,
5338,
328,
13,
578,
41811,
5338,
328,
374,
279,
1176,
2027,
315,
1202,
3169,
25,
264,
837
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
29356,
9818,
527,
912,
5129,
7636,
304,
33921,
264,
7447,
34220,
15439,
5507,
311,
21976,
449,
14683,
2585,
8316,
13,
51088,
7526,
527,
19019,
10671,
64284,
311,
53459,
3318,
304,
3230,
5789,
315,
7458,
13709,
11,
1778,
439,
74028,
282,
29163,
11,
5043,
50660,
2532,
17789,
627,
25503,
3432,
11,
912,
15439,
5507,
2216,
41834,
279,
3966,
315,
7458,
35416,
315,
420,
3169,
13,
1442,
814,
4934,
832,
11,
814,
1047,
311,
3780,
264,
34348,
323,
27877,
1646,
10825,
369,
279,
682,
21691,
45239,
11,
3582,
814,
1053,
2646,
1205,
1455,
315,
279,
5865,
433,
3984,
627,
7184,
11,
77324,
32,
706,
10409,
420,
13225,
304,
279,
3157,
449,
41811,
5338,
328,
13,
578,
41811,
5338,
328,
374,
279,
1176,
2027,
315,
1202,
3169,
25,
264,
837,
-100
] |
Create .m3u playlists with drag&drop of files and folders. For each folder a single .m3u file is created named after this folder. It will not replace already existing playlists of that name. The generated playlists do not contain EXTM3U information because a simple .m3u is nothing more than a text file with a list of file names per line. The initial tool for this task for me was a batch file ;-) As the generated playlists only include filenames without any path, as they are located along with the media files, they can be moved together with the files, for example for portable mp3 player usage.
Creates .m3u playlists for each folder.
Recursivly works on all included subfolders.
Creates .m3u playlist for dropped files.
Playlist named after folder name.
Preserves existing playlists (by the same name).
Playlists are sorted by filename.
Can be configured (a bit) in .ini file.
Drag files and folders onto compiled CreateM3U.exe (or a shortcut to it). This is limited, you can double click to open a window with an less limited drag and drop area instead.
where [*] are replaced with full paths (without ) and you are limited to nine files/folders to drop.
Better "About" box including help.
Is EXTM3U possible and useful?
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 4,279 |
[
128000,
4110,
662,
76,
18,
84,
69908,
449,
11161,
5,
6861,
315,
3626,
323,
30342,
13,
1789,
1855,
8695,
264,
3254,
662,
76,
18,
84,
1052,
374,
3549,
7086,
1306,
420,
8695,
13,
1102,
690,
539,
8454,
2736,
6484,
69908,
315,
430,
836,
13,
578,
8066,
69908,
656,
539,
6782,
4154,
22809,
18,
52,
2038,
1606,
264,
4382,
662,
76,
18,
84,
374,
4400,
810,
1109,
264,
1495,
1052,
449,
264,
1160,
315,
1052,
5144,
824,
1584,
13,
578,
2926,
5507,
369,
420,
3465,
369,
757,
574,
264,
7309,
1052,
2652,
63096,
1666,
279,
8066,
69908,
1193,
2997,
41640,
2085,
904,
1853,
11,
439,
814,
527,
7559,
3235,
449,
279,
3772,
3626,
11,
814,
649,
387,
7882,
3871,
449,
279,
3626,
11,
369,
3187,
369,
23665,
10710,
18
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
4110,
662,
76,
18,
84,
69908,
449,
11161,
5,
6861,
315,
3626,
323,
30342,
13,
1789,
1855,
8695,
264,
3254,
662,
76,
18,
84,
1052,
374,
3549,
7086,
1306,
420,
8695,
13,
1102,
690,
539,
8454,
2736,
6484,
69908,
315,
430,
836,
13,
578,
8066,
69908,
656,
539,
6782,
4154,
22809,
18,
52,
2038,
1606,
264,
4382,
662,
76,
18,
84,
374,
4400,
810,
1109,
264,
1495,
1052,
449,
264,
1160,
315,
1052,
5144,
824,
1584,
13,
578,
2926,
5507,
369,
420,
3465,
369,
757,
574,
264,
7309,
1052,
2652,
63096,
1666,
279,
8066,
69908,
1193,
2997,
41640,
2085,
904,
1853,
11,
439,
814,
527,
7559,
3235,
449,
279,
3772,
3626,
11,
814,
649,
387,
7882,
3871,
449,
279,
3626,
11,
369,
3187,
369,
23665,
10710,
18,
-100
] |
I have from time to time, I kid you not, threatened to punish our oldest son for punishing himself.
And here, right at the start, I have to confess something. I don't know what I'm talking about.
Generally when I write something to share with you good readers, I like to come from a position of some knowledge, personal experience, and understanding. But I simply don't comprehend where my son's behavior comes from, and I'm writing this in the hope that by discussing it here (and hopefully eliciting your comments) I can penetrate the murky waters of this – to me -most unfathomable phenomenon.
He does, however, deny himself pleasure when he feels that he's done something wrong, and is inappropriately hard on himself for minor, even accidental, behavioral infractions.
Me: You ate almost all of it, go ahead and get yourself something.
D: But there's still a lot left on my plate.
Me: No, D, there's not. You had a good dinner. You can have a treat.
D: I didn't eat all of the potatoes.
Me: Oh, fachrissakes, just get yourself a treat you whimpering, sniveling little sod! [No, no, that's just what I'm thinking. I actually say something more along the lines of…] Well, it's up to you. I'm telling you that it's okay for you to have a treat, but if you don't feel good about it, I'll leave that decision to you.
All right, so this is a petty bit of self-inflicted atonement for a crime uncommitted. Perhaps it's nothing more than him fishing for approval and affirmation. A minor martyr complex underpinned by shaky self-esteem.
For example. While sword-fighting with his little brother, he accidentally cracks the midget on the cranium. Little brother begins to cry. Now here's the thing. When I was a kid, my mind would have immediately begun spinning out stories, extenuating circumstances, justifications – the sun was in my eyes, I had a muscle spasm and the stick flew from my hand, high winds, planetary convergences, and he started it, anyway.
D, on the other hand, will begin to berate himself and even, on occasion, smack his own head in a febrile fit of self-reproach.
For example. The classic case of spilled milk. D knocks over a full glass at the dining room table, and we all jump up to grab towels and clear the mess.
He's disconsolate, reviling himself for his stupidity and slobbering abject apologies over and over as though his mother and I are going to burn him with cigars and toss him down a well, when what we're actually doing is telling him not to worry, accidents happen, it's no big deal.
Aha, you say. I think I've got it. The boy suffers from low self-esteem. Well yes, I've thought this before and this composition confirms in me that yes, of course he does. That's not the question. The question is why. And the follow-up question, of course, is: How do we fix this?
So if I can't really get a grasp on the why, how can I hope to help him? I've come up with three specific strategies that I plan to implement.
I'm going to give him more responsibilities around the house. We've assigned daily and weekly chores before, but they often slip by the wayside and are forgotten. I think that when kids are given tasks and are able to carry them out, they feel responsible, confident, and accomplished.
I'm going to sit down with him at least a few nights a week to study Spanish together. Not only will studying together provide us with a shared goal, but as his skills improve he will feel good about his few-found abilities.
I'm going to get him into martial arts lessons. While I have a certain antipathy towards pushing kids into doing a million extra-curricular activities and believe that it's actually detrimental to spread them too thinly, learning a skill thoroughly and performing it at a high level is a great way to feel good about yourself.
Well, I think this little exercise has been useful. If not for you, then certainly for me. Sometimes just laying something out on the page is all it takes to clarify it for you.
So what do you think? Does this seem reasonable and rational or am I missing something here? Do any of your kids ever punish themselves unnecessarily? If so, what did you do? I'd love to hear your stories and, please, your advice.
Is it Okay for Kids to Hit Back?
Be Nice, You Rancid Little Punk!
I use to put myself in time out when I was little.
I'd say around 2-9 years old.
I'd do it even if I had a bad thought in my head like call my mum a bad word, even though no one had heard me say it.
I actually use to enjoy the time out. It was my little solitude and it was always on the kitchen corner floor.
If kids want to do it then let them, if they do this they are analysing and processing their thoughts.
It's not a bad thing and is actually really good for their personal development because we learn how to problem solve and think out the box.
Okay, I stumbled upon this site while looking to the answer to my own issue, which is that my two year old LOVES to put herself in time-out….. only, not when she's done anything wrong, just in general. She knows that when she's done something bad and doesn't stop, Mommy brings a chair to the corner, and she has to sit there until time up is over. To be clear, she does not enjoy time-out. She cries and tries to wriggle her way out and eventually calms herself down and then it's over. She does not usually play when we put her into time out, and we never ever give her toys when she is in the time out chair. Nonetheless, lately, she loves pushing her chair to the time out corner an sitting in it when she is NOT in trouble. She will sit there and talk and sing to herself quite happily without any toys or books or reinforceement from us for an hour if we let her. I usually try to make her go do something else, but this is liable to trigger a genuine temper tantrum. She absolutely refuses to leave the time-out chair, and I'm not sure what to do.
But, back to your problem. I was very much like your son, and hearing you talk about him concerns me because I have struggled with anxiety (and depression to a lesser extent) since i was a child.. I had a hard time because I worried about a lot of things; I was not only concerned with doing the right thing, but I was also concerned about OTHERS doing the right thing as well….. which meant I tended to get into a lot of fights with teachers. For example, one day my kindergarten teacher had just had it with everything and decided to put everyone's name on the board. I was deeply upset by this because I did not believe I had done anything wrong and saw this as unjjust punishment. So, I went to my cubby, got my coat and my backpack and walked out of the building. Apparently, the police were out looking for me for the better part of two hours. I worried about everything as a kid: I worried about world hunger, I worried about people who used drugs, I worried about my grandparents dying (just because they were old, I guess). And I thought about things that most kids don't think about when they are young. My mother remembers me asking the pastor of our church why God lets people get into wars if he loves everyone when I was four years old. I was in elementary during the first war in Iraq, and I remember watching missiles being dropped on Iraq on the tv and crying in my room at night because, while we sang "God Bless the U.S.A" in music class, innocent people were dying. I was deeply troubled by a lot of things. Add to that, the fact that I was academically miles ahead of most of my peers, it's probably not a big surprise that I had a hard time making friends with kids my own age. As I got older, my loneliness and my concern with doing the "right" thing morphed into a sort of existential depression as I gradually came to realize that moral perfection was impossible to achieve. If we can not be "good", I wondered, then what use are people at all? Is goodness a made up construct simply designed to control society? Does God exist and does He care what we do at all? What makes something good, anyway? I began playing the piano at age three and competition piano at age five, so music was a big part of my life growing up. And, while I never had much of a gift for visual art, I often found myself deeply affected by it. I still wondered, what makes art good?. Does it have to do with technical skill? Does it have to do with correct expression? Does it have to do with profitability of whatever one produces? At one point, I was practicing the piano six hours a day outside school, and I was exhausted. In elementary, I had panic attacks if I missed a word on my spelling test. In 9th grade, I got a B+ in geometry and stopped caring about school entirely since my chances of getting a perfect high school GPA were already ruined. I dropped out of high school and back in again eleven times. In college, I faced real depression, and though I eventually made it through my darkest days and came to a better understanding of what makes life truly meaningful, I still struggle with anxiety and perfectionism even today. So, I don't know if your son is anything like me, but for me, the guilt and self flagellation was part of a much larger problem that has plagued me through my entire life to various extents, and it may be something you need to watch out for with your son.
As for your suggestions: #1 flatly would not have worked for me. At all. My sense of what needed to be done came from my own sense of practicality. I would clean the wash the dishes if there were no clean bowls to use. Being told to clean the dishes without being told WHY would have frustrated me. Being told to clean the dishes, "because I said so", would have made me downright angry and oppositional. #1 COULD have worked if someone had ASKED me to help out, say, because it makes life easier for Mom and Dad. If they had just explained that they work hard and are sometimes tired at the end of the day and that they would really appreciate it if I could help out with a few things around the house, I probably would felt a sense of responsiblity… no, more like obligation.. to help my parents because I loved them. My family was important to me. #2 is a good idea if you have the right goal in mind. First of all, spending time with your kid is never a bad thing. But, as you already know, going out of your way to boost a child's self-esteem can sometimes backfire. If you want to learn Spanish and he wants to learn Spanish (this is something he's expressed interest in, right?), then great, do it. But, don't lavish him with praise every time he does something well. Some kids have a pretty fine tuned bulls**t meter. Actually, I would say, make it as difficult for him as possible so that he gets frustrated, and then use it as an opportunity to learn frustration tolerance, patience, and coping skills. If you are going too fast, for example, teach him to ask questions and to advocate for himself. Show him how to break the problem down into smaller parts. Teach him persistence – If you make a mistake, try again. It's okay, etc. Be cautious with #3. Let him choose the activity, and focus on having fun. The absolute WORST thing you can do is to push the importance of mastery and technical skill. If your self-esteem is predicated on being able to do something at a very high skill level, you could be setting yourself up for failure down the road because there is almost certainly going to be someone who is better at ______ (eg; martial arts, etc.) out there somewhere. If you are one of the gifted few to truly excel in your chosen area, outside pressure and criticism can be intense, which may be really difficult for an already sensitive child to handle. Instead, give him as much security and unconditional love as you can (you probably already do this, but just keep on doing it). Help him find his passion. Teach him to pursue the things he loves for reasons OTHER than trying to be the best, never making mistakes, worrying about what other people think, etc. Try to figure out what he truly cares about, other than being "good". Does he care about his brother? Ask him to think of ways he can help his little brother. Maybe he can teach him a new skill. Does he care about animals? Maybe you can help him volunteer at an animal shelter. These are just a few examples. The point is, self-esteem comes from doing something meaningful and knowing that one has the power to make a difference, not just from doing things well.
You have no idea how much insight this has given me! As a parent, I remember the shyness, introverted personality I coped with. I do not remember anxiety until I was a bit older.
However, I have a child with signs of anxiety, and I had a brother with anxiety, (and I would say I do now).
What else can I do for her? What did you need to hear from your parents? What would have made it apparent that they understand, and love you no matter what, that life isn't about perfectionism, and that just finding joy is enough? She is very hard on herself, I find myself praising her a lot because she does a lot-crochet, art, violin, journaling, reading books beyond her level-all things she loves but brings to show me. How can I respond that doesn't make her feel like I'm BSing her, but gives her a real response?
From time to time over the past year, my now 5 year old son has been "telling on himself " for any and everything he does. Whether it be "I put my hand by my mouth, a little", or "I bumped my pee pee when I buckled my seatbelt". It is all day, every day. I dont know if it's a nervous thing or not. I try to remember if it happened before anything big is about to happen in his life. He just started kindergarten this month, and it had started up again maybe a month ago. I am one who gets easily annoyed, and feel awful that I feel that way. I just wish I knew what is causing it, or why he is doing it.
Hmm… "I put my hand by my mouth a little," or "I bumped my pee pee," seem odd things to confess. Are there any injunctions in your family about touching you mouth with your hands, for example? I know it seems strange to ask, but I'm wondering where he gets these particular ideas from. Do you think he's doing it for attention?
I'm no expert, to say the very, very least, but these seem like confessions of incredibly minor – pretty much non-existent- 'infractions' to me. Internalized stress, as you mention, could be a major factor, but who knows? We're going through some really rough times here right now – my wife has terminal cancer and we're taking a temporary (?) separation – and the other day our oldest lost his eyeglasses at the beach. I shouted at him (briefly), then discovered him a few minutes later in an upstairs bathroom hitting himself. Needless to say I felt really guilty about this, and had a long talk with him about it. It devastated me that he was abusing himself, and I told him that that was not ok. But you don't want to punish them for punishing themselves for being punished, if you see what I mean.
I wish I had an answer for you, Holly. The only thing I can say is that you wait a bit and see how things develop. He's only 5, so it may be something that he grows out of. We've considered the possibility of getting our oldest some counselling. He'd hate it – I mean, really, truly hate it – but it might help him in the long run. I wish I had more to offer, but we're pretty much flailing here too.
Anyway, thanks for reading, and for sharing your own thoughts and situation. Keep in touch to let me know how it goes.
In that example, it's pretty clear your child just manipulated you into allowing them to have a treat, despite not finishing their food. No offence, but that one seems obvious.
I was the same way, when I was a kid, still am now. But I quickly learned that trying that more than once was a great GREAT way to get people to start ignoring you, and allowing you to punish yourself.
Make sure he does it when nobody else is around, before youi make sure he's punishing himself, and not grabbing for sympathy first. I speak from my own personal experience, because I was that way as a child.
Yes, I have Asperger's, but eventually I spent some time with someone who wouldn't put up with that crap just because I was "special", and it changed my life.
I will wait and confront them when I am calm, I can be violent when necessary and I have no time for diplomacy. As a child I was not allowed to make mistakes but if I did I had to repeat whatever I had been doing until I could do it 5 or 6 times perfectly. Beware of putting your son into martial arts, he may begin to enjoy fighting with others and damaging them.
I find your post very interesting because I am now 25 and remember being as young as 4 and punishing myself by not allowing myself dessert, or I would break things I made/ tear things I wrote when they weren't exactly how I wanted them to be. My parents just laughed about it, particularly the not letting myself have dessert. My mum improved my habit of breaking things by letting me do it and showing me how much it hurt after when I was left without the thing and had to start again.
In my teens, when I got angry at myself for failing, I began a habit of hitting my head, and now when I'm in relationships and get very stressed during arguments (for example, I can get depressed and then lash out at the person closest to me and feel angry at myself for doing this and for being useless for not being able to control my mood), I sometimes hit my head or other parts of my body or punch the wall until my hands are bruised.
I'm not saying this to worry you but I feel that, for me personally, the behaviours are linked, since it is about me not being good enough and hating myself for this. When other people try to be nice to me it is almost worse and sometimes I say cruel things just so they will shout at me. Anyway, the idea that this is just a phase, as one person suggested, should be treated with caution. I'm in the process of seeking help but have not yet had the chance to be diagnosed with anything and don't know if I just struggle with the world and this is part of the effect of being ambitious and an over-achiever, or if I have a mental illness.
I think the only thing that my parents could have done (and they did try to, I think) was to teach me to love myself a bit more and maybe make me feel good just for being me, rather than that to be a good person I had to achieve something amazing, but that being average and just happy was also ok.
I also wish that they hadn't just seen me try to punish myself as some silly childhood thing. They don't know that I still self-harm and so laugh about it sometimes.
So, I guess my advice is to take it seriously (but don't over-worry). Don't be afraid to talk to your kid about it and be aware that such behaviours might develop into other negative habits when he is older and has to deal with more stress and so to create an environment in which your children feel able to talk to you about these things – be that sexuality to mental illness. But, as a parent, you can't fix everything. Some of these things we're born with and so much of our environment is outside of the home (school can be a breeding ground for negative coping mechanisms!).
Wow, balooinblue, thanks for sharing that with us. Yours is a really interesting story – I don't even know where to start. It's disheartening to hear that this has continued into adulthood for you, and I hope that you get some help if you need it. I think it's difficult for everyone – but particularly for perfectionists – to ask for help, and that's something we all struggle with. Thanks so much for your advice and for leaving your thoughts. I really appreciate it. Best of luck, and please report back if you find any coping mechanisms that you could share. Cheers!
Sounds to me like your son has a very rigid and dogmatic interpretation of your generally good intention that it is not healthy to receive reward without doing something first.
I do not know how many three year olds can read between a parent's lines, but I would not be surprised if it were less than 100%. So, you may need to explain gently and constructively that you do not literally mean everything you say.
One wacky possibility is that your son lies somewhere on the autistic spectrum, most likely Aspergers. There was a book written 50 odd years ago by parents of a more severely autistic child who actually responded really well to being spanked hard. The parents were terrified the first time they did it, but found that it stopped the yelling in ways that silence or reasoning never did. And their daughter started opening up much more. So maybe your wacky idea might have legs, in moderation, if your son were Aspergers?
Thanks for reading and leaving your thoughts. My son still has this tendency, but it's certainly not as strong as it once was. I think he's learning to ease up on himself. I'm pretty sure he's not on the autistic spectrum, and I definitely have no intention of spanking him. Ever. I appreciate you taking the time to leave your ideas, though!
Hi Mmtread, I was just reading this post and wondering how your son is going now?
my three year old daughter puts herself in timeout to avoid having to pick up her toys. We ask her to pick them up, try and make it a game but nothing works. She hates picking up her toys, knows that it will eventually end up in time out, so heads us off by putting herself in timeout. HELP!
My daughter is very similar to how you describe. I found the book "Mindset" by Carol Dweck to be very helpful in reinforcing a growth mindset rather than the fixed mindset that my daughter naturally gravitates towards. I highly recommend the book. It has a lot of great advice, such as avoiding telling her she is "smart" and instead applauding her efforts. We also try to emphasize that failure/mistakes are an important part of learning, and ask her every day to try to think of a mistake she made that allowed her to learn something. It's an ongoing process for all of us, but the ideas in the book have been tremendously helpful.
My son has been doing this for a couple of years. He started at 4 and in the beginning he would say terrible things like he wanted to die and even went to the extent of wrapping a sweatshirt around his neck and pulling tight. I was am very scared of what, why and how this all came to be I have learned that he would ride the bus home with his sister and some of the kids picked on him and threatened to throw him off the bus. I talked to the driver and stopped any further bullying there. And then we had a boy at the babysitters that was being mean and picking on him. I put a stop to that. He now only punished himself and there is not much talk of harming himself. But his behavior now is to throw his favorite in the garbage. I have tried so many things and still trying. I am going to go see a child psychiatrist now. I am at my wits end.
Someone he can talk openly with is probably a good thing if you've run out of options and ideas on your own. It's always a good idea to get help if you're at a loss. Sounds like he's headed in the right direction, though. I suppose the best thing to do is give him loads of love and support, from as many sources as possible.
Best of luck to you and your boy. I'm sure everything will turn out fine. It almost always does.
1) She needs more one on one attention. Thankfully she is outwardly expressing it … which is most likely a cry for help/attention… My husband and I are taking turns giving her at least 10 minutes a day of uninterrupted time… meaning no phones, no laptop.. (this goes for us as well as her) nada. 2) We are teaching and explaining the benefits and need to be "impeccable with your word"… If you have ever read The Four Agreements, I believe it is agreement number 2… this is a quick cliff notes – ish page to the agreement.
So far it is working… we had a little breakthrough tonight actually where she even talked about why she was feeling the way she was 🙂 Best Luck!!!
Hi, thanks for visiting and leaving your own experiences. It's great to see that you have a plan, and that it seems to be working! Spending time with the kids definitely helps – during the summer I'm with them all day long, and they clearly appreciate it. Our oldest is getting better about dealing with his perfectionism and allowing himself to make mistakes, but it's a process.
I'd never heard of the Four Agreements – interesting stuff. I'll have to give it a more careful read a bit later. Anyway, best of luck to you too, and thanks again for visiting!
Thank you for writing about this. I see a lot of the same thing in my oldest, too. Definitely explore the perfectionist angle. I think your action steps to build his self esteem is spot on, too. And it sounds like you're appropriately modeling how to deal with your own mistakes, too, from what I've read in this post. Thanks for sharing. I look forward to hearing more of the journey!
Thanks, Meghan. Our boy seems to be getting more adept at handling failure and frustration, and I think just spending a lot of time together helps. Best of luck with your oldest!
Perhaps I can say that I understand your son's behaviour, more because I see it as a reflection of my own old self growing up. Now a dad of a beautiful 1 year old baby girl, I promised that I'll let her be the way she wants to be, without ever undermining or restricting any of her choices. Not that I wasn't lucky in my age with any of these, but may be I was late in introspection. I also realised that I was a natural in camouflaging this to avoid parent detection. But you already know and that might be the biggest help your son has received, already. I think all that is required here is some good buddy talks (better than dad-son talks), drives, camping. Anything that he likes doing in particular? Well that's what he needs to do. And nothing beats when you are with him while he does it.
I too have a similar issue with my son, now nine years old. Before his two sisters were born it wasn't much of an issue. Now that he has two little sisters he seems to blow everything he does wrong out of proportion. I really beat myself up at the end of the day and wonder if i am too rough on my son when he does do wrong.
We also signed up my son for soccer this upcoming fall and hope that maybe he just needs a place to express himself a little more.
Please update us on how the steps you have taken work out.
This is a tough one. Perhaps participating in an activity where mistakes can be made and unmade – I was thinking pottery might be useful, because you get multiple attempts to make it just the way you want, and each attempt is a learning experience. I know I was very sensitive as a child, and would break down in tears if I even thought I had disappointed my parents, and then people would ask me why I was crying and then I would cry even harder because they said I was sensitive and I shouldn't cry.
I have a 2.5 year old son. I have found that bed time is the most calm and serene time for us. We have long chats before he drifts off to sleep. So I kind of blend the message I want to convey into a story, night after night. (So he knows what "Bruno" did when he was in the mall etc) And then one day I find him reminding himself of the moral of the story when opportunity comes..
I think you must try weaving a story when you have a peaceful one on one time and reiterate that it is ok to fail and that all great people have failed more than once..
I don't know how I missed this post until now. I'm sorry to hear you're have a difficult time with your son. Seems that you're already moving in the right direction with him, to help in through this stage of his development in the best way possible. As parents, most of us blame ourselves. Of course we must have done something to screw up our kid at some point without realizing it, resulting in the current challenge at hand.
In my own experience with our son (almost 8 yrs old), we've struggled since he was about 3. He's extremely willful and inflexible. Everything is a HUGE deal at certain moments and my husband and I sort of just blink at each other as if to say What the HELL is happening here? Why is he freaking out? After having him evaluated professionally just to make sure we weren't missing something important, we came up with this: This is his personality. I do think it's important to consider that much of a child's behavior is hard-wired.
We found the book – Transforming the Difficult Child: The Nurtured Heart Approach by Howard Glasser and Jennifer Easley, very helpful when we adhere to the recommendations. It's a decent amount of reading but well worth it.
Also, do you find that he's behaving this way in school? I often found myself hoping that my son's teacher would say that he had a meltdown in class and then guide me on how to respond, but in a school setting and amongst his friends he's as good as you'd hope your kid would be. Sometimes I think kids save certain aspects of their personalities for their parents, which is good bc it means they know how to – and more importantly, how not to – behave. Maybe you'd like to see what his teachers have to say about it?
Best of luck. It's not always easy. Either way, all your effort to help him along reinforces how much you love him.
Hello MM. This is intimate stuff and a great conversation you have opened up here.
Could D be punishing himself to prevent you from doing so? It's evident from reading you for a while that you have a relaxed, loving parental style and perhaps D values this so much he doesn't want to have to face the spectre of disapproval from you…or indeed, anyone. He may be one of the legion who don't like conflict or disapproval.
I was thinking about an incident in my childhood just a few days ago where I made myself more upset than I was after I accidentally damaged something so that my much-loved mother would not scold me (I realised when I was older that of course she wowuldn't have anyway). It is a slightly different situation but I throw it into the mix here in case it has some bearing. I like what Amy said about how he is probably much more forgiving of others than of himself. Would it be worth asking him, when he wants to punish himself, how he would respond if a good friend had done what he'd done? Or how that friend would respond.
Hmm, that's an interesting idea, Charu. I may have to give it a try if all else fails. Thanks for sharing!
I think this is a very interesting blog. A most read!
I really like your blog. Thanks for posting; will keep reading your blogs.
I suspect that someone outside of your immediate unit is or has given him cause to have low self esteem. I think he is telling you by way of third person when speaking out loud to himself or pointing it out to you. Eg soup.
None of my kids exhibit this behavior – although both my boys frequently bopped themselves on the head for no apparent reason – but I grew up pretty much doing just as your son does. I felt an internal need to be perfect at all times. That didn't seem to come from anything my parents did, or didn't do, it came from inside. I was the best at punishing myself for what I saw as failure. I was terrified of failure and the embarrassment I thought it would bring. Heaven forbid anyone figure out I wasn't perfect. Teaching him, though the activities you have planned, it's okay to fail, it's okay to be wrong, it's okay to not be perfect, is a great route to take. I think you handled the dinner and treat situation perfectly with that in mind.
Thanks, rgemom! Our son is, indeed, absolutely terrified of failure, which makes him avoid activities that are new or difficult. I guess we'll just keep trying to teach him that we all fail at times, and he's no different. Thanks for reading the post and sharing your thoughts!
I reckon it is a bit of a phase he is going through and certainly nothing to worry about. I have seen similar behaviour with two of my daughters at various times at a similar age. As a secondary (high school) teacher l do come across this behaviour in older kids – but they tend to be on the autistic spectrum. Your son aint autistic,Mat.At their age, understanding where they are coming from is bloody hard!
Autism has never even been a consideration, cravenology, so no worries there. It's not that I'm so concerned about this, it's just that I'd like to understand it better and do what I can to ease his obvious distress.
I would consider it a phase if he hadn't always been like this. Pensive and purgatorial. Wound tightly and tense. I suppose it's just his personality, but it just seems a boy of eight should be more easygoing and light-of-heart. But really, he's a great kid, and my concerns are probably unnecessarily exaggerated.
Thanks for taking the time to give it a read and share your thoughts – I appreciate it!
I just love the pictures of him, especially the one by the water! My father the child psychologist would probably say he has an "overdeveloped Id", (I think it's the Id that makes us feel like we don't deserve good things etc, may be the ego?) anyway, it does sound like you're doing the right things – are there any good counselors you could talk to who may have some more concrete advice? It's hard when we see our kids being so hard on themselves – I am sure he is much more forgiving of others and maybe if you continue to also point that out to him. You seem like a great Dad and I'm very impressed with your willingness so share and seek advice!
Hey Amy! Thanks for appreciating the pictures – it was late at night and I didn't spend too much time choosing them, but I went for ones that more or less showed him as a brooding, introspective loner. Which, I suppose, he largely is.
I hate to see D being hard on himself (that's my job and it steals my thunder), but I don't think we're quite ready for professional help or Prozac. The fact is, he's a sweet, sensitive, intelligent boy who just happens to beat himself up more than is strictly necessary. I feel he'll grow out of it. If not, I can probably just beat it out of him. A great dad? I aspire to be one. Am I succeeding? Only time will tell. In the meantime, I'll just keep plugging along, groping my way through this whole parenting thing, and teaching my kids anything and everything I know about the world. Which isn't much. But it's a start.
My son is 2 and is already starting to put himself in timeout if he sees Im unhappy about something that has happened, even if I don't chastise him.. I may have a perfectionist on my hands or a child who needs to make mom proud at all times…I am a big believer in a balance between nature&nurture and I think personality is mostly nature…I will just continue to let him know that time out is not always necessary and that I will always be proud of him & love him…it will be easier for me (I hope) once he becomes more verbal. We will see!
Two seems pretty early to be putting himself in timeout, but no need to worry, I should think. Worrying too much about your kids appears to be a common illness these days, but I think your attitude is perfect (for what it's worth), and from what you've written you sound a level-headed and sensible mom.
Thanks for the read and for leaving your thoughts, Ashly. Two is a great age, but it only gets better!
Loved the article! I will have to wait until my kids are a little older but as of my current situation my 2-year old and 10-m old are very different personalities. My oldest having always been mild tempered, never crying or fussing and my youngest is ready (at a moments notice) to scream&cry…my oldest will ask "What's wrong, sissy?" But if she keeps on he simply leaves the room without any fuss. Lol. I love them both but I can already see I have polar opposites on my hands. Please keep us informed on your journey with your boy..I'm very interested as it looks like I will be dealing with a similar situation!
Is like my children to show some remorse. They are on the other end of the spectrum.
Well, Tiefsa, there's nothing like a good whooping to put kids in a remorseful frame of mind. I'm kidding of course. No, really.
It's hard when kids are at the red or violet ends of the spectrum. Hey, I'm all for rainbows, but I always aim for yellow, green, or blue. It's all about balance in all things, I think.
Yeah. We had to have a talk with one about how we love them unconditionally, but that doesn't mean we love what you do or say.
Thanks for sharing this piece of your life. Our 7 year old will berate himself over his writing. His handwriting is poor at best, and we've been working on it with him. Backwards letters, no spacing or punctuation, and sloppy as the day is long. We gently direct him as he is writing. But when he realizes his own mistakes he'll hit himself on the head a few times, growl about, gets visibly frustrated, angry or upset. He's currently learning to ice skate and he's the same way. Every time he falls he growls and pounds the ice like it was the ice's fault he fell. Though I guess that's a bit different because he's taking it out on the ice instead of himself. The only thing I can think to do is keep gently reminding him how to do it. Have patience with him. Whenever I read the word "Perfectionist" I read OCD. I have taught many a student who were OCD and if things weren't exactly so they would be upset. Some would get angry and yell. Some would sit and cry. Some would scold themselves. As their teacher, the only thing I could do would be talk to them and try to help them fix the "problem" or at least calm them down and talk them through fixing it. How does D react when he gets it right? Does he get as excited about doing "perfect" as he gets down about not? (I'm guessing not because "perfect" is his standard so in his mind he's simply doing what he's supposed to be doing… but I'm speculating here). Our 7 year old won't punish himself when he actually does do something wrong, but he will try to punish his 3 year old brother. I have to remind him that he's not the dad, I'm the dad, he's not allowed to punish. Sorry for the long comment…. I can relate to your dilemma and I hope you find your way through it together successfully.
Hey, Stomperdad. Well, first of all, let me thank you for leaving such thorough and thoughtful comments. Secondly, what the hell is this whole hitting yourself thing? Both of our kids do it – when he was a toddler, our youngest would bang his head on the floor or furniture when he was angry. It was both harrowing and hilarious.
I don't imagine D is OCD – he doesn't exhibit any of the signs, really, but he does get upset when he tries and fails. D is very blase about getting things right – as you say, he's only doing what he feels he's supposed to be doing. That is an excellent point I'd not considered. D also bosses his younger brother around – not exactly trying to punish him, but acting as a proxy for what he thinks I would say or do in a given situation.
No doubt we'll be fine. It's not really a big problem, just something I find very curious. I'd like to see him lighten up a bit.
That's excellent advice, Greti. We're a pretty laid back household – I'm a fairly easygoing person as I'm sure you remember, but also prone to getting really stressed out, as I'm sure you also remember. 😉 You make an excellent point that he puts enough pressure on himself. Thanks for the reminder!
Feel your frustration! Matthew (11) goes in spurts with this sort of behavior. Puts me in a RAGE! His "go to" behavior is calling himself stupid. He has, in extream bouts if frustration, been self injurious! Clawing his face and punching his head! I can usually stay calm an discuss rationally…other times not so much! I think for Matt its anxiety!
Everything you've said in order to address this sounds great. I think you're doing all the right things. I have 5 year-old twins. The girl is fun-loving and takes everything in her stride. Her twin brother is much more intense and can be hard on himself… I think he had two 'accidents' while potty-training and he cried with disappointment. I don't know where it came from. Whenever they break something (even something expensive). I stipulate it was accident, let it go. It feels quite unusual to raise a child with a personality type different to your own. My son is more like my husband. I just try to respect what makes him tick and what's important to him.
It sounds like your little boy is just unsure about whether he deserves these things right now and I think its great that he can be so open with you.
Hugs are standard issue in this family, but while he'll let himself he hugged, he's not much of a hugger-backer. Anyway, hopefully we can boost his confidence a bit and get him to lighten up. He's the most serious-minded kid I think I've ever met.
Thanks for reading and taking the time to comment, Olivia!
I think your strategies sound perfect, particularly sitting down with him to learn Spanish if this something he's going to enjoy. A bit of reassurance and quality parent time can work wonders.
Thanx for sharing. To me, he is a perfectionist and although this might seem good, sometimes it is not; as he has to learn that it is good the way it is. Sorry for my English. I am from Madrid, Spain. And by the way, if you need help when learning spanish, here I am.
I think you're absolutely right – he's a complete perfectionist. We tell him all the time not to let perfect be the enemy of good, but he gets upset when something isn't perfect.
By they way, we live in Sitges, and please do not apologize for your English – It's far better than my Spanish. But I'm working on it!
Thank you for sharing. I have no answers or advice, only a sense of gratitude that you are willing to share. This is not easy. Different kids under the same parenting develop different habitual behaviour patterns. Becoming an adult human is a messy process.
I used to punish myself as a child… But I wouldn't tell anyone what I did wrong….
My daughter sounds a lot like your son. She is constantly seeking approval and berates herself if she doesn't feel she has met "expectations." I am trying to start a meditation practice with her to help her detach from these negative emotions.
Meditation. Now that's a thought. I'll have to look into it. Thanks for sharing, Anita.
Sounds to me like he might be a perfectionist. Not just in terms of doing tasks the way they are in his head (drawing, building etc) but also in how he views his actions. As if he should be perfect all the time. Maybe do some reading about the perfectionist child amongst the other resources you're tapping?
Oh man, FMNZ, he's a complete perfectionist. Homework time is fraught – he has to do everything perfectly and is reduced to tears if he makes a mistake. I can't believe I didn't investigate this line. But that's what I'm going to do right now. Thanks for the suggestion (and the comments)!
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 6,914 |
[
128000,
40,
617,
505,
892,
311,
892,
11,
358,
10585,
499,
539,
11,
21699,
311,
32151,
1057,
24417,
4538,
369,
75986,
5678,
627,
3112,
1618,
11,
1314,
520,
279,
1212,
11,
358,
617,
311,
48466,
2555,
13,
358,
1541,
956,
1440,
1148,
358,
2846,
7556,
922,
627,
77641,
994,
358,
3350,
2555,
311,
4430,
449,
499,
1695,
13016,
11,
358,
1093,
311,
2586,
505,
264,
2361,
315,
1063,
6677,
11,
4443,
3217,
11,
323,
8830,
13,
2030,
358,
5042,
1541,
956,
58389,
1405,
856,
4538,
596,
7865,
4131,
505,
11,
323,
358,
2846,
4477,
420,
304,
279,
3987,
430,
555,
25394,
433,
1618,
320,
438,
23127,
95360,
5977,
701,
6170,
8,
358,
649,
64855,
279,
97892,
21160,
315,
420,
1389,
311,
757,
482,
3646,
9662,
73555,
481,
25885
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
40,
617,
505,
892,
311,
892,
11,
358,
10585,
499,
539,
11,
21699,
311,
32151,
1057,
24417,
4538,
369,
75986,
5678,
627,
3112,
1618,
11,
1314,
520,
279,
1212,
11,
358,
617,
311,
48466,
2555,
13,
358,
1541,
956,
1440,
1148,
358,
2846,
7556,
922,
627,
77641,
994,
358,
3350,
2555,
311,
4430,
449,
499,
1695,
13016,
11,
358,
1093,
311,
2586,
505,
264,
2361,
315,
1063,
6677,
11,
4443,
3217,
11,
323,
8830,
13,
2030,
358,
5042,
1541,
956,
58389,
1405,
856,
4538,
596,
7865,
4131,
505,
11,
323,
358,
2846,
4477,
420,
304,
279,
3987,
430,
555,
25394,
433,
1618,
320,
438,
23127,
95360,
5977,
701,
6170,
8,
358,
649,
64855,
279,
97892,
21160,
315,
420,
1389,
311,
757,
482,
3646,
9662,
73555,
481,
25885,
-100
] |
Inside 'The Bank Job'
by Gerri Miller
Daniel Mays (left) and Jason Statham in "The Bank Job"
Jack English/Lionsgate
In 1971, a group of thieves broke into the safe deposit boxes at Lloyds Bank on Baker Street in London and stole more than 500,000 British pounds in cash and jewelry. Four of the thieves were caught, but the rest disappeared with the loot. The audacious heist -- and the accompanying scandal involving murder, corruption, blackmail and royal-family sex photos -- immediately captured the fascination of Great Britain. But three days later, all news coverage stopped because of a government gag order. It's obviously a perfect setup for a movie, but how do you create a script when there's so little information out there -- and so many of the major players are still on the lam?
Writers Dick Clement and Ian La Frenais ("The Commitments," "Flushed Away," "Across the Universe") found a way. They did their best to fill in the blanks, using secondhand accounts, media coverage -- what there was of it -- and transcripts of the thieves' walkie-talkie conversations to put together a script based as closely as possible on fact.
The result is "The Bank Job," directed by Roger Donaldson ("No Way Out," "Thirteen Days," "The Word's Fastest Indian") and starring Jason Statham ("The Italian Job," "Crank," "Transporter"). In this article, we'll sort fact from fiction, explain how the heist was planned and executed, and outline the challenges the production faced in shooting and re-creating the look of the '70s.
Video Gallery: Robberies
Thieves stole $28 million worth of diamonds from the ABN Amro Bank in the heart of the Diamond Quarter in Antwerp, Belgium. Antwerp is known as Europe's diamond capital because more than half of the world's diamonds pass through this district.
A Kentucky man entered a liquor store with his head covered in duct tape and unsuccessfully attempted to rob it. Learn about how robberies (sometimes don't) work in this news video from Reuters.
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 5,660 |
[
128000,
25488,
364,
791,
8715,
12280,
1270,
1729,
20524,
462,
17472,
198,
41686,
386,
954,
320,
2414,
8,
323,
18984,
800,
69476,
304,
330,
791,
8715,
12280,
702,
33731,
6498,
7586,
919,
25623,
198,
644,
220,
4468,
16,
11,
264,
1912,
315,
61667,
14760,
1139,
279,
6220,
16946,
15039,
520,
445,
2032,
5469,
8715,
389,
29492,
6825,
304,
7295,
323,
40606,
810,
1109,
220,
2636,
11,
931,
8013,
16701,
304,
8515,
323,
31817,
13,
13625,
315,
279,
61667,
1051,
10791,
11,
719,
279,
2800,
29496,
449,
279,
46403,
13,
578,
6264,
19995,
568,
380,
1198,
323,
279,
24442,
26681,
16239,
10102,
11,
21948,
11,
88263,
323,
30336,
17633,
1877,
7397,
1198,
7214,
17439,
279,
75444,
315,
8681,
13527,
13,
2030,
2380,
2919,
3010,
11,
682,
3754,
10401,
10717
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
25488,
364,
791,
8715,
12280,
1270,
1729,
20524,
462,
17472,
198,
41686,
386,
954,
320,
2414,
8,
323,
18984,
800,
69476,
304,
330,
791,
8715,
12280,
702,
33731,
6498,
7586,
919,
25623,
198,
644,
220,
4468,
16,
11,
264,
1912,
315,
61667,
14760,
1139,
279,
6220,
16946,
15039,
520,
445,
2032,
5469,
8715,
389,
29492,
6825,
304,
7295,
323,
40606,
810,
1109,
220,
2636,
11,
931,
8013,
16701,
304,
8515,
323,
31817,
13,
13625,
315,
279,
61667,
1051,
10791,
11,
719,
279,
2800,
29496,
449,
279,
46403,
13,
578,
6264,
19995,
568,
380,
1198,
323,
279,
24442,
26681,
16239,
10102,
11,
21948,
11,
88263,
323,
30336,
17633,
1877,
7397,
1198,
7214,
17439,
279,
75444,
315,
8681,
13527,
13,
2030,
2380,
2919,
3010,
11,
682,
3754,
10401,
10717,
-100
] |
Endoca's primary focus is to research and develop innovative hemp extracts and hemp based foods that can help to balance the human Endocannabinoid system making. Endoca believes that through targeting the human Endocannabinoid system with CBD Oil products they can help many people on this planet to restore their internal balance.
Endoca is dedicated to producing the finest and purest quality CBD hemp oil products while never compromising on quality. Endoca also produces other cannabinoids as there are over 80 different in nature.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 439 |
[
128000,
3812,
17270,
596,
6156,
5357,
374,
311,
3495,
323,
2274,
18699,
42063,
49062,
323,
42063,
3196,
15657,
430,
649,
1520,
311,
8335,
279,
3823,
4060,
511,
97853,
590,
1887,
3339,
13,
4060,
17270,
13919,
430,
1555,
25103,
279,
3823,
4060,
511,
97853,
590,
1887,
449,
22834,
15895,
3956,
814,
649,
1520,
1690,
1274,
389,
420,
11841,
311,
15301,
872,
5419,
8335,
627,
3812,
17270,
374,
12514,
311,
17843,
279,
28807,
323,
10748,
267,
4367,
22834,
42063,
5707,
3956,
1418,
2646,
76100,
389,
4367,
13,
4060,
17270,
1101,
19159,
1023,
93986,
439,
1070,
527,
927,
220,
1490,
2204,
304,
7138,
13,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
] |
[
3812,
17270,
596,
6156,
5357,
374,
311,
3495,
323,
2274,
18699,
42063,
49062,
323,
42063,
3196,
15657,
430,
649,
1520,
311,
8335,
279,
3823,
4060,
511,
97853,
590,
1887,
3339,
13,
4060,
17270,
13919,
430,
1555,
25103,
279,
3823,
4060,
511,
97853,
590,
1887,
449,
22834,
15895,
3956,
814,
649,
1520,
1690,
1274,
389,
420,
11841,
311,
15301,
872,
5419,
8335,
627,
3812,
17270,
374,
12514,
311,
17843,
279,
28807,
323,
10748,
267,
4367,
22834,
42063,
5707,
3956,
1418,
2646,
76100,
389,
4367,
13,
4060,
17270,
1101,
19159,
1023,
93986,
439,
1070,
527,
927,
220,
1490,
2204,
304,
7138,
13,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100
] |
Our Natural Shower Gel will leave your skin feeling soft and delicately fragranced with a tropical fruity aroma. We have added wheat protein, which will condition and hydrate your skin and also act as a natural barrier against the elements.
ARULA Shower Gel product is formulated with over 98% natural ingredients.
Our product is very beneficial for customers who are experiencing dry and sensitive skin conditions, including eczema. This is because our shower gel is: sulphate-free; Paraben-free; Silicone-free and formulated with non-toxic or 'greener' ingredients.
Furthermore, this product contain no synthetic dyes and synthetic preservatives. In addition, it will not cause any burning sensation to your eyes like other high street brands.
Arula Shower Gel will leave our customers skin feeling soft and delicately fragranced with a tropical fruity aroma. Furthermore, your skin will be conditioned and feel hydrated as a result of added wheat protein.
PEG-120 (Methyl Glucose Trioleate is a polyethylene glycol) is a corn derived chemical used to thickened the formulation. This chemical is considered to be a low hazard chemical and as a result, a safe ingredient in our product.
PEG-120 is highly adept at reducing the irritancy of surfactant systems, and is also ideal for formulating cleansing products for babies, children, the elderly, and those with sensitive skin. It provides luxurious sensory properties through the formation of a rich, creamy lather and a light, emollient after-feel.
This chemical is widely used in the formulation of baby shampoos.
Most of our customers prefer their shower gel to have a fragrance. Therefore, our shower gel is delicately fragranced with a mild, fruity and jasmine aroma. The fragrance in our product is considered to be synthetic. Hence, it contain minute quantities of allergens which are; Benzyl Alcohol (0.00004%); d-Limonene (0.00012%); and Linalol (0.000008%).
According to the HSE directive, the presence of these allergens must be indicated in the list of ingredients and also on the label when their concentration exceeds 0.01% in rinse-off products. Consequently, due to the minute quantities present in our product, it is not required for these allergens to be listed on the ingredient label.
For delivery details, please click the link below.
this is a lovely shower gel that feels smooth to my skin. It rehydrates and keep it soft for a long time. I will definitely recommend this.
Thick and rich shower gel that leaves skin moisturised longer than any other product I've used. Pleasantly fragranced too!
Lovely product, I bought it at a conference in the midlands. I wish I had bought more, because my skin feels smooth after use.
The composition is really good and safe and it has a really nice smell without being too present.
I think the price is quite good for the quality and size of the bottle.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 8,689 |
[
128000,
8140,
18955,
48471,
45482,
690,
5387,
701,
6930,
8430,
8579,
323,
61353,
2718,
8508,
81,
4979,
449,
264,
35148,
93032,
55279,
13,
1226,
617,
3779,
34153,
13128,
11,
902,
690,
3044,
323,
45705,
349,
701,
6930,
323,
1101,
1180,
439,
264,
5933,
22881,
2403,
279,
5540,
627,
946,
59169,
48471,
45482,
2027,
374,
60394,
449,
927,
220,
3264,
4,
5933,
14293,
627,
8140,
2027,
374,
1633,
24629,
369,
6444,
889,
527,
25051,
9235,
323,
16614,
6930,
4787,
11,
2737,
12208,
89,
9355,
13,
1115,
374,
1606,
1057,
17639,
18316,
374,
25,
83778,
349,
12862,
26,
4366,
41408,
12862,
26,
99900,
12862,
323,
60394,
449,
2536,
4791,
14668,
477,
364,
15893,
804,
6,
14293,
627,
57417,
11,
420,
2027,
6782,
912,
28367,
294,
9891,
323,
28367,
1685,
79629
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
8140,
18955,
48471,
45482,
690,
5387,
701,
6930,
8430,
8579,
323,
61353,
2718,
8508,
81,
4979,
449,
264,
35148,
93032,
55279,
13,
1226,
617,
3779,
34153,
13128,
11,
902,
690,
3044,
323,
45705,
349,
701,
6930,
323,
1101,
1180,
439,
264,
5933,
22881,
2403,
279,
5540,
627,
946,
59169,
48471,
45482,
2027,
374,
60394,
449,
927,
220,
3264,
4,
5933,
14293,
627,
8140,
2027,
374,
1633,
24629,
369,
6444,
889,
527,
25051,
9235,
323,
16614,
6930,
4787,
11,
2737,
12208,
89,
9355,
13,
1115,
374,
1606,
1057,
17639,
18316,
374,
25,
83778,
349,
12862,
26,
4366,
41408,
12862,
26,
99900,
12862,
323,
60394,
449,
2536,
4791,
14668,
477,
364,
15893,
804,
6,
14293,
627,
57417,
11,
420,
2027,
6782,
912,
28367,
294,
9891,
323,
28367,
1685,
79629,
-100
] |
Fifty years ago next Sunday, a 32-year-old engineer called Theodore Maiman switched on a gadget at Hughes Research Laboratories in California, and watched as pulses of light sprang from a pink ruby crystal.
When Maiman's great invention was announced, the response was essentially "Doh?" as people tried to figure out what it was and what use could be made of it.
That was swiftly followed by an "Eek!" when the press came up with some scary headlines.
"LA Man Discovers Science-Fiction Death Ray," said one, reflecting the zeitgeist of 1960, when the Cold War mixed promiscuously with B-movies about aliens.
It can be found in things as everyday as supermarket bar-code scanners – the first scanned object was a packet of Wrigley's chewing gum in 1974 – just as it can be found in hi-tech self-targeting bombs, sniper's sights, adaptive optics in astronomical telescopes and research into nuclear fusion, the ultimate in clean energy.
And if you are reading this story online, think of the lasers that got it to you – more than million lasers power the Internet, shuttling terabytes of data through optical fibre.
"The story of the laser is incredible," Tim Holt, head of the Institute of Photonics at the University of Strathclyde, Scotland, said in an interview.
It was 1953 before the US physicist Charles Townes put the phenomenon to the test, with a "maser" – Microwave Amplification by Stimulated Emission of Radiation – in which microwaves were used as the atom-exciter.
Townes and a colleague, Arthur Schawlow, then had the idea of using visible light rather than microwave, although it was Maiman who made the concept work. "Light" replaced "Microwave" in the acronym, and the word Laser entered the vocabulary.
Today, the top end of research is "femtosecond" lasers, in which ultra-fast lasers alter the "spin" of electrons in individual atoms to provide a more compact, denser storage on hard drives.
"Lasers have given us a step in capability that is truly mind-boggling," said David Hanna, a professor in opto-electronics at the University of Southampton, England.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 1,446 |
[
128000,
37,
39724,
1667,
4227,
1828,
7418,
11,
264,
220,
843,
4771,
6418,
24490,
2663,
77449,
386,
2706,
276,
30975,
389,
264,
59951,
520,
41411,
8483,
78717,
304,
7188,
11,
323,
15746,
439,
66557,
315,
3177,
8314,
526,
505,
264,
18718,
46307,
26110,
627,
4599,
386,
2706,
276,
596,
2294,
28229,
574,
7376,
11,
279,
2077,
574,
16168,
330,
35,
2319,
7673,
439,
1274,
6818,
311,
7216,
704,
1148,
433,
574,
323,
1148,
1005,
1436,
387,
1903,
315,
433,
627,
4897,
574,
56651,
8272,
555,
459,
330,
36,
1247,
9135,
994,
279,
3577,
3782,
709,
449,
1063,
29565,
31186,
627,
1,
18326,
2418,
11997,
9146,
10170,
7424,
2538,
16290,
13558,
1359,
1071,
832,
11,
42852,
279,
78258,
79500,
315,
220,
5162,
15,
11,
994,
279,
24062,
5111,
9709
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
37,
39724,
1667,
4227,
1828,
7418,
11,
264,
220,
843,
4771,
6418,
24490,
2663,
77449,
386,
2706,
276,
30975,
389,
264,
59951,
520,
41411,
8483,
78717,
304,
7188,
11,
323,
15746,
439,
66557,
315,
3177,
8314,
526,
505,
264,
18718,
46307,
26110,
627,
4599,
386,
2706,
276,
596,
2294,
28229,
574,
7376,
11,
279,
2077,
574,
16168,
330,
35,
2319,
7673,
439,
1274,
6818,
311,
7216,
704,
1148,
433,
574,
323,
1148,
1005,
1436,
387,
1903,
315,
433,
627,
4897,
574,
56651,
8272,
555,
459,
330,
36,
1247,
9135,
994,
279,
3577,
3782,
709,
449,
1063,
29565,
31186,
627,
1,
18326,
2418,
11997,
9146,
10170,
7424,
2538,
16290,
13558,
1359,
1071,
832,
11,
42852,
279,
78258,
79500,
315,
220,
5162,
15,
11,
994,
279,
24062,
5111,
9709,
-100
] |
The event is hosted on the second floor of the Smith Memorial Student Union on the Portland State University (PSU) campus at 1825 SW Broadway, Portland, OR 97201 [map].
We recommend using Portland's public transit system, TriMet, as the quickest and pain-free option. Portland's light rail runs from the PDX Airport to downtown Portland. MAX trains depart from the airport every 15 minutes on weekdays. Weekend trains run less frequently; every 20-25 minutes. Trains depart the airport from 5am to 11:45pm. Passes can be purchased at the vending machines inside the airport, or through the TriMet Tickets App.
----->To Downtown: Take the RED line MAX light rail train (toward Portland City Center & Beaverton TC).
----->To PSU: Once downtown, get off at the Mall/SW 5th Ave MAX Station. Transfer to the GREEN line MAX at Pioneer Courthouse Square/SW 5th Ave (toward City Center/PSU). Get off at PSU/5th & Mill St.
Learn more about hourly and visitor parking around the PSU campus here.
MAX Light Rail – MAX trains run about every 15 minutes most of the day, every day. Service is less frequent in the early morning, midday, and evening. $2.50 per single trip or $5.00 for the whole day.
Portland Street Car – The streetcar runs approximately every 15 minutes on weekdays and Saturdays and every 20 minutes on Sunday. Arrival times are posted on real-time electronic signs at most streetcar stops. $2.00 per single trip.
TriMet Bus – The TriMet bus, system which covers the city and its suburbs thoroughly, offers low fares, friendly drivers, and full wheelchair accessibility. $2.50 per single trip or $5.00 for the whole day.
Biketown – Portland's bike-share system. For a small fee (an average of $2.50 for 30 minutes of use), riders can check out bikes at one of the 20 kiosks or via the Biketown website or free smartphone app. When you're done riding, you can park the bike at any bike rack in the service area or return it to one of the system's 100 stations and receive a credit.
Rideshare – Uber and Lyft. Sign up in the smartphone app for use.
eScooter Share – Lime, Bird and Skip. Download the smartphone apps for use.
These are the hotels located closest to the venue.
We have set up a room block and discounted rate with Hotel Modera for our attendees. To reserve a room, please call the hotel directly (number listed above) or click this link. The group code, "235190310MOBIL" will be automatically populated - you will just need to update the date range. The group rooms are being held until Monday, February 18th, 2019. After that date passes, they will no longer be holding any rooms for us and the group rate will no longer be available. Additional reservations beyond the cutoff date will be subject to their best available rate and general availability. If you are interested in booking dates or room types outside of what has been contracted, please contact the hotel directly as you cannot do so online.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 5,953 |
[
128000,
791,
1567,
374,
21685,
389,
279,
2132,
6558,
315,
279,
9259,
27872,
11988,
9323,
389,
279,
23947,
3314,
3907,
320,
5119,
52,
8,
15679,
520,
220,
10828,
20,
13692,
37776,
11,
23947,
11,
2794,
220,
24425,
1721,
510,
2235,
27218,
1687,
7079,
1701,
23947,
596,
586,
25032,
1887,
11,
12639,
35773,
11,
439,
279,
81610,
323,
6784,
12862,
3072,
13,
23947,
596,
3177,
13881,
8640,
505,
279,
393,
17001,
21348,
311,
19441,
23947,
13,
8498,
28788,
11776,
505,
279,
17149,
1475,
220,
868,
4520,
389,
73095,
13,
48534,
28788,
1629,
2753,
14134,
26,
1475,
220,
508,
12,
914,
4520,
13,
1183,
1771,
11776,
279,
17149,
505,
220,
20,
309,
311,
220,
806,
25,
1774,
5298,
13,
10175,
288,
649,
387,
15075,
520,
279,
86144,
12933,
4871,
279
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
791,
1567,
374,
21685,
389,
279,
2132,
6558,
315,
279,
9259,
27872,
11988,
9323,
389,
279,
23947,
3314,
3907,
320,
5119,
52,
8,
15679,
520,
220,
10828,
20,
13692,
37776,
11,
23947,
11,
2794,
220,
24425,
1721,
510,
2235,
27218,
1687,
7079,
1701,
23947,
596,
586,
25032,
1887,
11,
12639,
35773,
11,
439,
279,
81610,
323,
6784,
12862,
3072,
13,
23947,
596,
3177,
13881,
8640,
505,
279,
393,
17001,
21348,
311,
19441,
23947,
13,
8498,
28788,
11776,
505,
279,
17149,
1475,
220,
868,
4520,
389,
73095,
13,
48534,
28788,
1629,
2753,
14134,
26,
1475,
220,
508,
12,
914,
4520,
13,
1183,
1771,
11776,
279,
17149,
505,
220,
20,
309,
311,
220,
806,
25,
1774,
5298,
13,
10175,
288,
649,
387,
15075,
520,
279,
86144,
12933,
4871,
279,
-100
] |
Abstract: Book reviewed in this article: A Collection of Close to-Death Analysis Readings: Scientijic Inquiries Into the Experiences of Persons Near Physical Death. If you ask them about other experiences from their life at the similar time, they are very fuzzy memories, if they have any at all. Moody thinks that shared death experiences act as strong proof for the view that the thoughts exists independently of the brain, mainly because the persons experiencing them are in no way getting impaired brain functions at the time. NDEs are encountered by individuals of all backgrounds, and most studies find the prevalence of NDEs to be 10-20 percent of persons who have come close to death.
Investigation on Close to Death Experiences (NDE): The Findings and a Case from an Intensive Care Prospective Study – Penny Sartori, Ph.D. Soon after she came back to life, she gained the ability to see the power around the earth and people's collective consciousness, she stated. Moody very first heard about shared death experiences in 1972 from a healthcare professor of his. Just after more than 30 years of study, Moody estimates that shared death experiences are as common as NDEs.
He compared a group of NDErs' accounts about their NDEs created 20 years apart and discovered that they remained closely identical more than time. The phenomenon was coined near-death experiences (NDEs) in the 1975 book Life Following Life" by Raymond Moody, M.D. and Ph.D. in philosophy and psychology. NDEs frequently include things like cognitive, affective, paranormal, and transcendental experiences. Often, the bystander would also experience the life overview of the dying individual.
A new study suggests that close to-death experiences could be the outcome of increased electrical activity in the brain near the moment of death. If you ask them about other experiences from their life at the very same time, they are incredibly fuzzy memories, if they have any at all. Moody thinks that shared death experiences act as sturdy proof for the view that the thoughts exists independently of the brain, simply because the people today experiencing them are in no way getting impaired brain functions at the time. NDEs are encountered by men and women of all backgrounds, and most studies find the prevalence of NDEs to be ten-20 % of men and women who have come close to death.
As he studied extra of these situations over the years, he identified that the attributes of shared death experiences are related to these of NDEs. A single of the most common capabilities of shared death experiences is that the shared death experiencer sees the spirit of the dying particular person, which appears as a transparent replica of the individual, or an oval or sphere of light leaving from the head or chest of the dying person's physical physique, Moody told The Epoch Occasions in an interview.
Even stronger proof, as Moody recounted through the interview with The Epoch Times, was the case of a priest and a nun in South Africa who had a automobile accident with each other and who each had cardiac arrest followed by an NDE. If you ask somebody about their close to-death expertise that happened 15 years ago, they tell it as if it happened yesterday. However, for the duration of his NDE, Alexander had such vivid experiences involving a number of senses, such as vision, hearing, and smell, that he mentioned he couldn't describe how awesome it was.
A further phenomenon connected to NDEs is shared death experiences, in which a particular person close to a dying individual experiences anything with the identical qualities as NDEs. Bruce Greyson, M.D. and director of the Division of Perceptual Research at the University of Virginia, mentioned NDEs are reputable mainly because the accounts by near-death experiencers (NDErs) of these events stay unchanged more than time.
Near Death Experiences Research – The phenomenon was coined close to-death experiences (NDEs) in the 1975 book Life Just after Life" by Raymond Moody, M.D. and Ph.D. in philosophy and psychology.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 8,574 |
[
128000,
16328,
25,
6017,
22690,
304,
420,
4652,
25,
362,
11348,
315,
13330,
311,
12,
40358,
18825,
4557,
826,
25,
21302,
3251,
292,
763,
33000,
32745,
279,
1398,
69597,
315,
63030,
31494,
28479,
16290,
13,
1442,
499,
2610,
1124,
922,
1023,
11704,
505,
872,
2324,
520,
279,
4528,
892,
11,
814,
527,
1633,
53833,
19459,
11,
422,
814,
617,
904,
520,
682,
13,
72401,
15849,
430,
6222,
4648,
11704,
1180,
439,
3831,
11311,
369,
279,
1684,
430,
279,
11555,
6866,
29235,
315,
279,
8271,
11,
14918,
1606,
279,
11434,
25051,
1124,
527,
304,
912,
1648,
3794,
50160,
8271,
5865,
520,
279,
892,
13,
452,
1170,
82,
527,
23926,
555,
7931,
315,
682,
36576,
11,
323,
1455,
7978,
1505,
279,
38009,
315,
452,
1170,
82,
311,
387,
220,
605
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
16328,
25,
6017,
22690,
304,
420,
4652,
25,
362,
11348,
315,
13330,
311,
12,
40358,
18825,
4557,
826,
25,
21302,
3251,
292,
763,
33000,
32745,
279,
1398,
69597,
315,
63030,
31494,
28479,
16290,
13,
1442,
499,
2610,
1124,
922,
1023,
11704,
505,
872,
2324,
520,
279,
4528,
892,
11,
814,
527,
1633,
53833,
19459,
11,
422,
814,
617,
904,
520,
682,
13,
72401,
15849,
430,
6222,
4648,
11704,
1180,
439,
3831,
11311,
369,
279,
1684,
430,
279,
11555,
6866,
29235,
315,
279,
8271,
11,
14918,
1606,
279,
11434,
25051,
1124,
527,
304,
912,
1648,
3794,
50160,
8271,
5865,
520,
279,
892,
13,
452,
1170,
82,
527,
23926,
555,
7931,
315,
682,
36576,
11,
323,
1455,
7978,
1505,
279,
38009,
315,
452,
1170,
82,
311,
387,
220,
605,
-100
] |
Q: Android Instrumentation - How to detect external activity call I am using Robotium for testing android applications and am interested in how to detect a situation when an external application activity (e.g. browser, camera, facebook) is being launched by tested app. There are several questions asking how to handle such situation, but all I need is to detect it, for example just log time when it's occurred. Is this possible using Robotium or simple Instrumentation?
BTW, I can't use the solution when system apps are being replaced by fake analogues, I can't even know what application (and when) will be launched, because the tests are random. (Yes, I know, it's a bad approach, but this is a project restriction).
A: See my answer here. The main idea is to detect current top Activity, which is Activity of external application. You can create Service, which will monitor top activities.
A: Anther way is to check when your activity looses focus. When it looses focus that means that the other external Activity has focus instead.
http://developer.android.com/reference/android/app/Activity.html#hasWindowFocus()
You can use Solo.waitForCondition() in conjunction with hasWindowFocus.
|
{
"redpajama_set_name": "RedPajamaStackExchange"
}
| 2,523 |
[
128000,
48,
25,
8682,
43405,
367,
482,
2650,
311,
11388,
9434,
5820,
1650,
358,
1097,
1701,
26945,
2411,
369,
7649,
2151,
8522,
323,
1097,
8173,
304,
1268,
311,
11388,
264,
6671,
994,
459,
9434,
3851,
5820,
320,
68,
1326,
13,
7074,
11,
6382,
11,
23795,
8,
374,
1694,
11887,
555,
12793,
917,
13,
2684,
527,
3892,
4860,
10371,
1268,
311,
3790,
1778,
6671,
11,
719,
682,
358,
1205,
374,
311,
11388,
433,
11,
369,
3187,
1120,
1515,
892,
994,
433,
596,
10222,
13,
2209,
420,
3284,
1701,
26945,
2411,
477,
4382,
43405,
367,
5380,
18066,
54,
11,
358,
649,
956,
1005,
279,
6425,
994,
1887,
10721,
527,
1694,
12860,
555,
12700,
24291,
1157,
11,
358,
649,
956,
1524,
1440,
1148,
3851,
320,
438,
994,
8,
690,
387,
11887
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
48,
25,
8682,
43405,
367,
482,
2650,
311,
11388,
9434,
5820,
1650,
358,
1097,
1701,
26945,
2411,
369,
7649,
2151,
8522,
323,
1097,
8173,
304,
1268,
311,
11388,
264,
6671,
994,
459,
9434,
3851,
5820,
320,
68,
1326,
13,
7074,
11,
6382,
11,
23795,
8,
374,
1694,
11887,
555,
12793,
917,
13,
2684,
527,
3892,
4860,
10371,
1268,
311,
3790,
1778,
6671,
11,
719,
682,
358,
1205,
374,
311,
11388,
433,
11,
369,
3187,
1120,
1515,
892,
994,
433,
596,
10222,
13,
2209,
420,
3284,
1701,
26945,
2411,
477,
4382,
43405,
367,
5380,
18066,
54,
11,
358,
649,
956,
1005,
279,
6425,
994,
1887,
10721,
527,
1694,
12860,
555,
12700,
24291,
1157,
11,
358,
649,
956,
1524,
1440,
1148,
3851,
320,
438,
994,
8,
690,
387,
11887,
-100
] |
Premature Ovarian Failure is diagnosed in 1 to 2% of women worldwide. This condition is associated with poor ovarian reserve, causing infertility among women who are under the age of 40. There are options for egg and/or embryo donation to achieve pregnancy through in-vitro fertilization with a reproductive endocrinologist. Barriers associated with this diagnosis are increased emotions, stress, poor health literacy, financial pressure, and psychological blocks when battling infertility. The importance of a well-rounded understanding of Premature Ovarian Failure is crucial to stay positive and focused during diagnosis and treatment. Emphasis should be laced within medical practices on providing resources for psychological and peer support after diagnosis. Patients diagnosed with Premature Ovarian Failure typically do not have enough information given to them on their initial consultations and this seeks to develop a well-rounded grasp of knowledge.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 1,318 |
[
128000,
42562,
1598,
507,
959,
1122,
33360,
374,
29704,
304,
220,
16,
311,
220,
17,
4,
315,
3278,
15603,
13,
1115,
3044,
374,
5938,
449,
8009,
85542,
21137,
11,
14718,
95643,
4315,
3278,
889,
527,
1234,
279,
4325,
315,
220,
1272,
13,
2684,
527,
2671,
369,
19151,
323,
5255,
87701,
25968,
311,
11322,
20209,
1555,
304,
8437,
275,
299,
36214,
2065,
449,
264,
42889,
842,
91153,
16549,
13,
4821,
17740,
5938,
449,
420,
23842,
527,
7319,
21958,
11,
8631,
11,
8009,
2890,
53082,
11,
6020,
7410,
11,
323,
24064,
10215,
994,
47336,
95643,
13,
578,
12939,
315,
264,
1664,
65318,
8830,
315,
12029,
1598,
507,
959,
1122,
33360,
374,
16996,
311,
4822,
6928,
323,
10968,
2391,
23842,
323,
6514,
13,
5867,
55249,
1288,
387,
326,
4535,
2949,
6593
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
42562,
1598,
507,
959,
1122,
33360,
374,
29704,
304,
220,
16,
311,
220,
17,
4,
315,
3278,
15603,
13,
1115,
3044,
374,
5938,
449,
8009,
85542,
21137,
11,
14718,
95643,
4315,
3278,
889,
527,
1234,
279,
4325,
315,
220,
1272,
13,
2684,
527,
2671,
369,
19151,
323,
5255,
87701,
25968,
311,
11322,
20209,
1555,
304,
8437,
275,
299,
36214,
2065,
449,
264,
42889,
842,
91153,
16549,
13,
4821,
17740,
5938,
449,
420,
23842,
527,
7319,
21958,
11,
8631,
11,
8009,
2890,
53082,
11,
6020,
7410,
11,
323,
24064,
10215,
994,
47336,
95643,
13,
578,
12939,
315,
264,
1664,
65318,
8830,
315,
12029,
1598,
507,
959,
1122,
33360,
374,
16996,
311,
4822,
6928,
323,
10968,
2391,
23842,
323,
6514,
13,
5867,
55249,
1288,
387,
326,
4535,
2949,
6593,
-100
] |
Fraps Free Download – Fraps is a video recording software which is easy to use and very popular among youtubers. You can easily record Fraps without hassle and view your frames per seconds on any game you play. Learn how to download and Install fraps for free in this article and be sure to share with your friends.
Fraps is a video recording software which is easy to use and very popular among youtube content creators. The number of users have skyrocketed over the years and remains one of the easiest and best recording software for videos. This article will tell you how to download Fraps full version for FREE. Not only can you record videos, but you can also take screenshots of the games you are playing and control the background audio of your game. You can record as long as you want and aren't capped with 30 seconds like the free version. On top of all that, you can even measure your frames per second while gaming.
Click install and let fraps install.
You're done! Double click on fraps on your desktop and from there, you can start recording!
Click the download button below to start Fraps Free Download. It is the full version of Fraps. Don't forget to share with your friends.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 9,617 |
[
128000,
23376,
2690,
3658,
8745,
1389,
2939,
2690,
374,
264,
2835,
14975,
3241,
902,
374,
4228,
311,
1005,
323,
1633,
5526,
4315,
379,
412,
392,
388,
13,
1472,
649,
6847,
3335,
2939,
2690,
2085,
47947,
323,
1684,
701,
14418,
824,
6622,
389,
904,
1847,
499,
1514,
13,
15281,
1268,
311,
4232,
323,
19796,
1448,
2690,
369,
1949,
304,
420,
4652,
323,
387,
2771,
311,
4430,
449,
701,
4885,
627,
23376,
2690,
374,
264,
2835,
14975,
3241,
902,
374,
4228,
311,
1005,
323,
1633,
5526,
4315,
28277,
2262,
34899,
13,
578,
1396,
315,
3932,
617,
75722,
291,
927,
279,
1667,
323,
8625,
832,
315,
279,
30689,
323,
1888,
14975,
3241,
369,
6946,
13,
1115,
4652,
690,
3371,
499,
1268,
311,
4232,
2939,
2690,
2539,
2373,
369,
16655,
13,
2876
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
23376,
2690,
3658,
8745,
1389,
2939,
2690,
374,
264,
2835,
14975,
3241,
902,
374,
4228,
311,
1005,
323,
1633,
5526,
4315,
379,
412,
392,
388,
13,
1472,
649,
6847,
3335,
2939,
2690,
2085,
47947,
323,
1684,
701,
14418,
824,
6622,
389,
904,
1847,
499,
1514,
13,
15281,
1268,
311,
4232,
323,
19796,
1448,
2690,
369,
1949,
304,
420,
4652,
323,
387,
2771,
311,
4430,
449,
701,
4885,
627,
23376,
2690,
374,
264,
2835,
14975,
3241,
902,
374,
4228,
311,
1005,
323,
1633,
5526,
4315,
28277,
2262,
34899,
13,
578,
1396,
315,
3932,
617,
75722,
291,
927,
279,
1667,
323,
8625,
832,
315,
279,
30689,
323,
1888,
14975,
3241,
369,
6946,
13,
1115,
4652,
690,
3371,
499,
1268,
311,
4232,
2939,
2690,
2539,
2373,
369,
16655,
13,
2876,
-100
] |
Did the 'Joker' director just confirm a major DCEU fan theory?
Joaquin Phoenix's Joker might not be who we think he is.
Jake Kleinman
10.7.2019 4:32 PM
Before Joker even came out, fans of the Batman arch-villain pointed out one weird inconsistency in the movie: If Joker's origin story takes place when Bruce Wayne is a young boy, how could Batman grow up to fight a criminal who would by then be an old man? The response, for many, was that Arthur Fleck (Joaquin Phoenix) isn't the actual "Joker," he's the guy who inspires the Joker. It was an interesting theory without much evidence to back it up, but now, a quote from the Joker director himself may confirm this theory was right all along.
Speaking to The Los Angeles Times, director Todd Phillips heavily implied that this Joker theory was right.
"Maybe Joaquin's character inspired the Joker," Phillips said. "You don't really know. His last line in the movie is, 'You wouldn't get it.' There's a lot going on in there that's interesting."
While I'd argue that Phoenix's last line of the movie isn't particularly interesting (they should have cut that whole final scene, to be honest), it is interesting to consider that Arthur Fleck isn't actually the Joker who's appeared for decades in comic books, cartoons, and DC movies.
But to really unpack what this means, we have to talk about Joker's ending. So consider yourself warned: Full spoilers for the end of Joker after this creepy GIF.
Joker in 'Batman: The Animated Series' DC Entertainment
By the end of Joker, Arthur Fleck has completed his transformation into killer clown. In the process, he's inspired a wave of riots across the city as angry Gothamites in clown masks and makeup burn, loot, and, in at least one case, murder. As Fleck dances in front of the crowd, we watch some other nameless clown gun down Thomas and Martha Wayne in front of their son, Bruce. This jarring juxtaposition seems to imply a connection between the two characters, Joker and Batman, but maybe that wasn't Phillips' intention.
After all, any of the faceless rioters we see at the end of Joker could easily transform into the Joker later on. It might even be the man who killed Bruce's parents in a nod to the backstory Tim Burton invented for Jack Nicholson's Joker in Batman (1989). We never learn this killer's age, name, or backstory, which also fits pretty well with the traditional portrayal of the Joker as an anarchic force of destruction with no clear identity or motivation.
Then again, this might not matter. Phillips has repeatedly said he has no plans to make a sequel, and there's no post-credits scene in Joker hinting at a broader Joker-verse. Then again, the director did recently say he'd consider making Joker 2 if he can come up with the right idea, so we wouldn't rule anything out. At the very least, don't be surprised if some future DC movie decides to sneak in an Arthur Flick reference as a way to tie this movie into the broader DC Extended Universe.
Joker is in theaters now.
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 3,459 |
[
128000,
7131,
279,
364,
41,
10668,
6,
7690,
1120,
7838,
264,
3682,
423,
2152,
52,
8571,
10334,
5380,
41,
20103,
36444,
23503,
596,
61377,
2643,
539,
387,
889,
584,
1781,
568,
374,
627,
95336,
43241,
1543,
198,
605,
13,
22,
13,
679,
24,
220,
19,
25,
843,
5975,
198,
10438,
61377,
1524,
3782,
704,
11,
7359,
315,
279,
27805,
5438,
8437,
484,
467,
14618,
704,
832,
16682,
97249,
304,
279,
5818,
25,
1442,
61377,
596,
6371,
3446,
5097,
2035,
994,
24785,
28640,
374,
264,
3995,
8334,
11,
1268,
1436,
27805,
3139,
709,
311,
4465,
264,
9337,
889,
1053,
555,
1243,
387,
459,
2362,
893,
30,
578,
2077,
11,
369,
1690,
11,
574,
430,
28686,
47790,
377,
320,
41,
20103,
36444,
23503,
8,
4536,
956,
279,
5150,
330,
41
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
7131,
279,
364,
41,
10668,
6,
7690,
1120,
7838,
264,
3682,
423,
2152,
52,
8571,
10334,
5380,
41,
20103,
36444,
23503,
596,
61377,
2643,
539,
387,
889,
584,
1781,
568,
374,
627,
95336,
43241,
1543,
198,
605,
13,
22,
13,
679,
24,
220,
19,
25,
843,
5975,
198,
10438,
61377,
1524,
3782,
704,
11,
7359,
315,
279,
27805,
5438,
8437,
484,
467,
14618,
704,
832,
16682,
97249,
304,
279,
5818,
25,
1442,
61377,
596,
6371,
3446,
5097,
2035,
994,
24785,
28640,
374,
264,
3995,
8334,
11,
1268,
1436,
27805,
3139,
709,
311,
4465,
264,
9337,
889,
1053,
555,
1243,
387,
459,
2362,
893,
30,
578,
2077,
11,
369,
1690,
11,
574,
430,
28686,
47790,
377,
320,
41,
20103,
36444,
23503,
8,
4536,
956,
279,
5150,
330,
41,
-100
] |
The Government of India's Department of Biotechnology (DBT) and Bharat Biotech announced positive results from a Phase III clinical trial of a rotavirus vaccine developed and manufactured in India. Data from the trial, presented on May 14, 2013 at the International Symposium on Rotavirus Vaccines for India—The Evidence and the Promise, showed ROTAVAC® to have an excellent safety and efficacy profile. The clinical study demonstrates for the first time that the India-developed rotavirus vaccine ROTAVAC® is efficacious in preventing severe rotavirus diarrhea in low-resource settings in India. ROTAVAC® significantly reduced severe rotavirus diarrhea by more than half—56 percent during the first year of life, with protection continuing into the second year of life. Moreover, the vaccine also showed impact against severe diarrhea of any cause. "This is an important scientific breakthrough against rotavirus infections, the most severe and lethal cause of childhood diarrhea, responsible for approximately 100,000 deaths of small children in India each year," said DBT Secretary Dr K. Vijay Raghavan. "The clinical results indicate that the vaccine, if licensed, could save the lives of thousands of children each year in India." The vaccine was developed through a unique social innovation partnership that brought together the experience and expertise of Indian and international researchers as well as the public and private sectors. The vaccine originated from an attenuated strain of rotavirus that was isolated from an Indian child at the All India Institute of Medical Sciences in New Delhi in 1985-86. Since then, partners have included DBT, Bharat Biotech, the US National Institutes of Health (NIH), the US Centers for Disease Control and Prevention (CDC), Stanford University School of Medicine, and the nongovernmental organization, PATH.
On May 10, 2013, the Journal of the American Heart Association published, online, the conclusive results from a study directed by Dr. Éric Thorin of the Montreal Heart Institute (MHI), which suggests for the first time that a blood protein contributes to the early development of atherosclerosis. Dr. Thorin, his team, and his collaborators discovered that the blood levels of angiopoietin-like protein 2 (angptl2) are six times higher in subjects with coronary heart disease than in healthy subjects of the same age. Their basic research study also revealed that angptl2, which is undetectable in young mice, increases with age in healthy subjects and increases prematurely in subjects who have high cholesterol and pre-atherosclerotic lesions. This study was conducted using an animal model consisting of three- to twelve-month-old mice. These results represent a major advance in the prevention and treatment of atherosclerosis. "Although much work remains to be done to broaden our knowledge of this protein's mechanisms of action, angiopoietin-like protein 2 may represent an early biomarker not only to prevent vascular damage, but also to predict atherosclerotic disease," explained Dr. Thorin. For 15 years, Dr. Thorin, a researcher at the MHI Research Centre and full professor at Université de Montréal, has been interested in the evolution of artery function during the aging process and in the underlying mechanisms of atherosclerosis. More specifically, over the past five years, he has looked at the role of this particular protein. Thanks to his work, we now know that angptl2 causes a high degree of vascular inflammation.
Ever notice how a dried coffee stain has a thicker outer rim, while the middle of the stain remains almost unsoiled? This "coffee ring effect" also occurs in other materials. Researchers from the Departments of Chemical Engineering and Chemistry at KU Leuven in Belgium have now discovered how to counteract coffee rings with "surfactants," i.e., soap. The key to the discovery was not a kitchen towel, but a bacterium that counteracts the coffee ring effect at the microscopic level. The findings were published on April 23, 2013 in an open-access article in Nature Communications. When a coffee ring dries, its edges become noticeably darker and thicker. This occurs because the coffee particles move toward the edge of the stain while the water in the liquid evaporates. At a microscopic level, this coffee ring effect can also be seen in liquids with particles of other materials such as plastic and wood. In various industrial applications – applying an even coat of paint or varnish, for example – the coffee ring effect can be particularly troublesome and scientists have long been seeking ways to counteract it. Dr. Raf De Dier and Dr. Wouter Sempels (Departments of Chemical Engineering and Chemistry) have now described a solution based on examples found in nature. Drs. De Dier and Sempels carried out experiments and calculations on nanomaterials, as well as on a particularly promising bacterium, Pseudomonas aeruginosa. Pseudomonas aeruginosa is a dangerous bacterium that can cause infections in open wounds. "A Pseudomonas aeruginosa bacteria colony wants to find as large a breeding ground as possible.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 1,090 |
[
128000,
791,
10423,
315,
6890,
596,
6011,
315,
12371,
52536,
320,
3590,
51,
8,
323,
67692,
266,
12371,
59196,
7376,
6928,
3135,
505,
264,
28673,
14767,
14830,
9269,
315,
264,
5868,
24713,
25474,
8040,
323,
28648,
304,
6890,
13,
2956,
505,
279,
9269,
11,
10666,
389,
3297,
220,
975,
11,
220,
679,
18,
520,
279,
7327,
74938,
389,
28460,
24713,
59788,
1572,
369,
6890,
2345,
791,
44796,
323,
279,
7451,
11,
8710,
59311,
8253,
1741,
12175,
311,
617,
459,
9250,
7296,
323,
41265,
5643,
13,
578,
14830,
4007,
32216,
369,
279,
1176,
892,
430,
279,
6890,
49061,
291,
5868,
24713,
25474,
59311,
8253,
1741,
12175,
374,
31914,
19995,
304,
27252,
15748,
5868,
24713,
69393,
304,
3428,
75890,
5110,
304,
6890,
13,
59311,
8253,
1741,
12175,
12207,
11293,
15748
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
791,
10423,
315,
6890,
596,
6011,
315,
12371,
52536,
320,
3590,
51,
8,
323,
67692,
266,
12371,
59196,
7376,
6928,
3135,
505,
264,
28673,
14767,
14830,
9269,
315,
264,
5868,
24713,
25474,
8040,
323,
28648,
304,
6890,
13,
2956,
505,
279,
9269,
11,
10666,
389,
3297,
220,
975,
11,
220,
679,
18,
520,
279,
7327,
74938,
389,
28460,
24713,
59788,
1572,
369,
6890,
2345,
791,
44796,
323,
279,
7451,
11,
8710,
59311,
8253,
1741,
12175,
311,
617,
459,
9250,
7296,
323,
41265,
5643,
13,
578,
14830,
4007,
32216,
369,
279,
1176,
892,
430,
279,
6890,
49061,
291,
5868,
24713,
25474,
59311,
8253,
1741,
12175,
374,
31914,
19995,
304,
27252,
15748,
5868,
24713,
69393,
304,
3428,
75890,
5110,
304,
6890,
13,
59311,
8253,
1741,
12175,
12207,
11293,
15748,
-100
] |
Quick, who bears the primary responsibility for training children to trust Christ and follow Christ as mature disciples? If you answered parents, you are correct.
An epidemic has developed in churches across the United States, one that actually is destroying our families and hurting our children in the process. We have unwittingly allowed the view that the church bears the primary responsibility for training our children in the nurture and admonition of the Lord to replace and undermine God's plan of discipleship.
When this view takes hold, the church tends to be viewed less as a family of faith in which we participate and more as a grocery store where we swing by to pick up our religious goods for the week. We drop our kids off in their respective children's or youth ministry and entrust that the trained leaders will teach our kids all they need to know to follow Jesus. The more we drop them off, the more they will learn. The more they learn, the safer they will be as they grow up into adults. Parents are divorced from the process, other than serving as leaders of various ministries.
This model has failed churches and families. The evidence is staggering. One recent study concluded that 70% of young adults (ages 23-30) dropped out of church for at least a year between the ages of 18-22. Many of those who dropped out were active members in their church youth group.
You will find all these commitments emphasized in the children and youth ministries of Homesteads Baptist Church.
The Family Equipping Ministry Model is not a one step, fix-all that makes parenting or even children's ministry easy. In fact, it is probably a much more difficult paradigm to employ than the traditional children's ministry model. At Homesteads Baptist Church, however, we believe our children (and yours) are worth the investment. We hope you will join us for the process of transformation.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 88 |
[
128000,
25310,
11,
889,
30824,
279,
6156,
12014,
369,
4967,
2911,
311,
7095,
3771,
323,
1833,
3771,
439,
15196,
49260,
30,
1442,
499,
19089,
6699,
11,
499,
527,
4495,
627,
2127,
42620,
706,
8040,
304,
31012,
4028,
279,
3723,
4273,
11,
832,
430,
3604,
374,
33812,
1057,
8689,
323,
48389,
1057,
2911,
304,
279,
1920,
13,
1226,
617,
95119,
398,
5535,
279,
1684,
430,
279,
8993,
30824,
279,
6156,
12014,
369,
4967,
1057,
2911,
304,
279,
79530,
323,
83085,
684,
315,
279,
10425,
311,
8454,
323,
43047,
4359,
596,
3197,
315,
49260,
2200,
627,
4599,
420,
1684,
5097,
3412,
11,
279,
8993,
28335,
311,
387,
19894,
2753,
439,
264,
3070,
315,
10082,
304,
902,
584,
16136,
323,
810,
439,
264,
30687,
3637,
1405,
584,
19336,
555,
311,
3820
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
25310,
11,
889,
30824,
279,
6156,
12014,
369,
4967,
2911,
311,
7095,
3771,
323,
1833,
3771,
439,
15196,
49260,
30,
1442,
499,
19089,
6699,
11,
499,
527,
4495,
627,
2127,
42620,
706,
8040,
304,
31012,
4028,
279,
3723,
4273,
11,
832,
430,
3604,
374,
33812,
1057,
8689,
323,
48389,
1057,
2911,
304,
279,
1920,
13,
1226,
617,
95119,
398,
5535,
279,
1684,
430,
279,
8993,
30824,
279,
6156,
12014,
369,
4967,
1057,
2911,
304,
279,
79530,
323,
83085,
684,
315,
279,
10425,
311,
8454,
323,
43047,
4359,
596,
3197,
315,
49260,
2200,
627,
4599,
420,
1684,
5097,
3412,
11,
279,
8993,
28335,
311,
387,
19894,
2753,
439,
264,
3070,
315,
10082,
304,
902,
584,
16136,
323,
810,
439,
264,
30687,
3637,
1405,
584,
19336,
555,
311,
3820,
-100
] |
Maintenance is defined as all actions that aim to preserve or restore an item to a state in which it can perform a required function. These actions include a combination of the corresponding technical and administrative actions .
Comprehensive preventive maintenance at the community , commercial or individual level (gardening , swimming pools, repairs, ... ) .
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 5,034 |
[
128000,
93835,
374,
4613,
439,
682,
6299,
430,
9395,
311,
21813,
477,
15301,
459,
1537,
311,
264,
1614,
304,
902,
433,
649,
2804,
264,
2631,
734,
13,
4314,
6299,
2997,
264,
10824,
315,
279,
12435,
11156,
323,
23541,
6299,
16853,
1110,
53999,
71123,
13709,
520,
279,
4029,
1174,
8518,
477,
3927,
2237,
320,
54592,
6147,
1174,
24269,
34239,
11,
31286,
11,
2564,
883,
662,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
] |
[
93835,
374,
4613,
439,
682,
6299,
430,
9395,
311,
21813,
477,
15301,
459,
1537,
311,
264,
1614,
304,
902,
433,
649,
2804,
264,
2631,
734,
13,
4314,
6299,
2997,
264,
10824,
315,
279,
12435,
11156,
323,
23541,
6299,
16853,
1110,
53999,
71123,
13709,
520,
279,
4029,
1174,
8518,
477,
3927,
2237,
320,
54592,
6147,
1174,
24269,
34239,
11,
31286,
11,
2564,
883,
662,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100
] |
Hisarya (en ), également connue sous les noms de Hisar, Hissar ou Hissarya, anciennement Toplitsa, en latin : Diocletianopolis ou Augusta, est une ville balnéaire de Bulgarie située dans la province (oblast) de Plovdiv et qui fait partie de l'obchtina de Khisarya.
Géographie
Établie à la périphérie de la chaîne de montagnes Sredna Gora, la ville a un climat très doux. Plus d'un vingtaine de sources d'eau minérale attirent les touristes, tant bulgares qu'étrangers.
En 2015, la ville comptait une population de .
Notes et références
Liens externes
Ville en Bulgarie
|
{
"redpajama_set_name": "RedPajamaWikipedia"
}
| 3,281 |
[
128000,
16366,
73268,
320,
268,
7026,
44767,
4635,
361,
32592,
3625,
308,
7085,
409,
5414,
277,
11,
473,
1056,
277,
6033,
473,
1056,
73268,
11,
459,
5979,
2734,
1133,
7054,
75,
1220,
64,
11,
665,
64019,
551,
7923,
511,
1169,
1122,
28765,
285,
6033,
84575,
11,
1826,
6316,
39973,
9839,
52235,
12267,
409,
13805,
12440,
648,
10109,
8047,
7010,
1208,
17271,
320,
677,
4354,
8,
409,
393,
36247,
614,
1880,
7930,
20272,
48901,
409,
326,
6,
677,
14244,
2259,
409,
20774,
285,
73268,
382,
38,
978,
3257,
648,
720,
27887,
6323,
11828,
3869,
1208,
84411,
764,
35285,
409,
1208,
13826,
25108,
818,
409,
20605,
3326,
288,
328,
1171,
3458,
480,
6347,
11,
1208,
39973,
264,
653,
11323,
266,
25945,
25595,
87,
13,
12623,
294,
22827,
348,
76019,
8511
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
16366,
73268,
320,
268,
7026,
44767,
4635,
361,
32592,
3625,
308,
7085,
409,
5414,
277,
11,
473,
1056,
277,
6033,
473,
1056,
73268,
11,
459,
5979,
2734,
1133,
7054,
75,
1220,
64,
11,
665,
64019,
551,
7923,
511,
1169,
1122,
28765,
285,
6033,
84575,
11,
1826,
6316,
39973,
9839,
52235,
12267,
409,
13805,
12440,
648,
10109,
8047,
7010,
1208,
17271,
320,
677,
4354,
8,
409,
393,
36247,
614,
1880,
7930,
20272,
48901,
409,
326,
6,
677,
14244,
2259,
409,
20774,
285,
73268,
382,
38,
978,
3257,
648,
720,
27887,
6323,
11828,
3869,
1208,
84411,
764,
35285,
409,
1208,
13826,
25108,
818,
409,
20605,
3326,
288,
328,
1171,
3458,
480,
6347,
11,
1208,
39973,
264,
653,
11323,
266,
25945,
25595,
87,
13,
12623,
294,
22827,
348,
76019,
8511,
-100
] |
The River Plantation Invitational was a golf tournament on the LPGA Tour from 1968 to 1969. It was played at the River Plantation Country Club in Conroe, Texas. Kathy Whitworth won both editions of the event.
Winners
River Plantation Women's Open
1969 Kathy Whitworth
River Plantation Invitational
1968 Kathy Whitworth
References
Former LPGA Tour events
Golf in Texas
Conroe, Texas
1968 establishments in Texas
1969 disestablishments in Texas
History of women in Texas
|
{
"redpajama_set_name": "RedPajamaWikipedia"
}
| 9,528 |
[
128000,
791,
11188,
18317,
367,
19337,
50924,
574,
264,
19665,
16520,
389,
279,
445,
59699,
14986,
505,
220,
5162,
23,
311,
220,
5162,
24,
13,
1102,
574,
6476,
520,
279,
11188,
18317,
367,
14438,
10349,
304,
1221,
35804,
11,
8421,
13,
65298,
27484,
20069,
2834,
2225,
47634,
315,
279,
1567,
382,
17400,
5079,
198,
86799,
18317,
367,
11215,
596,
5377,
198,
5162,
24,
65298,
27484,
20069,
271,
86799,
18317,
367,
19337,
50924,
198,
5162,
23,
65298,
27484,
20069,
271,
32812,
271,
31945,
445,
59699,
14986,
4455,
198,
38,
8255,
304,
8421,
198,
1128,
35804,
11,
8421,
198,
5162,
23,
63337,
304,
8421,
198,
5162,
24,
834,
34500,
1392,
304,
8421,
198,
13730,
315,
3278,
304,
8421,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
] |
[
791,
11188,
18317,
367,
19337,
50924,
574,
264,
19665,
16520,
389,
279,
445,
59699,
14986,
505,
220,
5162,
23,
311,
220,
5162,
24,
13,
1102,
574,
6476,
520,
279,
11188,
18317,
367,
14438,
10349,
304,
1221,
35804,
11,
8421,
13,
65298,
27484,
20069,
2834,
2225,
47634,
315,
279,
1567,
382,
17400,
5079,
198,
86799,
18317,
367,
11215,
596,
5377,
198,
5162,
24,
65298,
27484,
20069,
271,
86799,
18317,
367,
19337,
50924,
198,
5162,
23,
65298,
27484,
20069,
271,
32812,
271,
31945,
445,
59699,
14986,
4455,
198,
38,
8255,
304,
8421,
198,
1128,
35804,
11,
8421,
198,
5162,
23,
63337,
304,
8421,
198,
5162,
24,
834,
34500,
1392,
304,
8421,
198,
13730,
315,
3278,
304,
8421,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100
] |
I approve each and every comment, EXCEPT for those that are unsigned, and those that are Spam. I DO NOT want, need or like anonymous comments. If you do not have the conviction to stand behind your words, then they won't be published here.
Page 3 of the "return or destroy" letter. En Francais.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 9,521 |
[
128000,
40,
29251,
1855,
323,
1475,
4068,
11,
4154,
21752,
369,
1884,
430,
527,
3859,
11,
323,
1884,
430,
527,
82767,
13,
358,
9503,
4276,
1390,
11,
1205,
477,
1093,
22939,
6170,
13,
1442,
499,
656,
539,
617,
279,
29191,
311,
2559,
4920,
701,
4339,
11,
1243,
814,
2834,
956,
387,
4756,
1618,
627,
2732,
220,
18,
315,
279,
330,
693,
477,
7066,
1,
6661,
13,
2998,
31925,
936,
285,
13,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
] |
[
40,
29251,
1855,
323,
1475,
4068,
11,
4154,
21752,
369,
1884,
430,
527,
3859,
11,
323,
1884,
430,
527,
82767,
13,
358,
9503,
4276,
1390,
11,
1205,
477,
1093,
22939,
6170,
13,
1442,
499,
656,
539,
617,
279,
29191,
311,
2559,
4920,
701,
4339,
11,
1243,
814,
2834,
956,
387,
4756,
1618,
627,
2732,
220,
18,
315,
279,
330,
693,
477,
7066,
1,
6661,
13,
2998,
31925,
936,
285,
13,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100
] |
Male YouTube Artist
Jenny Aluning
Male YouTube Artist of the Month for March 2013!
For StarCentral Magazine's Male YouTube Artist of the Month for March 2013, Steve Aiston "…singing has always been my passion, and my inspiration comes from all my friends and family who have supported me from day 1…" He began performing when he was still a young boy in primary school but as pop music gained popularity in his teen years, he was inspired by the genre and Steve decided to start entering local singing competitions. One of his most memorable performances was when he first won of those same competitions – "…I was really nervous had all my friends watching me for the first time so felt I couldn't let them down!"
The 30-year old has come a long way from those early competitions. He has been a member of the boy band, 'Uptown', and as part of that band competed in 'The X-Factor' Boot Camp, appeared on UK's Sky One's 'Showbusiness', a performance documentary centred on choreographer, Louie Spence as he tries "to put on the best West End show London has ever seen!" Uptown was also featured in ITV2's 'The Only Way Is Essex' which is a part soap opera, part reality show that chronicles the outrageous lives of a group of real-life Essex guys and girls. As a soloist, his voice has earned him a glowing reputation all his own. He is a frequent featured act at many private functions, clubs, festivals, corporate events and hotels. "…I most enjoy the energy and buzz of the audience's reaction when performing and seeing people enjoy themselves…"
The future is wide open for the London-based singer/songwriter. He is currently working on his album with the first single, 'Is This Love' due for release very soon. "…I'd also one day love to do some acting…" With a "…a great family and partner…" behind him, as well as his incredible talent, it's pretty certain that Steve will be able to live up to his motto: "Never give up on your dream…"
Photos by: John Milton
Amazing Originals and Covers from StarCentral's A-List!
Male YouTube Artist of the Month for June 2013!
Male YouTube Artist of the Month for December 2013!
Male YouTube Artist of the Month for May 2013!
Male YouTube Artist of the Month for October 2013!
Male YouTube artistStarCentral Male YouTube Artist of the MonthStarCentral Male YouTube Artist of the Month for March 2013StarCentral YouTube artistSteve AistonSteve Aiston Male YouTube artistSteve Aiston Male YouTube artist March 2013Steve Aiston singerSteve Aiston StarCentralSteve Aiston StarCentral Male YouTube artistSteve Aiston StarCentral Male YouTube artist March 2013Steve Aiston StarCentral Male YouTube Artist of the MonthSteve Aiston StarCentral Male YouTube Artist of the Month for March 2013Steve Aiston YouTube artist of the monthUptown Steve AistonYoutube ArtistYoutube Artist Of The Month
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 6,181 |
[
128000,
37521,
13674,
29459,
198,
41,
18314,
1708,
38302,
198,
37521,
13674,
29459,
315,
279,
19961,
369,
5587,
220,
679,
18,
4999,
2520,
7834,
44503,
22168,
596,
19960,
13674,
29459,
315,
279,
19961,
369,
5587,
220,
679,
18,
11,
14129,
362,
59919,
330,
1981,
82,
24992,
706,
2744,
1027,
856,
11939,
11,
323,
856,
20343,
4131,
505,
682,
856,
4885,
323,
3070,
889,
617,
7396,
757,
505,
1938,
220,
16,
56209,
1283,
6137,
16785,
994,
568,
574,
2103,
264,
3995,
8334,
304,
6156,
2978,
719,
439,
2477,
4731,
18661,
23354,
304,
813,
9562,
1667,
11,
568,
574,
14948,
555,
279,
17779,
323,
14129,
6773,
311,
1212,
16661,
2254,
26139,
43682,
13,
3861,
315,
813,
1455,
33596,
24601,
574,
994,
568,
1176,
2834,
315,
1884,
1890,
43682,
1389,
330
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
37521,
13674,
29459,
198,
41,
18314,
1708,
38302,
198,
37521,
13674,
29459,
315,
279,
19961,
369,
5587,
220,
679,
18,
4999,
2520,
7834,
44503,
22168,
596,
19960,
13674,
29459,
315,
279,
19961,
369,
5587,
220,
679,
18,
11,
14129,
362,
59919,
330,
1981,
82,
24992,
706,
2744,
1027,
856,
11939,
11,
323,
856,
20343,
4131,
505,
682,
856,
4885,
323,
3070,
889,
617,
7396,
757,
505,
1938,
220,
16,
56209,
1283,
6137,
16785,
994,
568,
574,
2103,
264,
3995,
8334,
304,
6156,
2978,
719,
439,
2477,
4731,
18661,
23354,
304,
813,
9562,
1667,
11,
568,
574,
14948,
555,
279,
17779,
323,
14129,
6773,
311,
1212,
16661,
2254,
26139,
43682,
13,
3861,
315,
813,
1455,
33596,
24601,
574,
994,
568,
1176,
2834,
315,
1884,
1890,
43682,
1389,
330,
-100
] |
Home Contributors
Nick Rowley
Nick Rowley is an Adjunct Professor at the Sydney Democracy Network at the University of Sydney.
Nick currently works as a strategic policy consultant to a mix of business and NGO clients in Australia and overseas. Over the past fifteen years, Nick has worked at the centre of government on sustainability, climate change and broader policy and political strategy in Australia and the UK. In these roles he developed new policy approaches in a number of areas from the funding of cutting-edge medical research, protecting public lands bordering Sydney Harbour to helping establish the seminal Stern Review into the Economics of Climate Change.
From March 2004 to January 2006, Nick worked at 10 Downing Street as an advisor to Prime Minister Tony Blair. In this role he oversaw all domestic and international climate policy and was part of a small team advising the British Prime Minister on climate change prior to the G8 Summit at Gleneagles in July 2005. From 1995 to 2004 Nick was advisor to Bob Carr, Premier of NSW working primarily on policy on the environment, urban development and medical research.
From 2006 to 2009 Nick was the Strategic Director of the Copenhagen Climate Council: an affiliation of climate scientists, business CEOs and policy practitioners brought together to make the case for a more adequate global climate treaty in 2009. He was also also a director of Kinesis: a business focussed on achieving measurable emissions reductions for private and government clients. For nine months prior to the introduction of the then Australian government's Clean Energy Future package of measures (including the carbon price, Climate Change Authority and Clean Energy Finance Corporation Nick worked as an Assistant Secretary in the Department of Climate Change responsible for the communications of effective climate policy.
Nick has just been awarded the role of Adjunct Professor to lead a project on the future democratisation of decision making on the future of Antarctica.
By Nick Rowley
The SDGs: a first step to enduring progress
Here's why the United Nations Sustainable Development Goals (SDGs) are potentially a powerful driver for changing global economies, and how it would affect us all.
15 Jun 2016 Opinion
Paris climate talks could be a light in the darkness of terror
After the November 13 attacks on Paris, the global community might now be more determined to deliver a successful climate change conference, both for France and as a potent demonstration of the enduring effectiveness of the international diplomatic process.
17 Nov 2015 Opinion
Climate change: a problem we can do something about
Failure to reduce emissions and the risks will have significant economic, social, health and security implications for businesses and governments, but there are also benefits for businesses which move to clean technologies and processes, says sustainability consultant Nick Rowley.
4 Nov 2015 Opinion
Do you want trees with that? How to stop consumer products destroying the rainforests
The world's rainforests are still being slashed and burned at a dizzying rate to make consumer products. But now there are signs of real political will, especially in Asia, to rein in the destruction, says University of Sydney professor Nick Rowley.
20 Oct 2015 Opinion
Climate change: business risk or opportunity?
Why the Paris climate talks won't be another Copenhagen
8 Apr 2015 Opinion
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 5,332 |
[
128000,
7778,
40636,
198,
31456,
11035,
3258,
198,
31456,
11035,
3258,
374,
459,
64016,
20526,
17054,
520,
279,
21972,
48189,
8304,
520,
279,
3907,
315,
21972,
627,
31456,
5131,
4375,
439,
264,
19092,
4947,
31760,
311,
264,
6651,
315,
2626,
323,
47886,
8403,
304,
8494,
323,
25355,
13,
6193,
279,
3347,
37755,
1667,
11,
15341,
706,
6575,
520,
279,
12541,
315,
3109,
389,
41329,
11,
10182,
2349,
323,
27927,
4947,
323,
5054,
8446,
304,
8494,
323,
279,
6560,
13,
763,
1521,
13073,
568,
8040,
502,
4947,
20414,
304,
264,
1396,
315,
5789,
505,
279,
11006,
315,
14713,
48448,
6593,
3495,
11,
22973,
586,
12098,
3973,
287,
21972,
63107,
311,
10695,
5813,
279,
90773,
51850,
10506,
1139,
279,
35300,
315,
31636,
10604,
627,
3915,
5587,
220,
1049,
19,
311
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
7778,
40636,
198,
31456,
11035,
3258,
198,
31456,
11035,
3258,
374,
459,
64016,
20526,
17054,
520,
279,
21972,
48189,
8304,
520,
279,
3907,
315,
21972,
627,
31456,
5131,
4375,
439,
264,
19092,
4947,
31760,
311,
264,
6651,
315,
2626,
323,
47886,
8403,
304,
8494,
323,
25355,
13,
6193,
279,
3347,
37755,
1667,
11,
15341,
706,
6575,
520,
279,
12541,
315,
3109,
389,
41329,
11,
10182,
2349,
323,
27927,
4947,
323,
5054,
8446,
304,
8494,
323,
279,
6560,
13,
763,
1521,
13073,
568,
8040,
502,
4947,
20414,
304,
264,
1396,
315,
5789,
505,
279,
11006,
315,
14713,
48448,
6593,
3495,
11,
22973,
586,
12098,
3973,
287,
21972,
63107,
311,
10695,
5813,
279,
90773,
51850,
10506,
1139,
279,
35300,
315,
31636,
10604,
627,
3915,
5587,
220,
1049,
19,
311,
-100
] |
The Board of Directors and the staff of the Canadian Association of Movers wish you, your fellow employees and your families a Happy Christmas and a safe and prosperous New Year. 2012 has been a busy year for the association and 2013 is expected to offer new challenges. The support of CAM's members and the efforts of CAM's executive, directors and committee members are appreciated.
e-mail or in person, thank you.
Please accept this as CAM's holiday greeting to you. Instead of sending out greeting cards, a donation has been made to families in the Third World.
We also hope that you will enjoy the following good-news items.
The current moving contract has been extended to September 30, 2013, with an option to extend it to March 31, 2014.
The current contract will include a 1.2% increase.
A meeting with Public Works and Government Services Canada (PWGSC) that includes all stakeholders will be held on January 29 in Ottawa.
At first glance, it appears that our lobbying efforts with PWGSC last August made a difference.
We have often wondered how reputable movers can distinguish themselves from the fly-by-night operator who works out of a back room or basement office. Thanks to Nate Edwards of Starline Overseas Moving in Calgary who has advised CAM of a new service from Google. It's called Google Business View and it's like Google Street for your premises. It allows a consumer to take an online walk-through of your business. Check out this app via Starline's blog.
At their last meeting, CAM's Board of Directors approved a modest fee increase for mover's head offices, commencing in 2013. The annual membership fee for a mover has increased from $425 to $450 per year (plus applicable tax) after having been frozen for several years. There are no changes to any other membership fees.
The offices of the Canadian Association of Movers will be closed from noon on Monday, December 24, 2012, to the opening of business on Wednesday, January 2, 2013. During that time, we will be reviewing all voicemail and e-mail but will respond only to the most urgent.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 388 |
[
128000,
791,
8925,
315,
46085,
323,
279,
5687,
315,
279,
12152,
10229,
315,
386,
9146,
6562,
499,
11,
701,
12637,
8420,
323,
701,
8689,
264,
24241,
10280,
323,
264,
6220,
323,
71407,
1561,
9941,
13,
220,
679,
17,
706,
1027,
264,
13326,
1060,
369,
279,
15360,
323,
220,
679,
18,
374,
3685,
311,
3085,
502,
11774,
13,
578,
1862,
315,
29397,
596,
3697,
323,
279,
9045,
315,
29397,
596,
11145,
11,
28454,
323,
13093,
3697,
527,
26893,
627,
68,
11724,
477,
304,
1732,
11,
9901,
499,
627,
5618,
4287,
420,
439,
29397,
596,
13560,
43213,
311,
499,
13,
12361,
315,
11889,
704,
43213,
7563,
11,
264,
25968,
706,
1027,
1903,
311,
8689,
304,
279,
21530,
4435,
627,
1687,
1101,
3987,
430,
499,
690,
4774,
279,
2768,
1695,
46351
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
791,
8925,
315,
46085,
323,
279,
5687,
315,
279,
12152,
10229,
315,
386,
9146,
6562,
499,
11,
701,
12637,
8420,
323,
701,
8689,
264,
24241,
10280,
323,
264,
6220,
323,
71407,
1561,
9941,
13,
220,
679,
17,
706,
1027,
264,
13326,
1060,
369,
279,
15360,
323,
220,
679,
18,
374,
3685,
311,
3085,
502,
11774,
13,
578,
1862,
315,
29397,
596,
3697,
323,
279,
9045,
315,
29397,
596,
11145,
11,
28454,
323,
13093,
3697,
527,
26893,
627,
68,
11724,
477,
304,
1732,
11,
9901,
499,
627,
5618,
4287,
420,
439,
29397,
596,
13560,
43213,
311,
499,
13,
12361,
315,
11889,
704,
43213,
7563,
11,
264,
25968,
706,
1027,
1903,
311,
8689,
304,
279,
21530,
4435,
627,
1687,
1101,
3987,
430,
499,
690,
4774,
279,
2768,
1695,
46351,
-100
] |
Supplement to The Scroggins Papers by William G. Scroggins
Note: This is the next to the last update of this page at this url. See more below.
Mary Adelaide Stout
Mary Adelaide Stout1
F, #7939, (c 1870 - 1928)
Father*: William Stout
Mother*: Mary Nancy Spencer (c 1839 - )
Reference: [7939:F]
Birth*: Ca. 1870, Pike Co, MO.2
(Parents, daughter of William Stout & Mary Nancy Spencer).
Marriage*: 30 May 1889, Archibald Johnson Scroggins Jr., Pike Co, MO, 39.350000,-91.166667.
Marriage: 1909, John Linden.
Marriage: 1917, William Vaughn.
Death*: 28 Oct 1928, St. Louis, MO.2
Birth of son: Aft. 1879?, Frank Clarence Scoggins.
Birth of son: 21 Jul 1891, William 'Hap' Scoggins Sr., Pike Co, MO, 39.350000,-91.166667.
Birth of daughter: 6 Dec 1897, Ruth Belle Scoggins, Alton, Madison Co, IL, 385326N0901103W, Date uncertain.3
Birth of daughter: 14 Feb 1902, Mary Elizabeth Scoggins, St. Louis, MO.
Birth of son: 1 Mar 1905, George Lennard Scoggins, St. Louis, MO.
Birth of son: Bet. 1907 and 1908, John Scoggins, St. Louis, MO, Parentage in doubt.3
Married Name: Mary Adelaide Linden (Stout).
Married Name: Mary Adelaide Vaughn (Stout).
Married Name: Mary Adelaide Scroggins (Stout).
Children of Mary Adelaide Stout & Archibald Johnson Scroggins Jr.
Frank Clarence Scoggins+ (aft. 1879? - )
William 'Hap' Scoggins Sr.+ (21 Jul 1891 - 22 Feb 1936)
Ruth Belle Scoggins+ (6 Dec 1897 - 6 Jul 1967)
Mary Elizabeth Scoggins+ (14 Feb 1902 - 28 Aug 1983)
George Lennard Scoggins+ (1 Mar 1905 - 8 Jun 1974)
John Scoggins (bet. 1907 and 1908 - )
Last Edited: 20 Sep 2004
[S61] Unknown subject Forum message at genforum.genealogy.com, Stout Family, Message 3444, Greg Myers <e-mail address>, "Stout Indian Heritage via Spencer Line", 13 Sep 2004
http://genforum.genealogy.com/stout/messages/3444.html
[S109] Gregory Myers, Greg Myers' Scoggins/Scroggins Family Page, Url: http://www.members.tripod.com/-greg.
Compiler: John M. Scroggins
Note: The software used to maintain the database and these pages is no longer supported. The next (final) update will include instructions on finding replacement pages built with new software. on 28 Jun 2017 at 8:01:11 AM; 46,382 people
Page created by John Cardinal's Second Site v6.2.11. | Web site design by John Cardinal, modified by John M. Scroggins, 3 Jul 2007, 21 Sep 2013
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 5,635 |
[
128000,
10254,
2833,
311,
578,
2522,
12184,
70,
1354,
45231,
555,
12656,
480,
13,
2522,
12184,
70,
1354,
198,
9290,
25,
1115,
374,
279,
1828,
311,
279,
1566,
2713,
315,
420,
2199,
520,
420,
2576,
13,
3580,
810,
3770,
627,
42584,
50301,
92130,
198,
42584,
50301,
92130,
16,
198,
37,
11,
674,
24531,
24,
11,
320,
66,
220,
9674,
15,
482,
220,
5926,
23,
340,
62416,
81203,
12656,
92130,
198,
59978,
81203,
10455,
35336,
36221,
320,
66,
220,
10750,
24,
482,
1763,
9032,
25,
510,
24531,
24,
70956,
933,
32578,
81203,
14751,
13,
220,
9674,
15,
11,
69829,
3623,
11,
11672,
13,
17,
198,
5417,
1415,
82,
11,
10003,
315,
12656,
92130,
612,
10455,
35336,
36221,
4390,
12331,
8950,
81203,
220,
966,
3297,
220,
9367,
24,
11,
9683
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
10254,
2833,
311,
578,
2522,
12184,
70,
1354,
45231,
555,
12656,
480,
13,
2522,
12184,
70,
1354,
198,
9290,
25,
1115,
374,
279,
1828,
311,
279,
1566,
2713,
315,
420,
2199,
520,
420,
2576,
13,
3580,
810,
3770,
627,
42584,
50301,
92130,
198,
42584,
50301,
92130,
16,
198,
37,
11,
674,
24531,
24,
11,
320,
66,
220,
9674,
15,
482,
220,
5926,
23,
340,
62416,
81203,
12656,
92130,
198,
59978,
81203,
10455,
35336,
36221,
320,
66,
220,
10750,
24,
482,
1763,
9032,
25,
510,
24531,
24,
70956,
933,
32578,
81203,
14751,
13,
220,
9674,
15,
11,
69829,
3623,
11,
11672,
13,
17,
198,
5417,
1415,
82,
11,
10003,
315,
12656,
92130,
612,
10455,
35336,
36221,
4390,
12331,
8950,
81203,
220,
966,
3297,
220,
9367,
24,
11,
9683,
-100
] |
New stop signs in Downtown Jersey City aren't helping, residents say
Updated: Jan. 17, 2019, 4:34 p.m. | Published: Dec. 09, 2013, 8:23 p.m.
By Terrence T. McDonald | The Jersey Journal
new-stop-sign-9th-jersey.JPG
A new stop sign at Jersey Avenue and Ninth Street in Jersey City is partly obscured by an existing bus stop sign.
(Terrence T. McDonald/The Jersey Journal)
Jersey City has installed two dozen new stop signs at intersections throughout the Downtown, but residents say some are too close together, others are partially obscured by existing signs and there was no notice given that they would suddenly appear.
Even parents picking up their children at city schools today gave the new signs a thumbs down.
"Nobody's paying attention," said Deanna Day, who was waiting for her 10-year-old grandson outside School 16, where the city has installed two new stop signs for drivers on Washington Street.
Anna Rowley, outside the school to pick up her nearly 6-year-old son, agreed.
"People fly through here," Rowley said.
The sentiment was the same in the Hamilton Park area, where six new stops signs have appeared. Andrea Gaines, of West Hamilton Place, noted that one new sign, at West Hamilton Place and Ninth Street, is less than 300 feet away from another new one, at Jersey Avenue and Ninth Street.
And that sign, like one of the new ones at Washington and Sussex streets, is obscured by a bus stop sign.
"You can't see it," said Gaines, who admitted she accidentally blew through the intersection one day before she realized it was there.
On Tuesday morning, the day after The Jersey Journal published this story on nj.com, the stop sign at Ninth and Jersey had been moved in front of the bus stop sign instead of behind it.
The new signs are part of an initiative spearheaded by Ward E City Councilman Candice Osborne, who made pedestrian and traffic safety a priority after she was elected to her first term in May.
Most of the new signs were installed near Hamilton and Jones parks, with four more in the Paulus Hook neighborhood and three near Van Vorst Park.
Osborne said all drivers will not automatically stop at a new stop sign if they're used to the area.
"As drivers, we probably get into a little bit of autopilot," she said. "And if you've never seen one there, you're definitely not looking for it."
City crews are "working" on making some of the new signs more visible, she added. And once the weather is a little bit warmer, crews will paint "stop" onto streets just in front of the intersections with new signs, she said.
"That's definitely coming," Osborne said.
There are also two new speed humps, one on Ninth Street and the other Eighth Street, though they are low enough to the ground that cars can fly over them while barely slowing down.
State laws used to forbid municipalities from installing stop signs without prior state approval, but that was changed years ago for streets that are contained in only one municipality.
The councilwoman said she pushed for as many stop signs as she could, in addition to new no-turn-on-red signs on Grand Street and longer periods for pedestrians to cross Marin Boulevard.
"We're going to take another look next year to see what's working, what's not working," she said.
For Gaines, six new stop signs just around Hamilton Park is too much.
"Definitely 'A' for effort," she said, "but it should be a little more measured."
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 8,275 |
[
128000,
3648,
3009,
12195,
304,
44985,
16228,
4409,
7784,
956,
10695,
11,
11062,
2019,
198,
16593,
25,
4448,
13,
220,
1114,
11,
220,
679,
24,
11,
220,
19,
25,
1958,
281,
749,
13,
765,
30114,
25,
3799,
13,
220,
2545,
11,
220,
679,
18,
11,
220,
23,
25,
1419,
281,
749,
627,
1383,
10335,
16271,
350,
13,
32014,
765,
578,
16228,
10139,
198,
943,
42311,
29053,
12,
24,
339,
13636,
261,
8674,
65921,
198,
32,
502,
3009,
1879,
520,
16228,
17569,
323,
86858,
6825,
304,
16228,
4409,
374,
28135,
96801,
555,
459,
6484,
5951,
3009,
1879,
627,
4233,
261,
16271,
350,
13,
32014,
48848,
16228,
10139,
340,
45057,
8674,
4409,
706,
10487,
1403,
21030,
502,
3009,
12195,
520,
65357,
6957,
279,
44985,
11,
719,
11062,
2019,
1063,
527
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
3648,
3009,
12195,
304,
44985,
16228,
4409,
7784,
956,
10695,
11,
11062,
2019,
198,
16593,
25,
4448,
13,
220,
1114,
11,
220,
679,
24,
11,
220,
19,
25,
1958,
281,
749,
13,
765,
30114,
25,
3799,
13,
220,
2545,
11,
220,
679,
18,
11,
220,
23,
25,
1419,
281,
749,
627,
1383,
10335,
16271,
350,
13,
32014,
765,
578,
16228,
10139,
198,
943,
42311,
29053,
12,
24,
339,
13636,
261,
8674,
65921,
198,
32,
502,
3009,
1879,
520,
16228,
17569,
323,
86858,
6825,
304,
16228,
4409,
374,
28135,
96801,
555,
459,
6484,
5951,
3009,
1879,
627,
4233,
261,
16271,
350,
13,
32014,
48848,
16228,
10139,
340,
45057,
8674,
4409,
706,
10487,
1403,
21030,
502,
3009,
12195,
520,
65357,
6957,
279,
44985,
11,
719,
11062,
2019,
1063,
527,
-100
] |
Artiés y Garós es un tersón tradicional del Valle de Arán, (Cataluña, España) utilizado como circunscripción territorial para las elecciones al Consejo General de Arán. Resultado de la división del antiguo tercio de Garòs, su territorio se corresponde con la parte sur del municipio del Alto Arán: Artiés y Garós.
Elecciones
Desde la restauración de la estructura administrativa tradicional del Valle de Arán de 1990, elige 2 de los 13 Consejeros del Conselh Generau d'Aran.
Referencias
Tercios del Valle de Arán
|
{
"redpajama_set_name": "RedPajamaWikipedia"
}
| 360 |
[
128000,
9470,
72,
5512,
379,
12471,
29832,
1560,
653,
88730,
3244,
4790,
63893,
1624,
78594,
409,
1676,
11644,
11,
320,
34,
4306,
84,
18829,
11,
57608,
8,
99633,
8112,
4319,
11099,
745,
44428,
52482,
3429,
5252,
10732,
55803,
453,
67566,
7453,
3331,
409,
1676,
11644,
13,
5832,
2172,
409,
1208,
50468,
3739,
1624,
68937,
24012,
2024,
61930,
409,
12471,
22980,
82,
11,
924,
40456,
822,
513,
85164,
76149,
390,
1208,
20108,
1765,
1624,
20188,
822,
1624,
54370,
1676,
11644,
25,
5277,
72,
5512,
379,
12471,
29832,
382,
46439,
55803,
720,
85332,
1208,
7696,
5840,
409,
1208,
76509,
5808,
11398,
28952,
4790,
63893,
1624,
78594,
409,
1676,
11644,
409,
220,
2550,
15,
11,
658,
7404,
220,
17,
409,
2537,
220,
1032,
67566,
73,
6398,
1624,
7440,
301,
71,
2672
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
9470,
72,
5512,
379,
12471,
29832,
1560,
653,
88730,
3244,
4790,
63893,
1624,
78594,
409,
1676,
11644,
11,
320,
34,
4306,
84,
18829,
11,
57608,
8,
99633,
8112,
4319,
11099,
745,
44428,
52482,
3429,
5252,
10732,
55803,
453,
67566,
7453,
3331,
409,
1676,
11644,
13,
5832,
2172,
409,
1208,
50468,
3739,
1624,
68937,
24012,
2024,
61930,
409,
12471,
22980,
82,
11,
924,
40456,
822,
513,
85164,
76149,
390,
1208,
20108,
1765,
1624,
20188,
822,
1624,
54370,
1676,
11644,
25,
5277,
72,
5512,
379,
12471,
29832,
382,
46439,
55803,
720,
85332,
1208,
7696,
5840,
409,
1208,
76509,
5808,
11398,
28952,
4790,
63893,
1624,
78594,
409,
1676,
11644,
409,
220,
2550,
15,
11,
658,
7404,
220,
17,
409,
2537,
220,
1032,
67566,
73,
6398,
1624,
7440,
301,
71,
2672,
-100
] |
Decision Sciences & Systems
Research Transfer & Software
Prof. Dr. Felix Brandt
Florian Brandl, M.Sc., Christian Stricker, M.Sc.
Seminar SS 2018
Economics and Computation
In recent years, there has been an increasing interest in topics at the intersection of economics and computer science, as witnessed by the continued rapid rise of research areas such as algorithmic game theory and computational social choice. This development is due to the emergence of computational networks such as the Internet as well as the need to get a grip on algorithmic questions in economics.
The emphasis in this seminar lies on the independent study of classic economics papers, but also, and in particular, more recent papers from computer science. Among the topics to be covered are matching theory, mechanism design, and voting theory.
Registration (*please read carefully*)
Seats are split up between students from computer science and students from other programs. All interested students should attend the overview meeting and have to send an application by email briefly describing their background (including relevant courses) and motivation (up to 250 words) as well as 2-5 papers they are interested in. Deadline for applications: January 28, 2018 (11:59pm). Students from computer science additionally have to use the matching system for indication of interest. Notifications will be sent out on February 1 (mathematics) and Februrary 21 (computer science) and include assignment of papers and supervisors. Registration in TUMonline will be taken care of by the lecturers by end of February; no further action is required.
Time and venue
Overview (Vorbesprechung):
Wednesday, January 24, 13.00 - 14.00 , room 01.10.011 (slides)
First meeting:
Wednesday, April 18, 13.00 - 14.00, room 01.10.036
Talks:
Wednesday, May 9, 9.00 - 17.00, room 01.10.033
Friday, June 8, 9.00 - 17.00, room 01.10.033
Preliminary selection of articles
Algorithmic Game Theory
Mechanism Design
Randomized Social Choice
A. Bogomolnaia and M. O. Jackson. The stability of hedonic coalition structures. Games and Economic Behavior, 38(2):201–230, 2002.
X. Deng and C. H. Papadimitriou. On the complexity of cooperative solution concepts. Mathematics of Operations Research, 12(2):257–266, 1994.
D. C. Fisher and J. Ryan. Optimal strategies for a generalized "scissors, paper, and stone" game. American Mathematical Monthly, 99(10):935–942, 1992.
J. Hajdukovà. Coalition formation games: A survey. International Game Theory Review, 8(4):613–641, 2006.
J. A. Kroll, I. C. Davey, and E. W. Felten. The economics of Bitcoin mining, or Bitcoin in the presence of adversaries. In Proceedings of the 12th Workshop on the Economics of Information Security (WEIS 2013), 2013.
D. Monderer and L. S. Shapley. Potential games. Games and Economic Behavior, 14(1):124–143, 1996.
A. Abdulkadiroglu and T. Sönmez. House allocation with existing tenants. Journal of Economic Theory, 88(2):233–260, 1999.
A. Abdulkadiroglu and T. Sönmez. School choice: A mechanism design approach. American Economic Review, 93(3):729—747, 2003.
M. Balinski and T. Sönmez. A tale of two mechanisms: Student placement. Journal of Economic Theory, 84(1):73–94, 1999.
S. J. Brams and A. D. Taylor. An envy-free cake division protocol. The American Mathematical Monthly, 102(1):9–18, 1995.
U. Endriss, N. Maudet, F. Sadri, and F. Toni. Negotiating socially optimal allocations of resources. Journal of Artificial Intelligence Research, 25:315–348, 2006.
R. W. Irving. An efficient algorithm for the "stable roommates" problem. Journal of Algorithms, 6(4):577–595, 1985.
M. O. Jackson and A. Wolinsky. A strategic model of social and economic networks. Journal of Economic Theory, 71(1):44–74, 1996.
E. Maskin. Nash equilibrium and welfare optimality. Review of Economic Studies, 66(26):23–38, 1999.
N. Nisan and A. Ronen. Algorithmic mechanism design. Games and Economic Behavior, 35(1):166–196, 2001.
A. D. Procaccia and J. Wang. Fair enough: Guaranteeing approximate maximin shares. In Proceedings of the 15th ACM Conference on Economics and Computation (ACM-EC), pages 675–692, 2014.
T. Roughgarden and É. Tardos. How bad is selfish routing? Journal of the ACM, 49(2):236–259, 2002.
T. Sandholm. Algorithm for optimal winner determination in combinatorial auctions. Artificial Intelligence, 135(1–2):1–54, 2002.
J. Bartholdi, III, C. A. Tovey, and M. A. Trick. The computational difficulty of manipulating an election. Social Choice and Welfare, 6(3):227–241, 1989.
F. Brandt and C. Geist. Finding strategyproof social choice functions via SAT solving. Journal of Artificial Intelligence Research, 55:565–602, 2016.
V. Conitzer, T. Sandholm, and J. Lang. When are elections with few candidates hard to manipulate? Journal of the ACM, 54(3), 2007.
J. Duggan and T. Schwartz. Strategic manipulability without resoluteness or shared beliefs: Gibbard- Satterthwaite generalized. Social Choice and Welfare, 17(1):85–93, 2000.
P. Tang and F. Lin. Computer-aided proofs of Arrow's and other impossibility theorems. Artificial Intelligence, 173(11):1041–1053, 2009.
H. P. Young. Optimal voting rules. Journal of Economic Perspectives, 9(1):51–64, 1995.
S. Barberà. Majority and positional voting in a probabilistic framework. Review of Economic Studies, 46(2):379–389, 1979.
A. Bogomolnaia and H. Moulin. A new solution to the random assignment problem. Journal of Economic Theory, 100(2):295–328, 2001.
P. C. Fishburn. Probabilistic social choice based on simple voting comparisons. Review of Economic Studies, 51(4):683–692, 1984.
P. C. Fishburn. SSB utility theory: An economic perspective. Mathematical Social Sciences, 8(1):63–94, 1984.
A. Gibbard. Manipulation of schemes that mix voting with chance. Econometrica, 45(3):665–681, 1977.
M. D. Intriligator. A probabilistic model of social choice. Review of Economic Studies, 40(4):553–560, 1973.
Feedback guidelines (1) (2) [in addition to what is presented during the first meeting]
The seminar will be held in English (i.e., all presentations will have to be in English, too)
Module-Codes
IN2107 (Master-Seminar in the Master program Informatik)
IN0014 (Seminar in the Bachelor programs Informatik, Wirtschaftsinformatik)
For all other programs: Please check first whether this seminar fits in your curriculum. For example, mathematics students should find it listed as a mathematics seminar, too.
This email address is being protected from spambots. You need JavaScript enabled to view it., This email address is being protected from spambots. You need JavaScript enabled to view it.
Decision Sciences & Systems (DSS), Department of Informatics (I18), Technische Universität München, Boltzmannstr. 3, 85748 Garching, Germany
©2002-2019 DSS All Rights Reserved
Impressum, Privacy Policy, Copyright Information and Disclaimer
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 2,128 |
[
128000,
75946,
23199,
612,
15264,
198,
28528,
24078,
612,
4476,
198,
19105,
13,
2999,
13,
57023,
16835,
83,
198,
3968,
22865,
16835,
75,
11,
386,
18832,
2637,
9052,
4610,
5327,
11,
386,
18832,
627,
30599,
14080,
18679,
220,
679,
23,
198,
36,
81192,
323,
1219,
13623,
198,
644,
3293,
1667,
11,
1070,
706,
1027,
459,
7859,
2802,
304,
13650,
520,
279,
19801,
315,
28989,
323,
6500,
8198,
11,
439,
32126,
555,
279,
8738,
11295,
10205,
315,
3495,
5789,
1778,
439,
12384,
292,
1847,
10334,
323,
55580,
3674,
5873,
13,
1115,
4500,
374,
4245,
311,
279,
49179,
315,
55580,
14488,
1778,
439,
279,
8191,
439,
1664,
439,
279,
1205,
311,
636,
264,
25703,
389,
12384,
292,
4860,
304,
28989,
627,
791,
25679,
304,
420,
54675,
15812,
389,
279,
9678
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
75946,
23199,
612,
15264,
198,
28528,
24078,
612,
4476,
198,
19105,
13,
2999,
13,
57023,
16835,
83,
198,
3968,
22865,
16835,
75,
11,
386,
18832,
2637,
9052,
4610,
5327,
11,
386,
18832,
627,
30599,
14080,
18679,
220,
679,
23,
198,
36,
81192,
323,
1219,
13623,
198,
644,
3293,
1667,
11,
1070,
706,
1027,
459,
7859,
2802,
304,
13650,
520,
279,
19801,
315,
28989,
323,
6500,
8198,
11,
439,
32126,
555,
279,
8738,
11295,
10205,
315,
3495,
5789,
1778,
439,
12384,
292,
1847,
10334,
323,
55580,
3674,
5873,
13,
1115,
4500,
374,
4245,
311,
279,
49179,
315,
55580,
14488,
1778,
439,
279,
8191,
439,
1664,
439,
279,
1205,
311,
636,
264,
25703,
389,
12384,
292,
4860,
304,
28989,
627,
791,
25679,
304,
420,
54675,
15812,
389,
279,
9678,
-100
] |
The increasing amount of information available online has led to the development of technologies that help to deal with it. One of them is Interactive Question Answering (IQA), a research field that has emerged at the intersection of question answering and dialogue systems, and which allows users to find the answers to questions in an interactive way. During the answering process, the automatic system can initiate a dialogue with the user in order to clarify missing or ambiguous information, or suggest further topics for discussion. This chapter presents the state-of-the-art in the field of interactive question answering. Given that IQA inherits a lot of features from dialogue systems and question answering, these fields are also briefly presented. Analysis of the existing systems reveals that in general IQA systems rely on a scaled-down version of a dialogue system, sometimes built on top of question answering systems. Evaluation of IQA is also discussed, showing that it combines evaluation techniques from question answering and dialogue systems.
A computational device or agent that (a) engages in interaction with other human and/or computer participant(s); (b) uses human language in some form such as speech, text, or sign; and (c) typically engages in such interaction across multiple turns or sentences (http://www.jods.org/).
This definition highlights several important aspects of such systems. A dialogue system always has a user, who interacts with the system for a specific goal such as completing some tasks. The interaction involves a conversation in human language between two or more participants and can take several turns. The fact that human language is used in the interaction differentiates the field from other fields such as database access using computer languages (such as SQL) or interaction between software agents that communicate using XML or some other standard computer formats.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 4,871 |
[
128000,
791,
7859,
3392,
315,
2038,
2561,
2930,
706,
6197,
311,
279,
4500,
315,
14645,
430,
1520,
311,
3568,
449,
433,
13,
3861,
315,
1124,
374,
42966,
16225,
22559,
287,
320,
40,
48622,
705,
264,
3495,
2115,
430,
706,
22763,
520,
279,
19801,
315,
3488,
36864,
323,
21976,
6067,
11,
323,
902,
6276,
3932,
311,
1505,
279,
11503,
311,
4860,
304,
459,
21416,
1648,
13,
12220,
279,
36864,
1920,
11,
279,
17392,
1887,
649,
39201,
264,
21976,
449,
279,
1217,
304,
2015,
311,
38263,
7554,
477,
55861,
2038,
11,
477,
4284,
4726,
13650,
369,
10430,
13,
1115,
12735,
18911,
279,
1614,
8838,
10826,
38921,
304,
279,
2115,
315,
21416,
3488,
36864,
13,
16644,
430,
37811,
32,
76582,
264,
2763,
315,
4519,
505,
21976,
6067,
323,
3488,
36864,
11
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
791,
7859,
3392,
315,
2038,
2561,
2930,
706,
6197,
311,
279,
4500,
315,
14645,
430,
1520,
311,
3568,
449,
433,
13,
3861,
315,
1124,
374,
42966,
16225,
22559,
287,
320,
40,
48622,
705,
264,
3495,
2115,
430,
706,
22763,
520,
279,
19801,
315,
3488,
36864,
323,
21976,
6067,
11,
323,
902,
6276,
3932,
311,
1505,
279,
11503,
311,
4860,
304,
459,
21416,
1648,
13,
12220,
279,
36864,
1920,
11,
279,
17392,
1887,
649,
39201,
264,
21976,
449,
279,
1217,
304,
2015,
311,
38263,
7554,
477,
55861,
2038,
11,
477,
4284,
4726,
13650,
369,
10430,
13,
1115,
12735,
18911,
279,
1614,
8838,
10826,
38921,
304,
279,
2115,
315,
21416,
3488,
36864,
13,
16644,
430,
37811,
32,
76582,
264,
2763,
315,
4519,
505,
21976,
6067,
323,
3488,
36864,
11,
-100
] |
Abstract : During pregnancy the maternal immune system has to develop tolerance towards the developing fetus. These changes in maternal immunity can result in increased severity of certain infections, but also in amelioration of autoimmune diseases. Pregnancy-related hormones have been suggested to play a central role in the adaptation of the maternal immune system, but their specific effects on innate immune function is not well understood. In a longitudinal study of pregnant women, we investigated innate immune cell function in response to toll-like receptors (TLR) 4 and 7 stimulation, two TLR pathways playing a critical role in early innate immune recognition of bacteria and viruses. IFNα production by TLR7-stimulated pDCs was decreased in early pregnancy, and increased towards the end of pregnancy. In contrast, pro-inflammatory TLR4-induced TNFα production by monocytes was increased during early pregnancy, but declined after the first trimester. Changes in cytokine production were associated with changes in pregnancy-related hormones and monocyte subpopulations over the course of pregnancy. These data demonstrating a significant association between pregnancy-related hormones and modulation of innate immune responses mediated by TLRs provide novel insights into the immunological adaptations occurring during pregnancy.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 9,840 |
[
128000,
16328,
551,
12220,
20209,
279,
50150,
22852,
1887,
706,
311,
2274,
25065,
7119,
279,
11469,
76688,
13,
4314,
4442,
304,
50150,
40368,
649,
1121,
304,
7319,
31020,
315,
3738,
30020,
11,
719,
1101,
304,
126641,
2521,
367,
315,
88191,
19338,
13,
83627,
14228,
44315,
617,
1027,
12090,
311,
1514,
264,
8792,
3560,
304,
279,
34185,
315,
279,
50150,
22852,
1887,
11,
719,
872,
3230,
6372,
389,
65070,
22852,
734,
374,
539,
1664,
16365,
13,
763,
264,
68102,
4007,
315,
20895,
3278,
11,
584,
27313,
65070,
22852,
2849,
734,
304,
2077,
311,
26936,
12970,
44540,
320,
13778,
49,
8,
220,
19,
323,
220,
22,
41959,
11,
1403,
350,
20721,
44014,
5737,
264,
9200,
3560,
304,
4216,
65070,
22852,
18324,
315,
24032,
323,
42068,
13,
11812,
45,
19481,
5788
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
16328,
551,
12220,
20209,
279,
50150,
22852,
1887,
706,
311,
2274,
25065,
7119,
279,
11469,
76688,
13,
4314,
4442,
304,
50150,
40368,
649,
1121,
304,
7319,
31020,
315,
3738,
30020,
11,
719,
1101,
304,
126641,
2521,
367,
315,
88191,
19338,
13,
83627,
14228,
44315,
617,
1027,
12090,
311,
1514,
264,
8792,
3560,
304,
279,
34185,
315,
279,
50150,
22852,
1887,
11,
719,
872,
3230,
6372,
389,
65070,
22852,
734,
374,
539,
1664,
16365,
13,
763,
264,
68102,
4007,
315,
20895,
3278,
11,
584,
27313,
65070,
22852,
2849,
734,
304,
2077,
311,
26936,
12970,
44540,
320,
13778,
49,
8,
220,
19,
323,
220,
22,
41959,
11,
1403,
350,
20721,
44014,
5737,
264,
9200,
3560,
304,
4216,
65070,
22852,
18324,
315,
24032,
323,
42068,
13,
11812,
45,
19481,
5788,
-100
] |
ConfometRx is developing a platform of structure-based drug development tools to facilitate target validation, lead identification, and lead optimization for G protein coupled receptor (GPCR) targets. This platform includes: the production of pure, functional GPCRs for high-resolution structure determination by crystallography, and the engineering of GPCRs to facilitate crystal formation; the efficient and economical expression of GPCRs for NMR-based drug discovery applications; and the generation of receptor-specific, functional antibodies for target validation, therapeutics and GPCR crystallography.
The workshop goal is to bring together scientists to exchange ideas and discuss challenges related to drug development for G-protein coupled receptors in order to facilitate the discovery of more effective and selective therapeutics for unmet medical needs.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 2,168 |
[
128000,
15949,
4512,
51739,
374,
11469,
264,
5452,
315,
6070,
6108,
5623,
4500,
7526,
311,
28696,
2218,
10741,
11,
3063,
22654,
11,
323,
3063,
26329,
369,
480,
13128,
34356,
35268,
320,
70341,
49,
8,
11811,
13,
1115,
5452,
5764,
25,
279,
5788,
315,
10748,
11,
16003,
480,
4977,
43427,
369,
1579,
64036,
6070,
26314,
555,
26110,
848,
1976,
88,
11,
323,
279,
15009,
315,
480,
4977,
43427,
311,
28696,
26110,
18488,
26,
279,
11297,
323,
60618,
7645,
315,
480,
4977,
43427,
369,
452,
18953,
6108,
5623,
18841,
8522,
26,
323,
279,
9659,
315,
35268,
19440,
11,
16003,
59854,
369,
2218,
10741,
11,
9139,
88886,
323,
480,
74256,
26110,
848,
1976,
88,
627,
791,
26129,
5915,
374,
311,
4546,
3871,
14248,
311,
9473,
6848,
323,
4358,
11774,
5552,
311
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
15949,
4512,
51739,
374,
11469,
264,
5452,
315,
6070,
6108,
5623,
4500,
7526,
311,
28696,
2218,
10741,
11,
3063,
22654,
11,
323,
3063,
26329,
369,
480,
13128,
34356,
35268,
320,
70341,
49,
8,
11811,
13,
1115,
5452,
5764,
25,
279,
5788,
315,
10748,
11,
16003,
480,
4977,
43427,
369,
1579,
64036,
6070,
26314,
555,
26110,
848,
1976,
88,
11,
323,
279,
15009,
315,
480,
4977,
43427,
311,
28696,
26110,
18488,
26,
279,
11297,
323,
60618,
7645,
315,
480,
4977,
43427,
369,
452,
18953,
6108,
5623,
18841,
8522,
26,
323,
279,
9659,
315,
35268,
19440,
11,
16003,
59854,
369,
2218,
10741,
11,
9139,
88886,
323,
480,
74256,
26110,
848,
1976,
88,
627,
791,
26129,
5915,
374,
311,
4546,
3871,
14248,
311,
9473,
6848,
323,
4358,
11774,
5552,
311,
-100
] |
Diary We weren't really planning to go to Eldoret, but it just happened to be on the way to the Mount Elgon border with Uganda. Clare had been here when she was about 16 in 1898 and Queen Victoria was on the throne and a very different town it was then! In fact, she couldn't really remember anything about it at all. We stayed at a campsite owned by a real Asian/Kenyan character called Raj. He also helps his family run the Ken-Knit factory (school uniform jumpers and blankets) in town which employs 1600 people in a factory full of British machines from the '60's. We arrived at his "new" campsite about 20kms out of town when he had some of his family over from London to inspect his new resort. It is, without doubt, the best place we've camped on this trip so far. A little bizarre having a conversation with his sister after lunch, where she said she was thinking of emigrating from England to South Africa as "the schools are about a third of the price and England just has too many immigrants these days" Raj knew someone who had known Clare's sister's father in law many years ago and so before we knew it we had been invited to afternoon tea with a white colonial lady complete with aged black male retainer in a nylon pinny pushing a tea trolley with squeaky wheels with tea and cake at 4pm. What century have I landed in?? PIC SHOWS: KEN-KNIT FACTORY.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 8,132 |
[
128000,
22427,
661,
1226,
15058,
956,
2216,
9293,
311,
733,
311,
27965,
18673,
11,
719,
433,
1120,
7077,
311,
387,
389,
279,
1648,
311,
279,
10640,
4072,
11932,
3973,
449,
58149,
13,
62140,
1047,
1027,
1618,
994,
1364,
574,
922,
220,
845,
304,
220,
9378,
23,
323,
16657,
23225,
574,
389,
279,
44721,
323,
264,
1633,
2204,
6424,
433,
574,
1243,
0,
763,
2144,
11,
1364,
7846,
956,
2216,
6227,
4205,
922,
433,
520,
682,
13,
1226,
20186,
520,
264,
3190,
9703,
13234,
555,
264,
1972,
14875,
33954,
268,
8503,
3752,
2663,
26291,
13,
1283,
1101,
8779,
813,
3070,
1629,
279,
14594,
16222,
45168,
8803,
320,
22606,
14113,
7940,
388,
323,
65712,
8,
304,
6424,
902,
51242,
220,
6330,
15,
1274,
304,
264,
8803,
2539,
315,
8013,
12933
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
22427,
661,
1226,
15058,
956,
2216,
9293,
311,
733,
311,
27965,
18673,
11,
719,
433,
1120,
7077,
311,
387,
389,
279,
1648,
311,
279,
10640,
4072,
11932,
3973,
449,
58149,
13,
62140,
1047,
1027,
1618,
994,
1364,
574,
922,
220,
845,
304,
220,
9378,
23,
323,
16657,
23225,
574,
389,
279,
44721,
323,
264,
1633,
2204,
6424,
433,
574,
1243,
0,
763,
2144,
11,
1364,
7846,
956,
2216,
6227,
4205,
922,
433,
520,
682,
13,
1226,
20186,
520,
264,
3190,
9703,
13234,
555,
264,
1972,
14875,
33954,
268,
8503,
3752,
2663,
26291,
13,
1283,
1101,
8779,
813,
3070,
1629,
279,
14594,
16222,
45168,
8803,
320,
22606,
14113,
7940,
388,
323,
65712,
8,
304,
6424,
902,
51242,
220,
6330,
15,
1274,
304,
264,
8803,
2539,
315,
8013,
12933,
-100
] |
"This mix is just a taste of how my musical style has evolved over the years, with a carefully curated selection of old and new deep house tracks. I think this is a true reflection of my current passion for house music. Sit back, relax and enjoy."
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 1,472 |
[
128000,
22244,
6651,
374,
1120,
264,
12945,
315,
1268,
856,
18273,
1742,
706,
28995,
927,
279,
1667,
11,
449,
264,
15884,
58732,
6727,
315,
2362,
323,
502,
5655,
3838,
14242,
13,
358,
1781,
420,
374,
264,
837,
22599,
315,
856,
1510,
11939,
369,
3838,
4731,
13,
29508,
1203,
11,
12234,
323,
4774,
1210,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
] |
[
22244,
6651,
374,
1120,
264,
12945,
315,
1268,
856,
18273,
1742,
706,
28995,
927,
279,
1667,
11,
449,
264,
15884,
58732,
6727,
315,
2362,
323,
502,
5655,
3838,
14242,
13,
358,
1781,
420,
374,
264,
837,
22599,
315,
856,
1510,
11939,
369,
3838,
4731,
13,
29508,
1203,
11,
12234,
323,
4774,
1210,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100
] |
using System;
using System.Diagnostics;
using System.Diagnostics.CodeAnalysis;
using System.Reflection;
using System.ServiceProcess;
using System.Threading;
namespace SimpleNet.ServiceHost.Helpers
{
public static class ServiceRunner
{
[ExcludeFromCodeCoverage]
public static void RunInteractive(ServiceBase[] servicesToRun)
{
Console.WriteLine(@"Services running in interactive mode.");
Console.WriteLine();
MethodInfo onStartMethod = typeof(ServiceBase).GetMethod("OnStart",
BindingFlags.Instance | BindingFlags.NonPublic);
foreach (ServiceBase service in servicesToRun)
{
try
{
Trace.TraceInformation(@"Starting {0}...", service.ServiceName);
onStartMethod.Invoke(service, new object[] { new string[] { } });
Trace.TraceInformation(@"Starting {0}...... Started", service.ServiceName);
}
catch (Exception ex)
{
Trace.TraceError(@"Starting {0}...... FAILED!!{1}{1}{2}{1}", service.ServiceName, Environment.NewLine, ex);
}
}
Console.WriteLine();
Console.WriteLine();
Console.WriteLine(@"Press any key to stop the services and end the process...");
Console.Read();
Console.WriteLine();
MethodInfo onStopMethod = typeof(ServiceBase).GetMethod("OnStop", BindingFlags.Instance | BindingFlags.NonPublic);
foreach (ServiceBase service in servicesToRun)
{
Trace.Write( String.Format( @"Stopping {0}...", service.ServiceName ) );
onStopMethod.Invoke(service, null);
Trace.WriteLine(@"Stopped");
}
Console.WriteLine(@"All services stopped.");
// Keep the console alive for a second to allow the user to see the message.
Thread.Sleep(1000);
}
}
}
|
{
"redpajama_set_name": "RedPajamaGithub"
}
| 3,595 |
[
128000,
985,
744,
280,
985,
744,
22202,
280,
985,
744,
22202,
59605,
280,
985,
744,
26096,
280,
985,
744,
14181,
7575,
280,
985,
744,
7653,
401,
2280,
9170,
7099,
14181,
9480,
48751,
198,
517,
262,
586,
1118,
538,
5475,
20051,
198,
262,
341,
286,
510,
96339,
3915,
2123,
67412,
933,
286,
586,
1118,
742,
6588,
72786,
63774,
4066,
1318,
3600,
1271,
6869,
340,
286,
341,
310,
5390,
6158,
11181,
11271,
4401,
304,
21416,
3941,
7470,
310,
5390,
6158,
1454,
310,
24776,
58213,
3607,
284,
7960,
63774,
4066,
570,
86173,
446,
1966,
3563,
761,
394,
59778,
12977,
765,
59778,
31806,
12965,
317,
310,
4133,
320,
1898,
4066,
2532,
304,
3600,
1271,
6869,
340,
310,
341,
394,
1456,
198,
394,
341,
504,
28262,
48020,
15218,
11181,
25639,
314,
15,
92
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
985,
744,
280,
985,
744,
22202,
280,
985,
744,
22202,
59605,
280,
985,
744,
26096,
280,
985,
744,
14181,
7575,
280,
985,
744,
7653,
401,
2280,
9170,
7099,
14181,
9480,
48751,
198,
517,
262,
586,
1118,
538,
5475,
20051,
198,
262,
341,
286,
510,
96339,
3915,
2123,
67412,
933,
286,
586,
1118,
742,
6588,
72786,
63774,
4066,
1318,
3600,
1271,
6869,
340,
286,
341,
310,
5390,
6158,
11181,
11271,
4401,
304,
21416,
3941,
7470,
310,
5390,
6158,
1454,
310,
24776,
58213,
3607,
284,
7960,
63774,
4066,
570,
86173,
446,
1966,
3563,
761,
394,
59778,
12977,
765,
59778,
31806,
12965,
317,
310,
4133,
320,
1898,
4066,
2532,
304,
3600,
1271,
6869,
340,
310,
341,
394,
1456,
198,
394,
341,
504,
28262,
48020,
15218,
11181,
25639,
314,
15,
92,
-100
] |
Silenen (toponimo tedesco; in romancio Silauna, desueto) è un comune svizzero di 2 045 abitanti del Canton Uri.
Geografia fisica
Storia
Monumenti e luoghi d'interesse
Chiesa parrocchiale cattolica di Sant'Albino, eretta nel VII secolo, ricostruita nel 1754-1756.
Società
Evoluzione demografica
L'evoluzione demografica è riportata nella seguente tabella (fino al 1799 con Gurtnellen):
Amministrazione
Ogni famiglia originaria del luogo fa parte del cosiddetto comune patriziale e ha la responsabilità della manutenzione di ogni bene ricadente all'interno dei confini del comune.
Note
Altri progetti
Collegamenti esterni
|
{
"redpajama_set_name": "RedPajamaWikipedia"
}
| 6,713 |
[
128000,
50,
23684,
268,
320,
3565,
263,
11620,
42323,
94934,
26,
304,
10083,
1149,
822,
8211,
64,
8733,
11,
951,
84,
11269,
8,
11676,
653,
470,
2957,
13871,
8934,
2382,
1891,
220,
17,
4194,
23785,
671,
275,
15719,
1624,
73466,
17670,
382,
9688,
50901,
42723,
3074,
271,
626,
11015,
271,
11342,
1143,
72,
384,
25774,
540,
6151,
294,
6,
2295,
24201,
4815,
921,
552,
64,
1370,
28787,
331,
20487,
272,
1617,
337,
3074,
1891,
23746,
6,
2149,
65,
3394,
11,
2781,
27625,
25334,
46069,
513,
59348,
11,
94698,
496,
3159,
64,
25334,
220,
10005,
19,
12,
10005,
21,
382,
50,
2168,
95971,
271,
36,
12821,
55446,
2486,
26113,
3074,
720,
43,
6,
5230,
337,
55446,
2486,
26113,
3074,
11676,
25160,
403,
460,
37022,
82402,
6960,
5769,
6985,
320
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
50,
23684,
268,
320,
3565,
263,
11620,
42323,
94934,
26,
304,
10083,
1149,
822,
8211,
64,
8733,
11,
951,
84,
11269,
8,
11676,
653,
470,
2957,
13871,
8934,
2382,
1891,
220,
17,
4194,
23785,
671,
275,
15719,
1624,
73466,
17670,
382,
9688,
50901,
42723,
3074,
271,
626,
11015,
271,
11342,
1143,
72,
384,
25774,
540,
6151,
294,
6,
2295,
24201,
4815,
921,
552,
64,
1370,
28787,
331,
20487,
272,
1617,
337,
3074,
1891,
23746,
6,
2149,
65,
3394,
11,
2781,
27625,
25334,
46069,
513,
59348,
11,
94698,
496,
3159,
64,
25334,
220,
10005,
19,
12,
10005,
21,
382,
50,
2168,
95971,
271,
36,
12821,
55446,
2486,
26113,
3074,
720,
43,
6,
5230,
337,
55446,
2486,
26113,
3074,
11676,
25160,
403,
460,
37022,
82402,
6960,
5769,
6985,
320,
-100
] |
All Eyes on Tesla Motors Inc (TSLA) and NXP Semiconductors NV (NXPI) Earnings This Evening
Julie Lamb- October 26, 2016, 3:59 PM EDT
All eyes eagerly look to this evening's big earnings party as electric car giant Tesla Motors Inc (NASDAQ:TSLA) and chip maker NXP Semiconductors NV (NASDAQ:NXPI) are delivering third-quarter reports after market close. Oppenheimer weighs in sidelined on TSLA while offering perspective into an updated model for the company's restated financials, but decidedly more bullish on NXP Semiconductors amid the rumble of merger and acquisition chatter. Let's delve in deeper:
Tesla Motors Inc
Tesla's quarterly print is soon on its way, and investors anxiously wait to take a closer look at the company's forthcoming conversion to reporting GAAP revenue. Ahead of the print, Oppenheimer analyst Colin Rusch reiterates a Perform rating on shares of TSLA without listing a price target.
The analyst has tweaked his model to no longer include deferred revenue from lease accounting. For the third quarter, the analyst has lowered revenue from $2,327.4 million to $1,866.3 million. Adjusted gross margin will be reported based on GAAP revenue and gross profit will be altered for stock-based compensation after the reporting change. The analyst has reduced his gross margin estimate from 23.0% to 22.2%.
Rusch notes, "Given the potential confusion around TSLA's anticipated reporting changes, we are publishing an updated model to help investors navigate these adjustments. The critical change in proforma revenue is the recognition of lease revenue not total vehicle value. Due to high percentage of deferred revenue (~35%), this change also materially changes EPS. We expect cash flow estimates to remain intact."
"We also believe the new reporting structure highlights the importance of TSLA's lease partners and risks around TSLA's used vehicle market whether it is older vehicles cannibalizing new sales or the brisk pace of innovation limiting interest in older vehicles. While used Tesla's have held value well to date, we view the end of lease vehicle strategy as a critical variable for future cash needs," Rusch concludes.
As usual, we like to include the analyst's track record when reporting on new analyst notes to give a perspective on the effect it has on stock performance. According to TipRanks, four-star analyst Colin Rusch is ranked #440 out of 4,197 analysts. Rusch has a 47% success rate and gains 8.2% in his annual returns. When recommending TSLA, Rusch garners 88.9% in average profits on the stock.
TipRanks analytics exhibit TSLA as a Hold. Based on 11 analysts polled in the last 3 months, 2 rate a Buy on TSLA, 5 maintain a Hold, while 4 issue a Sell. The 12-month price target stands at $214.56, marking a 6% upside from where the shares last closed.
NXP Semiconductors NV
Ahead of NXP Semiconductors' third-quarter financial report, Oppenheimer analyst Rick Schafer remains bullish on what he deems "the leading semiconductor to the automotive market, our favorite multi-year vertical." Therefore, the analyst reiterates an Outperform rating on NXP while raising the price target from $100 to $110, which represents a 10% increase from current levels.
For the third quarter, the analyst projects EPS of $1.59. For the fourth quarter, the analyst forecasts EPS guidance of $2.02. Schafer likes the "multiple levers" the company has as leverage drivers for free cash flow (FCF) and EPS. Meanwhile, the company has guided OM of 27.5%. For the second half of 2017, the analyst expects OM to hit over 30% and for FCF to reach $2.2 billion in 2017. The company anticipates to close 2016 with over $300 million in annualized OpEx improvement, compared to its initial synergy target of $200 million.
Schafer opines, "We expect NT fundamentals to take a back seat to mgmt commentary surrounding recent reports (beginning with TheStreet.com on 9/29) that QCOM is in talks to acquire NXPI for ~$110/share. We also see NXPI as an accretive use of QCOM's nearly $30B offshore cash horde that can allow the company to diversify away from handsets. NXPI represents a relatively inexpensive non-US asset (at ~13x CY17E). We think a QCOM/NXPI deal is more likely than not. Near term shares likely react more to M&A speculation than fundamentals."
Additionally, the analyst likes NXPI's odds for a deal with QUALCOMM, Inc. (NASDAQ:QCOM). "While reports haven't been confirmed or denied by either company, we believe a QCOM/NXPI acquisition makes strategic sense. QCOM needs to diversify its ~100% mobile exposure and holds ~90% of its $31B cash stockpile offshore, making NXPI (a Dutch company) a logical target," Schafer concludes.
Long-term, the analyst expects growth of both EPS as well as FCF to be "significant."
Rick Schafer has a very good TipRanks score with a 72% success rate and he ranks at #13 out of 4,197 analysts. Schafer has a five-star rating and realizes 15.9% in his yearly returns. When suggesting NXP, Schafer yields 11.1% in average profits on the stock.
TipRanks analytics exhibit NXP as a Strong Buy. The consensus price target stands at $111.50, marking a nearly 12% increase from where the stock is currently trading.
Chip stocks face up to the grim implications of the coronavirus outbreak
Smarter Team, February 28, 2020
Bloggers CornerTechnology
This Week's Top Wall Street Upgrades: VMW, NXPI, PM
Smarter Team, October 4, 2019
Hedge Funds Pile Into Facebook (FB), PayPal (PYPL), NXPI Semiconductors (NXPI) Stocks
Harriet Lefton, August 27, 2018
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 4,001 |
[
128000,
2460,
43096,
389,
28298,
37792,
4953,
320,
51,
8143,
32,
8,
323,
452,
28475,
14582,
1965,
1076,
1105,
25464,
320,
44404,
1932,
8,
469,
15202,
1115,
57202,
198,
29185,
648,
32402,
12,
6664,
220,
1627,
11,
220,
679,
21,
11,
220,
18,
25,
2946,
5975,
49023,
198,
2460,
6548,
63473,
1427,
311,
420,
11714,
596,
2466,
24608,
4717,
439,
9249,
1841,
14880,
28298,
37792,
4953,
320,
88611,
69761,
8143,
32,
8,
323,
16797,
25214,
452,
28475,
14582,
1965,
1076,
1105,
25464,
320,
88611,
38794,
55,
1932,
8,
527,
24944,
4948,
58414,
6821,
1306,
3157,
3345,
13,
29710,
268,
35180,
50542,
304,
99048,
389,
350,
8143,
32,
1418,
10209,
13356,
1139,
459,
6177,
1646,
369,
279,
2883,
596,
2800,
660,
6020,
82,
11,
719,
77120,
810,
90221,
389
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
2460,
43096,
389,
28298,
37792,
4953,
320,
51,
8143,
32,
8,
323,
452,
28475,
14582,
1965,
1076,
1105,
25464,
320,
44404,
1932,
8,
469,
15202,
1115,
57202,
198,
29185,
648,
32402,
12,
6664,
220,
1627,
11,
220,
679,
21,
11,
220,
18,
25,
2946,
5975,
49023,
198,
2460,
6548,
63473,
1427,
311,
420,
11714,
596,
2466,
24608,
4717,
439,
9249,
1841,
14880,
28298,
37792,
4953,
320,
88611,
69761,
8143,
32,
8,
323,
16797,
25214,
452,
28475,
14582,
1965,
1076,
1105,
25464,
320,
88611,
38794,
55,
1932,
8,
527,
24944,
4948,
58414,
6821,
1306,
3157,
3345,
13,
29710,
268,
35180,
50542,
304,
99048,
389,
350,
8143,
32,
1418,
10209,
13356,
1139,
459,
6177,
1646,
369,
279,
2883,
596,
2800,
660,
6020,
82,
11,
719,
77120,
810,
90221,
389,
-100
] |
We compared 7 distinguished Simpli Home console tables over the previous 3 years. Check which Simpli Home console table is best. You can also Filter by type, material, style and model or choose one of our Simpli Home console table editorial picks.
Selecting a Simpli Home console table may be baffling. With various Simpli Home qualities, alternatives, types, console table sizes, colors all of which seem to be critical. At Consoletable.org we torment to make it effortless to locate unequivocally what Simpli Home console table you covet.
Simpli home console in dark walnut as stated in Homedepot. Spacious table in. Dimensions of the shelf w in. D h x d x in. For all or transient it has attractive lines draws low and.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 8,951 |
[
128000,
1687,
7863,
220,
22,
39575,
62342,
72,
5492,
2393,
12920,
927,
279,
3766,
220,
18,
1667,
13,
4343,
902,
62342,
72,
5492,
2393,
2007,
374,
1888,
13,
1472,
649,
1101,
12619,
555,
955,
11,
3769,
11,
1742,
323,
1646,
477,
5268,
832,
315,
1057,
62342,
72,
5492,
2393,
2007,
28144,
22657,
627,
3461,
287,
264,
62342,
72,
5492,
2393,
2007,
1253,
387,
65516,
2785,
13,
3161,
5370,
62342,
72,
5492,
29600,
11,
27548,
11,
4595,
11,
2393,
2007,
12562,
11,
8146,
682,
315,
902,
2873,
311,
387,
9200,
13,
2468,
1221,
708,
1169,
481,
2726,
584,
73770,
311,
1304,
433,
80192,
311,
25539,
84035,
511,
750,
1148,
62342,
72,
5492,
2393,
2007,
499,
22590,
295,
627,
50,
6517,
72,
2162,
2393,
304,
6453,
93729,
439,
11224,
304
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
1687,
7863,
220,
22,
39575,
62342,
72,
5492,
2393,
12920,
927,
279,
3766,
220,
18,
1667,
13,
4343,
902,
62342,
72,
5492,
2393,
2007,
374,
1888,
13,
1472,
649,
1101,
12619,
555,
955,
11,
3769,
11,
1742,
323,
1646,
477,
5268,
832,
315,
1057,
62342,
72,
5492,
2393,
2007,
28144,
22657,
627,
3461,
287,
264,
62342,
72,
5492,
2393,
2007,
1253,
387,
65516,
2785,
13,
3161,
5370,
62342,
72,
5492,
29600,
11,
27548,
11,
4595,
11,
2393,
2007,
12562,
11,
8146,
682,
315,
902,
2873,
311,
387,
9200,
13,
2468,
1221,
708,
1169,
481,
2726,
584,
73770,
311,
1304,
433,
80192,
311,
25539,
84035,
511,
750,
1148,
62342,
72,
5492,
2393,
2007,
499,
22590,
295,
627,
50,
6517,
72,
2162,
2393,
304,
6453,
93729,
439,
11224,
304,
-100
] |
Home/Russia/Russia responds to US 'false flag' claims
Russia responds to US 'false flag' claims
The Kremlin says there isn't any proof for 'hearsay' accusations of Russian 'operatives' set to border Ukraine
After Western media carried reviews, on Friday, citing unnamed US officers, claiming that Russia is making ready a "false flag" incident with a view to invade Ukraine, the Kremlin has dismissed them as baseless and unfounded rumors.
"So far, all these statements have been unsubstantiated and not backed up by anything," spokesman Dmitry Peskov informed TASS information company, when requested in regards to the tales making the rounds.
Earlier within the day, nameless US officers echoed claims from Kiev that Russia had despatched "operatives" into the disputed areas of Donetsk and Lugansk in jap Ukraine, to organize a "provocation" that might set off an "invasion."
CNN was the primary to report the assertion, adopted by the Guardian, Washington Post, Reuters, AFP and Bloomberg.
Washington has accused Moscow of planning to invade Ukraine for weeks now, although the Kremlin rejected this as "fake news." Similar warnings have been publicized in US media yearly since 2015, all the time based mostly on nameless sources, and have by no means come to move.
The quotes from an nameless US official apparently dovetailed with insinuations by US National Security Advisor Jake Sullivan about Russia "laying the groundwork to have the option of fabricating a pretext for an invasion," introduced up at Thursday's White House press briefing. Pentagon spokesman John Kirby referred to those claims at his personal briefing on Friday, however then added the US evaluation is that Russia hasn't but determined whether or not to invade.
Washington's claims come after every week of high-level talks with Russia, NATO and the OSCE to debate Moscow's proposals for bettering collective safety in Europe. The US and its allies rejected Russia's key proposals out of hand.
Senior Navalny allies added to Russia's 'terrorist & extremist' checklist
Christina Ricci says ladies in Hollywood are 'attending to play increasingly more attention-grabbing characters'
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 108 |
[
128000,
7778,
19945,
29578,
19945,
29578,
31680,
311,
2326,
364,
3934,
5292,
6,
8349,
198,
45606,
31680,
311,
2326,
364,
3934,
5292,
6,
8349,
198,
791,
51424,
2795,
1070,
4536,
956,
904,
11311,
369,
364,
383,
1590,
352,
6,
36569,
315,
8690,
364,
3376,
5983,
6,
743,
311,
3973,
19278,
198,
6153,
11104,
3772,
11953,
8544,
11,
389,
6740,
11,
33264,
53873,
2326,
9808,
11,
21039,
430,
8524,
374,
3339,
5644,
264,
330,
3934,
5292,
1,
10672,
449,
264,
1684,
311,
75272,
19278,
11,
279,
51424,
706,
27292,
1124,
439,
2385,
1752,
323,
9662,
13382,
35492,
627,
48058,
3117,
11,
682,
1521,
12518,
617,
1027,
80304,
4811,
10234,
323,
539,
22126,
709,
555,
4205,
1359,
18367,
88231,
57956,
52767,
16369,
350,
5045,
2038,
2883,
11,
994,
11472,
304
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
7778,
19945,
29578,
19945,
29578,
31680,
311,
2326,
364,
3934,
5292,
6,
8349,
198,
45606,
31680,
311,
2326,
364,
3934,
5292,
6,
8349,
198,
791,
51424,
2795,
1070,
4536,
956,
904,
11311,
369,
364,
383,
1590,
352,
6,
36569,
315,
8690,
364,
3376,
5983,
6,
743,
311,
3973,
19278,
198,
6153,
11104,
3772,
11953,
8544,
11,
389,
6740,
11,
33264,
53873,
2326,
9808,
11,
21039,
430,
8524,
374,
3339,
5644,
264,
330,
3934,
5292,
1,
10672,
449,
264,
1684,
311,
75272,
19278,
11,
279,
51424,
706,
27292,
1124,
439,
2385,
1752,
323,
9662,
13382,
35492,
627,
48058,
3117,
11,
682,
1521,
12518,
617,
1027,
80304,
4811,
10234,
323,
539,
22126,
709,
555,
4205,
1359,
18367,
88231,
57956,
52767,
16369,
350,
5045,
2038,
2883,
11,
994,
11472,
304,
-100
] |
Organic semiconductors have received a great deal of attention in recent years due to their optoelectronical properties and their potential applications in industry. In this thesis, resonance Raman calculations of a family of organic semiconductor dyes: pery-lene tetracarboxylic diimides (PTCD) derivatives, are presented. Also the UV/vis, infrared and normal Raman spectra, structures and molecular orbitals, excitation ener- gies, dipole moments, (hyper) polarizabilities and charge population analyses of these molecules are calculated. The calculations in this thesis are performed using a Time- Dependent Density Functional Theory (TDDFT) method, and for that, the Amsterdam Density Functional program package (ADF2009) is used. The molecular properties cal-culated for the molecules in this thesis are then compared with experimental data and the main spectroscopic characteristics are found to be, for the most part, due to the perylene plane and therefore, the same for all of its derivatives.
Torabi, MohammadAmin, "Resonance Raman Spectral Calculations of Organic Semiconductor Molecules" (2011). Electronic Theses and Dissertations. 377.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 2,884 |
[
128000,
24105,
292,
5347,
1965,
1076,
1105,
617,
4036,
264,
2294,
3568,
315,
6666,
304,
3293,
1667,
4245,
311,
872,
3469,
4748,
70329,
950,
6012,
323,
872,
4754,
8522,
304,
5064,
13,
763,
420,
17850,
11,
58081,
432,
13005,
29217,
315,
264,
3070,
315,
17808,
87836,
294,
9891,
25,
281,
727,
2922,
1994,
259,
17820,
582,
277,
2054,
88,
416,
1891,
318,
3422,
320,
2898,
6620,
8,
43645,
11,
527,
10666,
13,
7429,
279,
30136,
14,
2749,
11,
59141,
323,
4725,
432,
13005,
63697,
11,
14726,
323,
31206,
27605,
1147,
11,
3521,
7709,
38556,
12,
342,
552,
11,
24522,
1286,
14269,
11,
320,
69292,
8,
25685,
450,
8623,
323,
6900,
7187,
29060,
315,
1521,
35715,
527,
16997,
13,
578,
29217,
304,
420,
17850,
527,
10887,
1701,
264,
4212
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
24105,
292,
5347,
1965,
1076,
1105,
617,
4036,
264,
2294,
3568,
315,
6666,
304,
3293,
1667,
4245,
311,
872,
3469,
4748,
70329,
950,
6012,
323,
872,
4754,
8522,
304,
5064,
13,
763,
420,
17850,
11,
58081,
432,
13005,
29217,
315,
264,
3070,
315,
17808,
87836,
294,
9891,
25,
281,
727,
2922,
1994,
259,
17820,
582,
277,
2054,
88,
416,
1891,
318,
3422,
320,
2898,
6620,
8,
43645,
11,
527,
10666,
13,
7429,
279,
30136,
14,
2749,
11,
59141,
323,
4725,
432,
13005,
63697,
11,
14726,
323,
31206,
27605,
1147,
11,
3521,
7709,
38556,
12,
342,
552,
11,
24522,
1286,
14269,
11,
320,
69292,
8,
25685,
450,
8623,
323,
6900,
7187,
29060,
315,
1521,
35715,
527,
16997,
13,
578,
29217,
304,
420,
17850,
527,
10887,
1701,
264,
4212,
-100
] |
More than 23,000 stateless people have been granted Thai nationality in the last four years as the Southeast Asian nation seeks to end statelessness by 2024.
CHIANG RAI, Thailand – When Sawitree* turned seven, she tried to apply for her Thai nationality card.
But despite having been born in the Southeast Asian nation she could not meet the application requirements as her Thai father could not be found.
"I felt like I had no future, no options. It was like a dead end," says Sawitree, now 12, of the state of limbo that she found herself in.
Her mother and father separated when she was barely a year old. Her step-father went to district offices in Chiang Rai and Bangkok, unsuccessfully trying to get the right documents so that Sawitree could obtain an identity card showing that she was, in fact, a Thai national.
"I felt like I failed her for not being able to obtain the card. Without it, she did not have many choices," says her step-father, referring to the limitations on freedom of movement and higher educational opportunities that stateless people face.
When staff with the Adventist Relief and Development Agency (ADRA), a local NGO partner working with UNHCR, the UN Refugee Agency, assisted the family in resolving the case, they finally obtained the documentation they needed, and in mid-October, Sawitree received her national identity card.
The young Thai student's plight is far too common. There are around 10 million people worldwide who are denied a nationality. UNHCR works with governments around the world to identify, prevent, and resolve their situations, and in 2014 launched the global #IBelong Campaign that seeks to end statelessness over a decade.
In Thailand alone, there are currently 438,821 people registered as stateless. Many stateless people and those at risk of statelessness in Thailand come from areas where national borders have changed, leaving their nationality in question. Some belong to "hill tribes" living in remote areas with limited access to information about nationality procedures and who, in the past, lived without registration or identity documentation.
The Thai government has pledged to attain zero statelessness by 2024. Earlier this year, it requested all districts in the country to identify and issue legal status – which could range from Thai nationality to permanent or temporary residence – to eligible stateless students in its database. This move could benefit up to 65,000 students like Sawitree, who might otherwise struggle to pursue higher education.
To help address the challenge, UNHCR has been working with the Thai authorities and, most recently with ADRA, to open "service points" in different schools in Chiang Rai's Mae Fah Luang and Mae Chan districts.
Stateless students and their families living there have been able to obtain nationality-related information and eventually lodge applications for birth registration, nationality, permanent residency and related civil status and identity documentation.
The project has engaged government officials at the district level, school principals, community leaders as well as local civil society, who have also been working on the issue of statelessness. The target – to prepare 10,000 applications by the end of 2016.
The Chiang Rai project has also helped Manee, 39, a single mother with two daughters from the Lahu hill tribe. She lives and works on a farm, but in the past as a stateless person could not travel freely outside of her district for work or to visit relatives.
When she tried to apply for nationality, she faced several challenges. Born in a remote area, her birth was not registered and she found it difficult to get through the administrative procedures without assistance.
With this support, she acquired Thai nationality last month. She says this has brought her peace of mind. She looks forward to exercising her full rights, including the right to travel to see her cousins, and access public services that will also benefit her children. She also feels empowered to be more active and vocal in village activities.
*Names of all stateless people have been changed for protection reasons.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 6,559 |
[
128000,
7816,
1109,
220,
1419,
11,
931,
1614,
1752,
1274,
617,
1027,
11938,
27490,
59343,
304,
279,
1566,
3116,
1667,
439,
279,
36664,
14875,
7140,
26737,
311,
842,
1614,
33028,
555,
220,
2366,
19,
627,
2198,
40,
5330,
432,
15836,
11,
30567,
1389,
3277,
38107,
275,
770,
9,
6656,
8254,
11,
1364,
6818,
311,
3881,
369,
1077,
27490,
59343,
3786,
627,
4071,
8994,
3515,
1027,
9405,
304,
279,
36664,
14875,
7140,
1364,
1436,
539,
3449,
279,
3851,
8670,
439,
1077,
27490,
7126,
1436,
539,
387,
1766,
627,
7189,
6612,
1093,
358,
1047,
912,
3938,
11,
912,
2671,
13,
1102,
574,
1093,
264,
5710,
842,
1359,
2795,
38107,
275,
770,
11,
1457,
220,
717,
11,
315,
279,
1614,
315,
4671,
754,
430,
1364,
1766,
11937,
304,
627,
21364,
6691
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
7816,
1109,
220,
1419,
11,
931,
1614,
1752,
1274,
617,
1027,
11938,
27490,
59343,
304,
279,
1566,
3116,
1667,
439,
279,
36664,
14875,
7140,
26737,
311,
842,
1614,
33028,
555,
220,
2366,
19,
627,
2198,
40,
5330,
432,
15836,
11,
30567,
1389,
3277,
38107,
275,
770,
9,
6656,
8254,
11,
1364,
6818,
311,
3881,
369,
1077,
27490,
59343,
3786,
627,
4071,
8994,
3515,
1027,
9405,
304,
279,
36664,
14875,
7140,
1364,
1436,
539,
3449,
279,
3851,
8670,
439,
1077,
27490,
7126,
1436,
539,
387,
1766,
627,
7189,
6612,
1093,
358,
1047,
912,
3938,
11,
912,
2671,
13,
1102,
574,
1093,
264,
5710,
842,
1359,
2795,
38107,
275,
770,
11,
1457,
220,
717,
11,
315,
279,
1614,
315,
4671,
754,
430,
1364,
1766,
11937,
304,
627,
21364,
6691,
-100
] |
George Bettinger's Mom & Pop Shop: The Interviews
Di: George Bettinger
Letto da: George Bettinger, Madeline Kahn, Rose Marie, Judy Tenuta, Kathy Garver, Drew Friedman, Lily Tomlin
Editore: BearManor Media
Comedian and radio personality George Bettinger talked with Emmy-winning actress-comedienne Lily Tomlin (Laugh-In) just after she received her Kennedy Center Honors Award. Tomlin, who currently stars opposite Jane Fonda, Martin Sheen, and Sam Waterston in the hit Netflix original series Grace and Frankie, is just one of 10 rare and unedited celebrity interviews in this collection: Madeline Kahn, Julie Newmar, Rose Marie, Robert Morse, Judy Tenuta, Butch Patrick, Michael Learned, Kathy Garver, Drew Friedman, and Lily Tomlin.
In 1982, George Bettinger got started in TV with Movie Magic, a series he wrote, produced, and hosted. Movie Magic (later The George Bettinger Show) ran on Manhattan Cable Television from 1982 to 1987. It was cartoons, comedy, and fun. Soupy Sales said George made him think of himself when he was young. During the making of Movie Magic, George was also appearing on The Joe Franklin Show and The Uncle Floyd Show. Next he moved on to doing radio and TV commercials. Soon he was producing and directing corporate training films for companies such as Pfizer, Hoffman LaRoche, The Cancer Institute, and Maimonides Medical Center. His CDs have been given fine mention in The Daily News, New York Post, and Herald Tribune newspapers.
Currently George is teaching acting and creative writing and continuing to host live shows. You can hear George Bettinger live on his Internet radio show The Mom & Pop Shop on Mondays, Wednesdays, and Fridays at 4:00 pm EDT and 1:00 PM Pacific. The show is an entertaining mixture of classic and rare music and features zany comedy skits with his unique impersonations.
©2014 George Bettinger (P)2014 Joe Bevilacqua and Waterlogg Productions
George Bettinger's Mom & Pop Variety Shop
George Bettinger's Mom & Pop Shop Interviews & Variety, Box Set
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 2,447 |
[
128000,
40052,
85085,
261,
596,
22969,
612,
10466,
14355,
25,
578,
89172,
198,
22427,
25,
10058,
85085,
261,
198,
10267,
998,
3067,
25,
10058,
85085,
261,
11,
9671,
4939,
99666,
11,
16344,
33116,
11,
55158,
18165,
16382,
11,
65298,
12471,
424,
11,
41208,
62458,
11,
48390,
8529,
3817,
198,
4126,
461,
25,
24941,
1692,
269,
7972,
198,
1110,
29667,
323,
9063,
17743,
10058,
85085,
261,
15243,
449,
67112,
27875,
24577,
11733,
291,
26584,
48390,
8529,
3817,
320,
84967,
32600,
8,
1120,
1306,
1364,
4036,
1077,
24573,
5955,
16958,
1105,
17768,
13,
8529,
3817,
11,
889,
5131,
9958,
14329,
22195,
435,
18693,
11,
11826,
3005,
268,
11,
323,
8388,
10164,
7876,
304,
279,
4295,
23469,
4113,
4101,
32171,
323,
94545,
11,
374,
1120,
832,
315,
220,
605,
9024,
323
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
40052,
85085,
261,
596,
22969,
612,
10466,
14355,
25,
578,
89172,
198,
22427,
25,
10058,
85085,
261,
198,
10267,
998,
3067,
25,
10058,
85085,
261,
11,
9671,
4939,
99666,
11,
16344,
33116,
11,
55158,
18165,
16382,
11,
65298,
12471,
424,
11,
41208,
62458,
11,
48390,
8529,
3817,
198,
4126,
461,
25,
24941,
1692,
269,
7972,
198,
1110,
29667,
323,
9063,
17743,
10058,
85085,
261,
15243,
449,
67112,
27875,
24577,
11733,
291,
26584,
48390,
8529,
3817,
320,
84967,
32600,
8,
1120,
1306,
1364,
4036,
1077,
24573,
5955,
16958,
1105,
17768,
13,
8529,
3817,
11,
889,
5131,
9958,
14329,
22195,
435,
18693,
11,
11826,
3005,
268,
11,
323,
8388,
10164,
7876,
304,
279,
4295,
23469,
4113,
4101,
32171,
323,
94545,
11,
374,
1120,
832,
315,
220,
605,
9024,
323,
-100
] |
In Southern Maryland and parts of Virginia, a heating system is a necessary investment. As a homeowner, the best way to protect that investment is through proactive maintenance of your furnace. No matter the style, age, or brand of the equipment, performance will gradually decline without regular service. Wear and tear, the buildup of contaminants, and minor issues can result in higher utility bills, costly repairs, premature system failure, poor air quality, diminished comfort, excess noise, and even risk your household's safety. And without seasonal maintenance, your warranty will not apply. By scheduling convenient and affordable heating service from T. N. Bowes Heating & Air Conditioning, Inc., you'll keep your furnace in peak condition.
Our team of service technicians is extensively trained, regularly updated, and proficient in all makes and models. We adhere to strict service procedures to maximize the efficiency, longevity, and reliability of all components. By focusing on quality, we deliver superior results that will save you time and money and ensure your comfort even on the coldest days. Rest assured, our dedicated technicians arrive at your doorstep right on time, keep you informed, and leave no mess behind. Exceeding your expectations for service and performance, T. N. Bowes Heating & Air Conditioning, Inc. is your trusted source for furnace and heating service across Leonardtown, Waldorf, Lusby, La Plata, St. Leonard, California, Prince Frederick, Hughesville & Hollywood, MD. Call us at 301-880-0161 for information about our preventative maintenance plans.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 4,271 |
[
128000,
644,
16642,
23481,
323,
5596,
315,
13286,
11,
264,
24494,
1887,
374,
264,
5995,
9341,
13,
1666,
264,
67798,
11,
279,
1888,
1648,
311,
6144,
430,
9341,
374,
1555,
59314,
13709,
315,
701,
55564,
13,
2360,
5030,
279,
1742,
11,
4325,
11,
477,
6883,
315,
279,
7241,
11,
5178,
690,
27115,
18174,
2085,
5912,
2532,
13,
42162,
323,
18036,
11,
279,
86765,
315,
88959,
11,
323,
9099,
4819,
649,
1121,
304,
5190,
15919,
19123,
11,
34348,
31286,
11,
42227,
1887,
8060,
11,
8009,
3805,
4367,
11,
54182,
6981,
11,
13937,
12248,
11,
323,
1524,
5326,
701,
14048,
596,
7296,
13,
1628,
2085,
36899,
13709,
11,
701,
8642,
690,
539,
3881,
13,
3296,
38952,
17125,
323,
17049,
24494,
2532,
505,
350,
13,
452,
13,
18943,
288,
68954,
612
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
644,
16642,
23481,
323,
5596,
315,
13286,
11,
264,
24494,
1887,
374,
264,
5995,
9341,
13,
1666,
264,
67798,
11,
279,
1888,
1648,
311,
6144,
430,
9341,
374,
1555,
59314,
13709,
315,
701,
55564,
13,
2360,
5030,
279,
1742,
11,
4325,
11,
477,
6883,
315,
279,
7241,
11,
5178,
690,
27115,
18174,
2085,
5912,
2532,
13,
42162,
323,
18036,
11,
279,
86765,
315,
88959,
11,
323,
9099,
4819,
649,
1121,
304,
5190,
15919,
19123,
11,
34348,
31286,
11,
42227,
1887,
8060,
11,
8009,
3805,
4367,
11,
54182,
6981,
11,
13937,
12248,
11,
323,
1524,
5326,
701,
14048,
596,
7296,
13,
1628,
2085,
36899,
13709,
11,
701,
8642,
690,
539,
3881,
13,
3296,
38952,
17125,
323,
17049,
24494,
2532,
505,
350,
13,
452,
13,
18943,
288,
68954,
612,
-100
] |
Dad and I are just back from a week on the road. We went to Adelaide to look at our new 4WD trailer which will be ready in 8 weeks or so. It will be great for me because there's plenty of shade and places to sleep. Plus we will camp in forests and beaches and in the desert.
We seem to get on even better now than we used to. We don't have to talk that much or do master/dog things. It's more that we just hang out together. I know some things really annoy Dad, like barking when the car starts or when I see a bike. But I can't stop it. I'm half cattle dog. Any way Dad should talk. He only drinks green tea from corrugated white paper cups. And always two of them. Whats that about?
The best part of the trip was in Gunning because there were these sheep that didn't move outside a cafe.
I like it when we slept by the side of the road as well. The air flows over me and brings smells all night that make me dream of chasing horses and sheep.
In Adelaide we stayed at our friends Leigh and Janet's place and I played with Molly the German Shepherd. She is nice but young and frisky so I have to snarl at her sometimes.
Then we came back along the Great Ocean Road. I got to surf. I am really good. I learnt when I was young when we lived near the beach. When the stick gets thrown into the water, the trick is to wait for a wave to crash and then go. Sometimes people cheer me because I catch the wave back in. Dad calls me the Surfing Dingo.
We are Amooti,Dubbo and Tinkerbell and we are joyous and jealous of your new home Tommie. We are inner city dudes but we go bush as often as we can. I (Dubbo) have just had to give a statement to the dog ranger because I nipped the parking officer outside my house. I tried to tell Mum I didn't mean to but.... really I did.
Hi Amooti, Dubbo and Tinkerbell. You guys sound great. What sort of dogs are you? I understand your position on the dog ranger. Why WOULDN''T you nip him. Dad has to lock the gate everyday because I do the same with the postie. Its the bike I hate. Talk soon. I think I am going to the beach this weekend.
Hi Tommie You are so lucky. My mum and dad are planning to head off wiithout me. They think I don't know, but I'm smarter than they think.
Hi Tommie! I reckon your new trailer sounds just the thing for canine travel. My mum's just bought a tent and we tested it out last week. I am a 2 yr old retrievador so I got a bit caught up in the zippered doorway. Why do humans have to do it like that? The good part was that I got to sleep next to her which she doesn't let me do at the house. We went to Fisherman's Bay in SA where I got to run and run on the beach without anyone giving me a hard time.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 8,664 |
[
128000,
35,
329,
323,
358,
527,
1120,
1203,
505,
264,
2046,
389,
279,
5754,
13,
1226,
4024,
311,
50301,
311,
1427,
520,
1057,
502,
220,
19,
18023,
19809,
902,
690,
387,
5644,
304,
220,
23,
5672,
477,
779,
13,
1102,
690,
387,
2294,
369,
757,
1606,
1070,
596,
11510,
315,
28601,
323,
7634,
311,
6212,
13,
12623,
584,
690,
3190,
304,
36658,
323,
35909,
323,
304,
279,
24521,
627,
1687,
2873,
311,
636,
389,
1524,
2731,
1457,
1109,
584,
1511,
311,
13,
1226,
1541,
956,
617,
311,
3137,
430,
1790,
477,
656,
7491,
3529,
540,
2574,
13,
1102,
596,
810,
430,
584,
1120,
15020,
704,
3871,
13,
358,
1440,
1063,
2574,
2216,
38292,
33621,
11,
1093,
293,
34552,
994,
279,
1841,
8638,
477,
994,
358,
1518,
264,
13260
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
35,
329,
323,
358,
527,
1120,
1203,
505,
264,
2046,
389,
279,
5754,
13,
1226,
4024,
311,
50301,
311,
1427,
520,
1057,
502,
220,
19,
18023,
19809,
902,
690,
387,
5644,
304,
220,
23,
5672,
477,
779,
13,
1102,
690,
387,
2294,
369,
757,
1606,
1070,
596,
11510,
315,
28601,
323,
7634,
311,
6212,
13,
12623,
584,
690,
3190,
304,
36658,
323,
35909,
323,
304,
279,
24521,
627,
1687,
2873,
311,
636,
389,
1524,
2731,
1457,
1109,
584,
1511,
311,
13,
1226,
1541,
956,
617,
311,
3137,
430,
1790,
477,
656,
7491,
3529,
540,
2574,
13,
1102,
596,
810,
430,
584,
1120,
15020,
704,
3871,
13,
358,
1440,
1063,
2574,
2216,
38292,
33621,
11,
1093,
293,
34552,
994,
279,
1841,
8638,
477,
994,
358,
1518,
264,
13260,
-100
] |
AARP Bulletin: "Two years after undergoing a double mastectomy and chemotherapy so severe she was hospitalized in intensive care for several weeks, breast cancer survivor Denise Hicks should be following what her doctors call 'the plan,'" which includes additional medications and treatments. But she can't. "Hicks has health insurance but already reached her coverage limits. So the CT scan that her oncologist 'strongly advised" months ago to check a possible recurrence remains undone. … She's also skipping recommended medications. … Hick's disquieting predicament is not unique. At least 2 million Americans—roughly one in six cancer survivors—decide to forgo at least some of their recommended follow-up medical care because of the cost, according to recent research" published in the medical journal Cancer (Kirchheimer, 8/27).
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 5,170 |
[
128000,
32,
43893,
61063,
25,
330,
11874,
1667,
1306,
47397,
264,
2033,
19218,
72783,
323,
62730,
779,
15748,
1364,
574,
70340,
304,
37295,
2512,
369,
3892,
5672,
11,
17659,
9572,
49748,
81349,
74605,
1288,
387,
2768,
1148,
1077,
16410,
1650,
364,
1820,
3197,
49982,
902,
5764,
5217,
31010,
323,
22972,
13,
2030,
1364,
649,
956,
13,
330,
39,
5908,
706,
2890,
8276,
719,
2736,
8813,
1077,
10401,
13693,
13,
2100,
279,
19084,
8737,
430,
1077,
78970,
16549,
364,
4620,
398,
26160,
1,
4038,
4227,
311,
1817,
264,
3284,
76293,
8625,
79841,
13,
4696,
3005,
596,
1101,
43759,
11349,
31010,
13,
4696,
79028,
596,
834,
44750,
287,
79434,
2908,
374,
539,
5016,
13,
2468,
3325,
220,
17,
3610,
9053,
2345,
1458,
398,
832,
304,
4848,
9572,
32696,
2345,
8332
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
32,
43893,
61063,
25,
330,
11874,
1667,
1306,
47397,
264,
2033,
19218,
72783,
323,
62730,
779,
15748,
1364,
574,
70340,
304,
37295,
2512,
369,
3892,
5672,
11,
17659,
9572,
49748,
81349,
74605,
1288,
387,
2768,
1148,
1077,
16410,
1650,
364,
1820,
3197,
49982,
902,
5764,
5217,
31010,
323,
22972,
13,
2030,
1364,
649,
956,
13,
330,
39,
5908,
706,
2890,
8276,
719,
2736,
8813,
1077,
10401,
13693,
13,
2100,
279,
19084,
8737,
430,
1077,
78970,
16549,
364,
4620,
398,
26160,
1,
4038,
4227,
311,
1817,
264,
3284,
76293,
8625,
79841,
13,
4696,
3005,
596,
1101,
43759,
11349,
31010,
13,
4696,
79028,
596,
834,
44750,
287,
79434,
2908,
374,
539,
5016,
13,
2468,
3325,
220,
17,
3610,
9053,
2345,
1458,
398,
832,
304,
4848,
9572,
32696,
2345,
8332,
-100
] |
The secret to out-earning your colleagues begins with asking -- here's how to effectively do it
Joe Avella, Libby Kane
Susie Moore is a career coach and the author of "What If It Does Work Out?". In this video, she explains when and how to ask for more money. The following is a transcript of the video.
People who earn more ask.
Ask, ask, ask, ask. Asking's a massive form of action. Typically you'll notice if someone's moved around in different companies, over time, they tend to earn more because there's always that conversation at the recruitment and selection phase of, "What are you looking for?" So there's a negotiation that can happen organically.
If you're in the same job though, for a while, having that conversation can be more tricky. Especially if you don't have annual reviews. They don't exist everywhere anymore.
The most important thing is to understand what your role is being paid, in the market, again everything is market research. Understand kind of where you sit, if you think it's too low you're probably right. Really get that information, be equipped with it, understand why you're good at what you do. I mean, that's the other thing people will just remember their mistakes. They made a mistake on a report four months ago, and then still feel sick about it, versus going, "Well I grew the team, and I increased profit, and I streamlined this process and I bring a lot of value to the office in terms of my attitude."
So, really being able to understand your value as an employee, doing the research, knowing what other people like you are getting paid, those two things are really critical. And then you just have to ask, you have to set up a conversation, with your boss, and go for it.
I mean, think about it. A few minutes of an awkward conversation over time, like how that will compound your income, it's absolutely worth it. Actually, I love what Tim Ferriss says about this, he says that a successful person can almost be measured by the number of uncomfortable conversations they're willing to have.
That conversation, those few minutes, when you think of the upside versus the downside, it's absolutely worth doing.
More from Libby Kane:
The secret to out-earning your colleagues begins with asking — here's how to effectively do it
A finance expert shares advice on whether you should save for retirement or pay off your student loans
A woman who left her $US500,000-a-year job to work for herself says a side hustle can't be successful without 3 factors
A woman who left her $US500,000-a-year job to work for herself debunks the most common excuse she hears from would-be entrepreneurs
First job, first car, first apartment: The smartest things to do with your money in your 20s
bi original video bi-video career careers green screen money strategy susie moore video video-us
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 8,384 |
[
128000,
791,
6367,
311,
704,
12,
16933,
701,
18105,
12302,
449,
10371,
1198,
1618,
596,
1268,
311,
13750,
656,
433,
198,
41444,
7671,
6985,
11,
5806,
1729,
46656,
198,
78524,
648,
20832,
374,
264,
7076,
7395,
323,
279,
3229,
315,
330,
3923,
1442,
1102,
12838,
5664,
4470,
30,
3343,
763,
420,
2835,
11,
1364,
15100,
994,
323,
1268,
311,
2610,
369,
810,
3300,
13,
578,
2768,
374,
264,
36815,
315,
279,
2835,
627,
16298,
889,
7380,
810,
2610,
627,
27264,
11,
2610,
11,
2610,
11,
2610,
13,
96249,
596,
264,
11191,
1376,
315,
1957,
13,
46402,
499,
3358,
5406,
422,
4423,
596,
7882,
2212,
304,
2204,
5220,
11,
927,
892,
11,
814,
8541,
311,
7380,
810,
1606,
1070,
596,
2744,
430,
10652,
520,
279,
34102,
323,
6727,
10474
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
791,
6367,
311,
704,
12,
16933,
701,
18105,
12302,
449,
10371,
1198,
1618,
596,
1268,
311,
13750,
656,
433,
198,
41444,
7671,
6985,
11,
5806,
1729,
46656,
198,
78524,
648,
20832,
374,
264,
7076,
7395,
323,
279,
3229,
315,
330,
3923,
1442,
1102,
12838,
5664,
4470,
30,
3343,
763,
420,
2835,
11,
1364,
15100,
994,
323,
1268,
311,
2610,
369,
810,
3300,
13,
578,
2768,
374,
264,
36815,
315,
279,
2835,
627,
16298,
889,
7380,
810,
2610,
627,
27264,
11,
2610,
11,
2610,
11,
2610,
13,
96249,
596,
264,
11191,
1376,
315,
1957,
13,
46402,
499,
3358,
5406,
422,
4423,
596,
7882,
2212,
304,
2204,
5220,
11,
927,
892,
11,
814,
8541,
311,
7380,
810,
1606,
1070,
596,
2744,
430,
10652,
520,
279,
34102,
323,
6727,
10474,
-100
] |
import os, sys
from setuptools import setup
from setuptools.extension import Extension
from glob import glob
with open('README.rst') as readme_file:
readme = readme_file.read()
with open('HISTORY.rst') as history_file:
history = history_file.read()
requirements = [
#'Click>=6.0',
#'numpy',
#'matplotlib',
#'scipy',
#'pandas',
#'ipython',
]
test_requirements = [
# TODO: put package test requirements here
]
setup(
name='sfg2d',
version='0.6.0',
description="Python Toolkit for Analsys of 2d-sfg spectra",
long_description=readme + '\n\n' + history,
author="Malte Deiseroth",
author_email='[email protected]',
url='https://github.com/deisi/sfg2d',
packages=[
'sfg2d',
'sfg2d.data',
'sfg2d.data.calib',
'sfg2d.io',
'sfg2d.utils',
],
package_dir={'sfg2d':
'sfg2d'},
entry_points={
'console_scripts': [
'sfg2d=sfg2d.cli:main'
]
},
package_data={
'sfg2d': ['data/calib/params_Ne_670.npy']
},
include_package_data=True,
install_requires=requirements,
license="MIT license",
zip_safe=False,
keywords='sfg2d',
classifiers=[
'Development Status :: 2 - Pre-Alpha',
'Intended Audience :: Scientists',
'License :: OSI Approved :: MIT License',
'Natural Language :: English',
'Programming Language :: Python :: 3.5',
],
test_suite='tests',
tests_require=test_requirements,
)
|
{
"redpajama_set_name": "RedPajamaGithub"
}
| 5,288 |
[
128000,
475,
2709,
11,
5826,
198,
1527,
64897,
1179,
6642,
198,
1527,
64897,
41575,
1179,
27571,
198,
1527,
13509,
1179,
13509,
1432,
4291,
1825,
493,
55775,
1783,
267,
873,
439,
84784,
2517,
512,
262,
84784,
284,
84784,
2517,
4217,
2892,
4291,
1825,
493,
39,
36686,
1783,
267,
873,
439,
3925,
2517,
512,
262,
3925,
284,
3925,
2517,
4217,
2892,
71226,
284,
2330,
262,
43137,
2677,
10123,
21,
13,
15,
756,
262,
43137,
36183,
756,
262,
43137,
81527,
756,
262,
43137,
2445,
23799,
756,
262,
43137,
79,
56533,
756,
262,
43137,
23799,
4690,
756,
2595,
1985,
90732,
284,
2330,
262,
674,
5456,
25,
2231,
6462,
1296,
8670,
1618,
198,
2595,
15543,
1021,
262,
836,
1151,
82,
4922,
17,
67,
756,
262,
2373,
1151,
15,
13,
21,
13,
15,
756
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
475,
2709,
11,
5826,
198,
1527,
64897,
1179,
6642,
198,
1527,
64897,
41575,
1179,
27571,
198,
1527,
13509,
1179,
13509,
1432,
4291,
1825,
493,
55775,
1783,
267,
873,
439,
84784,
2517,
512,
262,
84784,
284,
84784,
2517,
4217,
2892,
4291,
1825,
493,
39,
36686,
1783,
267,
873,
439,
3925,
2517,
512,
262,
3925,
284,
3925,
2517,
4217,
2892,
71226,
284,
2330,
262,
43137,
2677,
10123,
21,
13,
15,
756,
262,
43137,
36183,
756,
262,
43137,
81527,
756,
262,
43137,
2445,
23799,
756,
262,
43137,
79,
56533,
756,
262,
43137,
23799,
4690,
756,
2595,
1985,
90732,
284,
2330,
262,
674,
5456,
25,
2231,
6462,
1296,
8670,
1618,
198,
2595,
15543,
1021,
262,
836,
1151,
82,
4922,
17,
67,
756,
262,
2373,
1151,
15,
13,
21,
13,
15,
756,
-100
] |
Today's food can be a source of artificial growth hormones and pesticides, causing health risks to everyone, especially children who are most vulnerable because of their still developing immune systems. One of the best ways to limit children's exposure to harmful chemicals is to introduce them to organic food at an early stage.
Parents are advised to introduce kids to healthy living at an early age.
Food and organic product organizations confirm that organic products have become more mainstream in recent years as the demand for healthier and more natural products have risen.
Nowadays, organic products are not just found in specialty stores or drug stores but are made available at groceries, supermarkets, and online.
Nourishing kids with the healthiest food to ensure their growth and development is always a priority among moms. By establishing an organic lifestyle for the family, they are ensuring their kids get the best of nature and the best of health.
Produce such as potatoes and apples can be consumed without removing its skin, but these have the highest amounts of pesticide residue. To avoid intake of harmful chemicals and ensure to get the good minerals and vitamins, switch to organic. In addition, it is best to buy food items that have the lowest chemical residue like eggplant, cabbage, mango, and onion.
When making meals, use organic meat, eggs, and dairy. One of the few organic milk brands available in the market today is Promil Organic, which is made from 100 percent organic dairy milk from certified European farms.
Through a quick search online, anyone can review and choose among different farm-to-table concept restaurants that offer organic meals that fit their budget.
It will only require little time and space to grow one's own organic produce. The most important element in organic gardening is good and healthy soil. Home testing kits, available at gardening stores and online, can check the soil's pH level and breakdown of nutrients to ensure that it is free of harmful chemicals.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 4,767 |
[
128000,
15724,
596,
3691,
649,
387,
264,
2592,
315,
21075,
6650,
44315,
323,
61257,
11,
14718,
2890,
15635,
311,
5127,
11,
5423,
2911,
889,
527,
1455,
20134,
1606,
315,
872,
2103,
11469,
22852,
6067,
13,
3861,
315,
279,
1888,
5627,
311,
4017,
2911,
596,
14675,
311,
28856,
26333,
374,
311,
19678,
1124,
311,
17808,
3691,
520,
459,
4216,
6566,
627,
62869,
527,
26160,
311,
19678,
6980,
311,
9498,
5496,
520,
459,
4216,
4325,
627,
27844,
323,
17808,
2027,
11351,
7838,
430,
17808,
3956,
617,
3719,
810,
21391,
304,
3293,
1667,
439,
279,
7631,
369,
39345,
323,
810,
5933,
3956,
617,
41482,
627,
7184,
35076,
11,
17808,
3956,
527,
539,
1120,
1766,
304,
36974,
10756,
477,
5623,
10756,
719,
527,
1903,
2561,
520,
66508,
11,
77599,
11,
323,
2930
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
15724,
596,
3691,
649,
387,
264,
2592,
315,
21075,
6650,
44315,
323,
61257,
11,
14718,
2890,
15635,
311,
5127,
11,
5423,
2911,
889,
527,
1455,
20134,
1606,
315,
872,
2103,
11469,
22852,
6067,
13,
3861,
315,
279,
1888,
5627,
311,
4017,
2911,
596,
14675,
311,
28856,
26333,
374,
311,
19678,
1124,
311,
17808,
3691,
520,
459,
4216,
6566,
627,
62869,
527,
26160,
311,
19678,
6980,
311,
9498,
5496,
520,
459,
4216,
4325,
627,
27844,
323,
17808,
2027,
11351,
7838,
430,
17808,
3956,
617,
3719,
810,
21391,
304,
3293,
1667,
439,
279,
7631,
369,
39345,
323,
810,
5933,
3956,
617,
41482,
627,
7184,
35076,
11,
17808,
3956,
527,
539,
1120,
1766,
304,
36974,
10756,
477,
5623,
10756,
719,
527,
1903,
2561,
520,
66508,
11,
77599,
11,
323,
2930,
-100
] |
For other places with the same name, see Jasper (disambiguation).
Jasper is a town in Jasper National Park in the Alberta Rockies region of Alberta, Canada.
Located in the eastern slope of the Rocky Mountains, Jasper is the shy cousin to bold, brassy Banff and offers a more serene experience. Approximately four hours north of Banff and four hours west of Edmonton, capital of the province of Alberta. Most people enter the park via Edmonton.
It can be reached by road or train. Highway 16, the Yellowhead Highway, connects Jasper to Edmonton, which has an international airport and is served by many major airlines. Highway 93, the Icefields Parkway, connects south through the Rockies to Lake Louise and onto Banff. Drivers will have to pay for a park admission pass as the town of Jasper is located within Jasper National Park. There are bus and train connections from Edmonton, Banff and Vancouver.
Those driving to the park should be aware that a park fee of $20 per night applies. Annual passes are available for $89.
There is no public transit. A few taxis and rental cars are available, and bikes can be rented in the summer.
Hitchhikers report that getting a lift in Jasper is very easy.
Jasper is small enough that most places can be easily walked to. Tour companies offer bus service to most of the major tourist destination including Maligne Lake, The Columbia Icefields and Miette Hot Springs. The two major companies in town are Brewster and SunDog. Both these companies offer similar tours at the same price. The only difference is the size of the tour group: Brewster uses large buses (seating approx. 50) and SunDog has smaller vehicles (seating approx. 12).
Maligne Lake, which has regular boat trips to see the wonder of a lake with water two different colours in different parts. Spirit Island, made famous by a photography competition, is in the middle of Maligne Lake. The most common way to see it is to take the boat cruise (1.5 hours round trip). You can also rent canoes and kayaks, and, if you are prepared to camp at the other end of the lake, paddle the 8+ hours it takes to get there.
Angel Glacier, take highway 93A to Mount Edith Cavell and take the 500m hike from the parking lot to view Angel glacier and its lake full of icebergs. A longer 5km trail (500m up) is also available for even better views. Note for 2017 a pass is required (line up at the Jasper into center to obtain for free, from 8-10am daily). The pass is required due to the continuing reconstruction of the parking lot and limited parking available. A few years prior ghost glacier (a smaller glacier next to Angel glacier), fell off the mountain into the lake, causing a tidal wave that destroyed the trail and parking lot below.
Excellent camping in the summer. Cross Country and downhill skiing in winter.
Rent a bike. Plenty of sport shops offer bikes, which are a great way to see the area.
Miette Hot Springs, , are an excellent place to go and relax after a day of hiking. There is also a challenging 4 hour hike advertising the best views in the park that starts from the parking lot next to the hot springs.
Maligne Canyon, . The river, up near Maligne Lake, flows in a mighty torrent. When it enters Medicine Lake (named in reference to bad medicine) it does so with great volumes. It does not, however, appear to flow out. Indeed, through the summer months Medicine Lake disappears until, in mid autumn all that is left is a thin channel that itself disappears into the shoreline. In Winter, walk on the canyon bottom to see masses of blue ice where waterfalls flow in the summertime!
Athabasca Falls - The Athabasca River thunders through a narrow gorge where the walls have been smoothed and potholes are created by the sheer force of the rushing water carrying sand and rock. Parking and restroom facilities. Paved trail and picnic sites available.
Mount Robson (The park's namesake peak is the highest in the Canadian Rockies at 3954 metres. Mount Robson Provincial Park is 217,200 hectares of stunning mountains, valleys, and waterways. Forests of spruce, fir, balsam, cedar and alder are home to an abundant array of wildlife including mountain goats, moose, deer, elk, marmot and squirrels and chipmunks. The Rearguard Falls of the Fraser River is the furthest migration point of the Pacific salmon.
GyPSy Guide, Take a guided tour in your own car. The GyPSy Guide is a small electronic device that uses GPS technology to automatically trigger commentary about history, geology, directions and quirky Canadian stories, through your car stereo.
Most of the hotels in or near Jasper have restaurants that will meet or beat your expectations for dining in a small community. Some are higher-toned than others, but across the board food is generally very good and portions tend to be hearty. In addition to the hotel restaurants, the town has several noteworthy dining spots as well as numerous restaurants catering to the hungry but penny-pinching skiers and other visiting outdoors-folk.
Syrah's of Jasper, 606 Patricia St. Fine dining restauarant with great wine selection. Prices are moderately high, but not out of line for the finer restaurants in Jasper.
Dead Dog Bar & Grill, a favorite among locals, the $7 burger & fries deal is probably the cheapest meal in town. Happy hour is on Fridays. Located on Connaught Street.
Spooners, 610 Patricia St. Juice bar, gourmet coffees, teas, smoothies, homemade desserts and pastries, sandwiches, soups and salads.
Something Else, 621 Patricia St. A Greek restaurant that serves up a variety of cuisine, and pizza.
North Face Pizza, 618 Connaught Dr. Excellent pizza, great after a day at Marmot Basin!
Sorrentino's Bistro Bar, 96 Geike St. Italian restaurant, popular Canadian chain.
Sayuri's Japanese Restaurant, 410 Connaught Dr. Japanese cuisine, sushi.
Kimchi House, 407 Patricia St, +1 780 852-5022. Korean cuisine.
Jasper Brewing Company, 624 Connaught Dr, +1 780 852-4111. Unique in Jasper, beer is brewed on-site.
Earls, 600 Patricia St. Western Canadian cuisine. A restaurant and bar in one, this can get fairly packed out. The staff aren't fantastic, but the food more than makes up for the wait to be seated.
Buckles Saloon, Maligne Lodge, W end Connaught Dr, +1 780 852-7074.
De'd Dog Bar & Grill, Bsmt, 620 Connaught Dr, +1 780 852-9449.
Villa Caruso, Corner, Connaught Dr and Hazel Ave, +1 780 852-3920.
The 'B' Ale House, 510 Patricia St, +1 780 852-3386.
Chateau Jasper, 96 Geikie St, +1 780 852-5644.
Nick's Bar, Juniper St between Connaught Dr and Geikie St, +1 780 852-4966.
In the summer, there is excellent camping in the campgrounds operated by Parks Canada. Throughout the year, it is possible to stay in private home accommodation - rooms or suites in residents' homes. Prices vary, but standards are high - full details of what is available can be found in the tourist office in town, or at . Hostelling International runs a series of hostels throughout Jasper National Park. Jasper International is the closest to town (7km south-west from Jasper on Whistlers Road, and is open all year), and four Wilderness Hostels which cater more towards trekkers and those people who "enjoy getting in touch with nature."
There are numerous motels and inns within the town, as well as in Hinton, just outside the eastern entrance to the park.
Austrian Haven B&B Located at 812 Patricia Street, +1 780 852-4259. Owned and run by a very friendly and welcoming Austrian lady. There are only 2 rooms, which are beautifully decorated with the biggest most comfortable beds you will ever sleep in. Both the rooms have excellent views of the mountains around Jasper.
Mount Robson Inn 902 Connaught Drive, +1 780 852-3327. A comfortable well equipped two storey motel style inn. All rooms have air conditioning, cable television, and plenty of room. Reasonable prices.
Patricia Lake Bungalows, +1 780 852-3560.
Chateau Jasper 96 Geikie St, +1 780 852-5644 Offers exceptional accommodations and amenities complemented by a young, personable staff who extend a warm and sincere welcome. Each guest room is retains the charm of the majestic Rockies. A restaurant and conference facilities are available on-site. Rates vary with season - late Fall being the lowest and Summer being the highest $100-$300.
Mount Robson Inn, 902 Connaught Drive, +1 780 852-3327. Offers 80 luxurious, well appointed rooms and suites. $94-325.
Jasper Inn Alpine Resort, 98 Geikie Street, +1 780 852-4461. Cozy and tastefully appointed fireside accommodation, comfortable amenities and casual fine dining amid the rugged beauty of the Canadian Rockies. $93-413.
Lobstick Lodge, 94 Geikie Street, A favourite among the skiing crowd, Lobstick Lodge features a restaurant onsite, as well as FIVE hot tubs and an indoor pool. Rates: varies with season; Fall has the cheapest rates (<$100 is common), while summer is high season (June - August).
Fairmont Jasper Park Lodge, Old Lodge Rd (turn south off highway 16 just east of the townsite, and make a quick right after crossing the river), +1 780 852-3301 or 800-257-7544 ([email protected], fax 780-852-5107), . The Jasper Park Lodge is the premier hotel in Jasper, and one of the most famous and storied mountain resorts in the Canada. Started as a tenting site in conjunction with the Grand Trunk Railway in 1915, it has grown into a renowned destination resort. $400 and up.
Bears looking for food are a potential problem. The city has installed bear proof garbage cans and business use bear proof dumpsters.
It is a good solution until the bears learn how to read!
VIA Rail runs routes through Jasper, which you can often include as a free stop-over if heading between Vancouver and Edmonton.
It is also popular to travel down to Lake Louise and Banff, making stops at Athabasca Falls, Peyto Lake, Crowfoot Glacier, and Bow Lake.
This page was last edited on 10 September 2017, at 00:28.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 6,485 |
[
128000,
2520,
1023,
7634,
449,
279,
1890,
836,
11,
1518,
71584,
320,
4338,
3042,
343,
4090,
4390,
41,
33361,
220,
374,
264,
6424,
304,
71584,
5165,
5657,
304,
279,
33654,
80151,
5654,
315,
33654,
11,
7008,
627,
48852,
304,
279,
24024,
31332,
315,
279,
48641,
41114,
11,
71584,
374,
279,
33394,
36271,
311,
14265,
11,
1437,
27801,
23565,
544,
323,
6209,
264,
810,
95863,
3217,
13,
79904,
3116,
4207,
10411,
315,
23565,
544,
323,
3116,
4207,
9909,
315,
46387,
11,
6864,
315,
279,
17271,
315,
33654,
13,
7648,
1274,
3810,
279,
6246,
4669,
46387,
627,
2181,
649,
387,
8813,
555,
5754,
477,
5542,
13,
29866,
220,
845,
11,
279,
26541,
2025,
29866,
11,
34161,
71584,
311,
46387,
11,
902,
706,
459,
6625,
17149,
323,
374,
10434,
555,
1690
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
2520,
1023,
7634,
449,
279,
1890,
836,
11,
1518,
71584,
320,
4338,
3042,
343,
4090,
4390,
41,
33361,
220,
374,
264,
6424,
304,
71584,
5165,
5657,
304,
279,
33654,
80151,
5654,
315,
33654,
11,
7008,
627,
48852,
304,
279,
24024,
31332,
315,
279,
48641,
41114,
11,
71584,
374,
279,
33394,
36271,
311,
14265,
11,
1437,
27801,
23565,
544,
323,
6209,
264,
810,
95863,
3217,
13,
79904,
3116,
4207,
10411,
315,
23565,
544,
323,
3116,
4207,
9909,
315,
46387,
11,
6864,
315,
279,
17271,
315,
33654,
13,
7648,
1274,
3810,
279,
6246,
4669,
46387,
627,
2181,
649,
387,
8813,
555,
5754,
477,
5542,
13,
29866,
220,
845,
11,
279,
26541,
2025,
29866,
11,
34161,
71584,
311,
46387,
11,
902,
706,
459,
6625,
17149,
323,
374,
10434,
555,
1690,
-100
] |
All aboard the Kerry bandwagon. ..
WASHINGTON (PAI) – Union membership declined again in 2003, dropping to 12.9 percent of the nation's workforce, compared to 13.3 percent in 2002. The federal Bureau of Labor Statistics (BLS) reported the numbers last month. ..
The huge dust cloud from the imploded J.L. Hudson's building on Woodward Avenue in 1998 has been long been settled. ..
It looks like a lighthouse. It was modeled after a lighthouse. But the U.S. Coast Guard prefers that we don't call it a lighthouse. ..
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 9,587 |
[
128000,
2460,
37606,
279,
40340,
7200,
94219,
13,
55638,
51577,
320,
8201,
40,
8,
1389,
9323,
16250,
19284,
1578,
304,
220,
1049,
18,
11,
26156,
311,
220,
717,
13,
24,
3346,
315,
279,
7140,
596,
32027,
11,
7863,
311,
220,
1032,
13,
18,
3346,
304,
220,
1049,
17,
13,
578,
6918,
22555,
315,
15277,
25647,
320,
33,
7416,
8,
5068,
279,
5219,
1566,
2305,
13,
55638,
791,
6908,
16174,
9624,
505,
279,
11866,
6879,
622,
1236,
13,
39253,
596,
4857,
389,
87218,
17569,
304,
220,
2550,
23,
706,
1027,
1317,
1027,
23183,
13,
55638,
2181,
5992,
1093,
264,
326,
59009,
13,
1102,
574,
62653,
1306,
264,
326,
59009,
13,
2030,
279,
549,
815,
13,
16377,
12542,
55064,
430,
584,
1541,
956,
1650,
433,
264,
326,
59009,
13,
5354
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
2460,
37606,
279,
40340,
7200,
94219,
13,
55638,
51577,
320,
8201,
40,
8,
1389,
9323,
16250,
19284,
1578,
304,
220,
1049,
18,
11,
26156,
311,
220,
717,
13,
24,
3346,
315,
279,
7140,
596,
32027,
11,
7863,
311,
220,
1032,
13,
18,
3346,
304,
220,
1049,
17,
13,
578,
6918,
22555,
315,
15277,
25647,
320,
33,
7416,
8,
5068,
279,
5219,
1566,
2305,
13,
55638,
791,
6908,
16174,
9624,
505,
279,
11866,
6879,
622,
1236,
13,
39253,
596,
4857,
389,
87218,
17569,
304,
220,
2550,
23,
706,
1027,
1317,
1027,
23183,
13,
55638,
2181,
5992,
1093,
264,
326,
59009,
13,
1102,
574,
62653,
1306,
264,
326,
59009,
13,
2030,
279,
549,
815,
13,
16377,
12542,
55064,
430,
584,
1541,
956,
1650,
433,
264,
326,
59009,
13,
5354,
-100
] |
Turkey meatballs + whole grain pasta is the perfect thing to make when you want a dinner that's healthy, but also cozy and comforting.
I absolutely love a classic plate of spaghetti and meatballs. When it's done right, it can be so satisfying. I mean, all pasta dishes are really satisfying. But there's something about the meatballs that add a nostalgic element to the meal. Sometimes, though, we crave it a bit more done up. I've added modern day healthy twists and some extra ingredients to turn the big ole plate of spaghetti into a dish that's a little bit more refined.
Healthy pastas are one of my go-to weeknight meals. It's so simple to just boil some whole grain pasta, sear a chicken breast, and sauté some veggies on the side for a quick dinner. Since I do that so often, I decided to treat myself with something more extravagant, but still just as easy to make. I also created this dish with Valentine's day in mind…if you cut the recipe in half, it's the perfect healthy date-night dinner for two, so you and your special someone can enjoy a delicious meal while feeling amazing afterwards. But, if you're in the mood for dessert and MORE pasta, you can follow it up with my Dark Chocolate Ravioli.
Back to the dish. It's a super simple tomato sauce that gets simmered until thickened. While the sauce does its thing on the stove, I mix the ground turkey with sautéed onion, garlic, and spices, and roll out the meatballs. The ground turkey provides a different texture than your usual ground beef or pork, but it's leaner and still stays moist. The key is to not overcook it. I sear the meatballs quickly just until they turn golden brown, and then I remove them from the skillet (it's ok for them to be underdone at this point). Once the veggies get cooked and the sauce gets added, I return the meatballs to the pan to gently finish cooking, keeping all the moisture inside. A quick toss with some whole grain linguine and you're good to go. I would definitely recommend to start eating more whole grain pastas. I still love and eat regular pasta on occasion, but on a weekly basis, whole grain is much more nutritious for you. It has a nutty taste and a strong bite, which I've grown to really like.
Other than that, you just want lots of freshly grated parmesan cheese inside the pasta as well as thick shaves on top. Some fresh basil, and maybe a glass of red wine, and you've got yourself quite the treat. Enjoy!
Coat the bottom of a pot with olive oil and heat over medium heat. Add the onions and sauté until golden brown, stirring often. Meanwhile, pulse the tomatoes in a food processor until chunky. Add the garlic to the pot and cook for a minute. Pour in the tomatoes and add the red pepper flakes. Let the sauce come to a boil and then reduce the heat to a low simmer. Cook for 30 minutes covered and then 30 minutes uncovered, stirring often. Stir in the basil at the end and season with salt and pepper to taste. Keep warm until serving.
Coat a large sauté pan with olive oil and heat over medium heat. Add the onions and sauté until golden brown, stirring often. Add the garlic and cook for a minute. Season with salt and pepper. Let the mixture cool. In a mixing bowl, combine the ground turkey, onion mixture, red pepper flakes, dried oregano, anise seed, salt and black pepper. Mix until incorporated. Form the ground turkey mixture into 8 meatballs. Season with a little more salt and pepper. Coat a large skillet with canola oil. Sear the meatballs on each side until golden brown. Transfer the meatballs to a plate.
To the meatball pan, add the leeks, along with a little more olive oil if needed. Sauté for 5 minutes. Add the red kale and kumato tomatoes. Season with salt and pepper. Cook until the kale is wilted and the tomatoes have softened. Add the tomato basil sauce to the pan, along with the seared meatballs. Simmer until the meatballs have finished cooking through. Stir in the pasta and freshly grate lots of parmesan cheese into the skillet. Stir to coat.
Serve the pasta with the meatballs on top and a spoonful of sauce over everything. Use a peeler to shave thick pieces of parmesan as a garnish. Finish with a sprinkling of basil.
I usually make a double or triple batch of the sauce so I can use it in other meals during the week. If you don't get through it all, you can freeze it for the next time you want some homemade tomato sauce.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 6,887 |
[
128000,
73819,
13339,
46618,
489,
4459,
24875,
37547,
374,
279,
4832,
3245,
311,
1304,
994,
499,
1390,
264,
14177,
430,
596,
9498,
11,
719,
1101,
43535,
323,
69130,
627,
40,
11112,
3021,
264,
11670,
12235,
315,
88010,
323,
13339,
46618,
13,
3277,
433,
596,
2884,
1314,
11,
433,
649,
387,
779,
37154,
13,
358,
3152,
11,
682,
37547,
26863,
527,
2216,
37154,
13,
2030,
1070,
596,
2555,
922,
279,
13339,
46618,
430,
923,
264,
83374,
2449,
311,
279,
15496,
13,
18156,
11,
3582,
11,
584,
81413,
433,
264,
2766,
810,
2884,
709,
13,
358,
3077,
3779,
6617,
1938,
9498,
62990,
323,
1063,
5066,
14293,
311,
2543,
279,
2466,
32878,
12235,
315,
88010,
1139,
264,
12269,
430,
596,
264,
2697,
2766,
810,
38291,
627,
97213,
3347,
300,
527,
832
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
73819,
13339,
46618,
489,
4459,
24875,
37547,
374,
279,
4832,
3245,
311,
1304,
994,
499,
1390,
264,
14177,
430,
596,
9498,
11,
719,
1101,
43535,
323,
69130,
627,
40,
11112,
3021,
264,
11670,
12235,
315,
88010,
323,
13339,
46618,
13,
3277,
433,
596,
2884,
1314,
11,
433,
649,
387,
779,
37154,
13,
358,
3152,
11,
682,
37547,
26863,
527,
2216,
37154,
13,
2030,
1070,
596,
2555,
922,
279,
13339,
46618,
430,
923,
264,
83374,
2449,
311,
279,
15496,
13,
18156,
11,
3582,
11,
584,
81413,
433,
264,
2766,
810,
2884,
709,
13,
358,
3077,
3779,
6617,
1938,
9498,
62990,
323,
1063,
5066,
14293,
311,
2543,
279,
2466,
32878,
12235,
315,
88010,
1139,
264,
12269,
430,
596,
264,
2697,
2766,
810,
38291,
627,
97213,
3347,
300,
527,
832,
-100
] |
Boston Red Sox tickets are maybe the most desirable tickets in all of sports. With among the most faithful fan bases anywhere, the Red Sox constantly offer out even more games than any various other MLB group. At Tixbag it has been our speciality to serve Red Sox Nation with the best Boston Red Sox Tickets at the best costs. Our choice and also knowledge of Red Sox tickets can not be defeated. Tixbag is additionally comfortably located simply 2 blocks from Fenway Park, providing our clients a free and easy means to grab their Red Sox tickets.
Boston's franchise business was founded as the Boston Americans in 1901, although they would alter their name to the Red Sox in 1907 and also have actually been a sports custom in Beantown ever since. In spite of a notorious title dry spell that started in 1918, the Red Sox appeared in 2004 for a Globe Series title throughout an exceptional play-off run and have actually included 2 even more titles since then, solidifying their red sox standing as one of the premier round clubs of the contemporary era.
It's difficult to match the custom that comes with participating in a Boston Red Sox game, a group that has actually been at the center of the Boston sports scene for decades. The group's background has actually been punctuated by a few of one of the most memorable moments in World Collection history, consisting of Cardinal Enos Slaughter's mad dashboard to home plate in 1946, the "Impossible Dream" of 1967, Carlton Fisk's foul-pole clearing crowning achievement against the Reds in 1975, and also first baseman Expense Buckner's well known error against the Mets in 1986. Boston's listing of greats runs deep, led by Ted Williams, Carl Yastrzemski, Fisk, Jim Rice, Pedro Martinez, David Ortiz, Wade Boggs, Roger Clemens and also Manny Ramirez.
The declared "Curse of the Young child" was front as well as center throughout the 2004 ALCS versus Boston's vowed adversary and also rival, the New York Yankees. Falling back 3 games to none, they resisted history as well as stormed back to win the pennant, taking the last two games of the collection in Yankee Stadium as well as blowing out the Bronx Bombers in Game 7.
TixBag is pleased to give you with the opportunity to cheer on the Boston Red Sox Tickets on Opening Day from virtually anywhere inside Fenway Park, from behind home base to atop the Environment-friendly Beast.
Upon finding the Red Sox Opening Day tickets that meet your demands, press "Buy" to advance to check out. Both new and returning clients will be motivated to sign in with a legitimate e-mail address to complete the purchase online. Make certain to review all transaction details prior to sending the Red Sox Opening Day tickets for handling. When confirmed by the vendor, your order will be covered under our 100% Buyer Assurance.
There are always good deals to be found at Tixbag. The get-in cost, or most affordable cost for a ticket to a Boston Red Sox residence big game, is $8. Costs will certainly change based on lots of factors such as stock and also need, so make certain to obtain your cheap Boston Red Sox tickets before it's too late!
Boston Red Sox tickets are perhaps the most coveted tickets in all of sports. At Tixbag best place to buy red sox tickets it has actually been our specialized to offer Red Sox Nation with the ideal Red Sox tickets at the ideal prices. The get-in cost, or lowest cost for a ticket to a Boston Red Sox residence big game, is $8. Rates will change based on numerous variables such as inventory and need, so be certain to get your cheap Boston Red Sox tickets before it's as well late!
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 4,518 |
[
128000,
65432,
3816,
39645,
14741,
527,
7344,
279,
1455,
35946,
14741,
304,
682,
315,
10034,
13,
3161,
4315,
279,
1455,
37513,
8571,
23963,
12660,
11,
279,
3816,
39645,
15320,
3085,
704,
1524,
810,
3953,
1109,
904,
5370,
1023,
41217,
1912,
13,
2468,
350,
953,
21946,
433,
706,
1027,
1057,
3361,
488,
311,
8854,
3816,
39645,
17671,
449,
279,
1888,
10406,
3816,
39645,
36715,
520,
279,
1888,
7194,
13,
5751,
5873,
323,
1101,
6677,
315,
3816,
39645,
14741,
649,
539,
387,
24164,
13,
350,
953,
21946,
374,
37938,
45769,
7559,
5042,
220,
17,
10215,
505,
45406,
3195,
5657,
11,
8405,
1057,
8403,
264,
1949,
323,
4228,
3445,
311,
11894,
872,
3816,
39645,
14741,
627,
65432,
596,
19562,
2626,
574,
18538,
439,
279,
10406,
9053,
304,
220,
7028,
16,
11
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
65432,
3816,
39645,
14741,
527,
7344,
279,
1455,
35946,
14741,
304,
682,
315,
10034,
13,
3161,
4315,
279,
1455,
37513,
8571,
23963,
12660,
11,
279,
3816,
39645,
15320,
3085,
704,
1524,
810,
3953,
1109,
904,
5370,
1023,
41217,
1912,
13,
2468,
350,
953,
21946,
433,
706,
1027,
1057,
3361,
488,
311,
8854,
3816,
39645,
17671,
449,
279,
1888,
10406,
3816,
39645,
36715,
520,
279,
1888,
7194,
13,
5751,
5873,
323,
1101,
6677,
315,
3816,
39645,
14741,
649,
539,
387,
24164,
13,
350,
953,
21946,
374,
37938,
45769,
7559,
5042,
220,
17,
10215,
505,
45406,
3195,
5657,
11,
8405,
1057,
8403,
264,
1949,
323,
4228,
3445,
311,
11894,
872,
3816,
39645,
14741,
627,
65432,
596,
19562,
2626,
574,
18538,
439,
279,
10406,
9053,
304,
220,
7028,
16,
11,
-100
] |
Read, Listen, & Watch
Immerse yourself in sustainable media
How to Let Go of the World and Love All The Things Climate Can't Change
Helmed by Josh Fox, this stunning documentary brings viewers on a journey around the world to uncover what climate change can't destroy.
Cradle to Cradle: Remaking the Way We Make Things
Author: Edward Abbey
Braungart's visionary novel argues that the highly popularized "reduce, reuse recycle" mantra perpetuates a one-way, "cradle to grave" manufacturing model. Abbey proposes a different solution: altering the very composition of the products we produce.
Author: Jonathan Safran Foer
Part memoir and part investigative report, Eating Animals is the groundbreaking moral examination of vegetarianism, farming, and the food we eat every day that inspired the documentary of the same name.
Drawdown: The Most Comprehensive Plan Ever Proposed to Reverse Global Warming
Author: Paul Hawken
Drawdown offers an in depth look at the 100 most substantive solutions to global warming, covering its history, the carbon impact that it provides, the relative cost, and more.
Sand County Almanac
Author: Aldo Leopold
Divided into a series of essays, A Sand Count Almanac envelops readers with scenic descriptions of nature and conservation methodologies. Here, Leopold blends natural history, nature writing, and scientific philosophy in a true mosaic of literary distinction.
Rust Belt Arcana
Authors: Matt Stansberry & David Wilson
By juxtaposing the characteristics of the cards of the Tarot's Major Arcana to the creatures and plants around us, Stansberry and Wilson weave tales of resilience in the industrial Midwest.
The End of Nature
Author: Bill McKibben
A classic read in environmental literature, McKibben's impassioned plea for radical change still resonates today. Proposing that the survival of the globe is dependent on a philosophical shift in human nature is more relevant today than ever.
Author: Rachel Carson
Marine biologist and conservationist Rachel Carson captured the attention of the world in the late 1950s with her research on synthetic pesticides. The result of which was Silent Spring, which eventually spurred a nationwide ban on DDT for agricultural uses.
Rising: Dispatches from the New American Shore
Author: Elizabeth Rush
Sea level rise is one of the hallmarks of climate change and has already changed the coastline of the United States in remarkable ways. Weaving testimonials with wildlife profiles, Rising reminds us that climate change is neither distant nor imagined.
Author: Naomi Klein
This Changes Everything is a call to action against disillusionment. Through an economic lens, this novel analyses the profit and growth associated with free markets and connects them to environmental degradation.
The Everglades: River of Grass
Author: Marjory Stoneman Douglas
First published in 1947, Marjory Stoneman Douglas' emphatic descriptions of the Everglades are what brought global attention to one of Florida's most cherished natural areas.
With sharp language and a powerful perspective, Abbey scrutinizes the urbanization of America's landscapes. Desert Solitaire manages to balance frustration with appreciation, as much as a call to action as it is an ode to the magnificent vastness of canyon land.
Authors: Douglas Adams & Mark Carwardine
Both hilarious and profound, as only Douglas Adams can be, Last Chance to See takes readers on a journey through the animal kingdom, bringing them not only face to face with the natural world, but it's imminent peril.
Cowspiracy
Released in 2014, this documentary reveals the impact of animal agriculture on the environment. Focusing on deforestation, water consumption, and greenhouse gases, the film displays the consequences of animal agriculture, and the corporate attempts to mask these truths.
Ten years after An Inconvenient Truth brought climate change into the forefront of mainstream culture, this compelling follow-up shows that while the stakes have never been higher, the solutions to the climate crisis are still within our reach.
Get a glimpse of the lives of concerned citizens living at the frontiers of extreme oil and gas extraction, from the mayor of an Inuit village in Canada's high arctic, to a river conservationist on the shores of the Colorado River.
Perhaps one of the most influential films of the 21st century, An Inconvenient Truth follows former vice president Al Gore on his quest to alert the public to the concerns of climate change. Using slideshows, and real-life footage, this film illustrates a powerful message.
Rising Tides explores the reality of climate change and rising seas, and the effects it will have on coastal communities through interviews with scientists, nonprofits, homeowners, government officials, and other groups offering potentail solutions.
This climate change documentary follows Leonardo DiCaprio on a tumultuous journey around the world. Deeply ambitious and undeniably sincere, Before the Flood focused on the current threats facing the planet, as well as actions individuals can take to reduce their impact.
This groundbreaking TV series was filmed over four years across 64 different countries. David Attenborough celebrates the amazing variety of the natural world in an epic documentary series.
A challenging, and sometimes disheartening series, Our Planet examines the beauty of our world, but also how deeply climate change impacts all living creatures. Some scenes are notably graphic, and viewer discretion is advised.
Gasland follows filmmaker Josh Fox through a series of interviews in communities being affected by natural gas drilling. By drawing connections between drilling, water contamination, air pollution, and chronic health problems, Fox exposes a dark side of "fracking".
The True Cost invites viewers into the lives of the people and places behind our clothes, pulling back the curtain on the untold story of production, and asking us: who really pays the price for our clothing?
Pixar's hit animated feature Wall-e follows an exceptionally curious robot responsible for compacting trash on an Earth that humans have left uninhabitable in a dystopian future. At its core, Wall-e is a call to action that can be enjoyed by kids and adults alike.
Biggest Little Farm
In 2011 John and Molly Chester left their jobs in Los Angeles to start Apricot Lanes farms in 2011. Follow their journey to create a farm with a well-balanced eco-system while treating the environment and animals with respect.
Podcasts & Talk Shows
Dive into one of the curiously delightful conversations overheard around National Geographic's headquarters. You'll be introduced to the explorers, photographers and scientists at the edges of our big, bizarre, and beautiful world.
Recommended Episodes:
- "The Hidden Cost of the Perfect Selfie"
An independently produced radio show created by Roman Mars that focuses on design and architecture.
- "Oyster-tecture"
- "A Sea Worth its Salt"
Science Vs.
Science Vs takes on fads, trends, and the opinionated mob to find out what's fact, what's not, and what's somewhere in between.
- "Climate Change...the Apocalypse?"
- "Vegans: Are They Right?"
- "Plastics: The Final Straw?"
- "The Bee-pocalypse"
The 2020 presidential race is not just about who's going to take on President Trump--it's about the future of our country. And it all starts in Iowa. Climate change has become a key issue for Democratic voters. See how the climate crisis is already affecting the everyday lives of Iowans and how they're voting.
Recommended Episode:
- "The climate crisis hits Iowa"
Costing the Earth
A program on BBC Radio 4 looking at man's effect on the environment, questioning accepted truths, and reporting on progress towards improving the world.
- "Soil Saviors"
- "Cruising: A Dirty Secret"
An American comedy podcasting network founded by Scott Aukerman and Jeff Ullrich in August 2010.
- "Who is Enforcing the Clean Water Act?" With Bobby Kennedy Jr.
- "What's Really Going on with Recycling Hunny?" With Beth Porter
- "What's The Cutest Way to Fight Climate Change?" With Anasasia Khoo
The past is never past. Every headline has a history. Join us every week as we go back in time to understand the present. These are stories you can feel and sounds you can see from the moments that shaped our world.
- "The Grid"
- "A year of wonder"
- "The litter myth"
An award-winning podcast and video series published by Stuff Media and hosted by Josh Clark and Charles W..
- "The Ins and Outs of Bee Keeping"
- "How Environmental Psychology Works"
- "We Are Running Out of Sand and it Actually Matters"
Energy Gang
A weekly digest on energy, clean tech and the environment produced by Green tech Media. They delve into the technological, political and market forces driving energy and environmental issues.
- "From the Vault: Elon Musk and the Grid Fan Fiction"
- "What's Changed Most in Energy Over the Last Four Years?"
Radiolab is a radio program produced by WNYC, a public radio station in New York City, that investigates this strange world.
- "From Tree to Shining Tree"
Hear thought-provoking ideas on every subject imaginable - from AI to Zoology - given by the world's leading thinkers and doers.
- "The biggest risks facing cities - and some solutions" with Robert Muggah
- "Why wildfires have gotten worse - and what we can do about it" with Paul Hessburg
*These suggestions are cultivated from recommendations by Sustainability Initiatives employees as well as students, faculty, and staff like you. The university does not officially endorse or benefit from works included in the Read, Listen and Watch lists.
For more publications, visit the Center for Global Economic and Environmental Opportunity's website.
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 3,860 |
[
128000,
4518,
11,
33249,
11,
612,
10573,
16067,
198,
1453,
1195,
325,
6261,
304,
22556,
3772,
198,
4438,
311,
6914,
6122,
315,
279,
4435,
323,
10919,
2052,
578,
20695,
31636,
3053,
956,
10604,
198,
33813,
2106,
555,
18747,
13585,
11,
420,
20441,
25999,
12716,
22511,
389,
264,
11879,
2212,
279,
1917,
311,
45063,
1148,
10182,
2349,
649,
956,
7066,
627,
34,
13584,
273,
311,
4656,
72455,
25,
5031,
1802,
279,
12424,
1226,
7557,
20695,
198,
7279,
25,
22653,
65555,
198,
33,
969,
2234,
472,
596,
87490,
11775,
29633,
430,
279,
7701,
5526,
1534,
330,
27369,
11,
27068,
61843,
1,
68009,
22313,
52085,
264,
832,
27896,
11,
330,
5192,
72455,
311,
25165,
1,
15266,
1646,
13,
65555,
55725,
264,
2204,
6425,
25,
60923,
279,
1633,
18528,
315,
279,
3956
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
4518,
11,
33249,
11,
612,
10573,
16067,
198,
1453,
1195,
325,
6261,
304,
22556,
3772,
198,
4438,
311,
6914,
6122,
315,
279,
4435,
323,
10919,
2052,
578,
20695,
31636,
3053,
956,
10604,
198,
33813,
2106,
555,
18747,
13585,
11,
420,
20441,
25999,
12716,
22511,
389,
264,
11879,
2212,
279,
1917,
311,
45063,
1148,
10182,
2349,
649,
956,
7066,
627,
34,
13584,
273,
311,
4656,
72455,
25,
5031,
1802,
279,
12424,
1226,
7557,
20695,
198,
7279,
25,
22653,
65555,
198,
33,
969,
2234,
472,
596,
87490,
11775,
29633,
430,
279,
7701,
5526,
1534,
330,
27369,
11,
27068,
61843,
1,
68009,
22313,
52085,
264,
832,
27896,
11,
330,
5192,
72455,
311,
25165,
1,
15266,
1646,
13,
65555,
55725,
264,
2204,
6425,
25,
60923,
279,
1633,
18528,
315,
279,
3956,
-100
] |
Conferences \ Advances in Drug Discovery \ Academic Drug Discovery \ Agenda \ Per Arvidsson
Per Arvidsson's Biography
Per Arvidsson, Director, Karolinska Institutet
Per I Arvidsson is Director of the national Swedish Drug Discovery & Development Platform at Science for Life Laboratory (SciLifeLab DDD). He is also fractional research professor at the College of Heath Science, UKZN, South Africa. Before joining SciLifeLab & Karolinska Institutet in 2013 to set up the SciLifeLab DDD platform, Prof. Arvidsson held various roles at the CNS & Pain iMED at AstraZeneca, Södertälje, the last being Director for late stage pre-clinical and clinical neuroscience drug discovery. His accolades include several professorship positions at Uppsala University in Sweden and postdoctoral studies with Prof. D. Seebach at ETH Zurich, Switzerland. Prof. Arvidsson is named inventor on some 15+ patent applications, and co-author to some 100 peer-reviewed publications, two of which have won "most cited papers" awards.
Open for Collaboration – An Academic Platform for Drug Discovery & Development at SciLifeLab
Monday, 6 March 2017 at 16:30
Add to Calendar ▼2017-03-06 16:30:002017-03-06 17:30:00Europe/LondonOpen for Collaboration – An Academic Platform for Drug Discovery and Development at [email protected]
SciLifeLab DDD is a newly established national Swedish infrastructure for academic drug discovery. I will describe the build-up, operation, and international collaboration of the platform, with the ambition to share learnings and best practice with academic drug discovery centers globally.
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 3,094 |
[
128000,
1128,
5006,
1144,
91958,
304,
26166,
39193,
1144,
42170,
26166,
39193,
1144,
58662,
1144,
3700,
1676,
1325,
31031,
198,
3976,
1676,
1325,
31031,
596,
79269,
198,
3976,
1676,
1325,
31031,
11,
10783,
11,
13528,
337,
1354,
4657,
96562,
295,
198,
3976,
358,
1676,
1325,
31031,
374,
10783,
315,
279,
5426,
31209,
26166,
39193,
612,
11050,
17916,
520,
10170,
369,
9601,
32184,
320,
91792,
26833,
30146,
423,
4195,
570,
1283,
374,
1101,
69309,
3495,
14561,
520,
279,
9304,
315,
48562,
10170,
11,
6560,
45947,
11,
4987,
10384,
13,
13538,
18667,
41472,
26833,
30146,
612,
13528,
337,
1354,
4657,
96562,
295,
304,
220,
679,
18,
311,
743,
709,
279,
41472,
26833,
30146,
423,
4195,
5452,
11,
8626,
13,
1676,
1325,
31031,
5762,
5370,
13073,
520,
279,
93643,
612,
35567
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
1128,
5006,
1144,
91958,
304,
26166,
39193,
1144,
42170,
26166,
39193,
1144,
58662,
1144,
3700,
1676,
1325,
31031,
198,
3976,
1676,
1325,
31031,
596,
79269,
198,
3976,
1676,
1325,
31031,
11,
10783,
11,
13528,
337,
1354,
4657,
96562,
295,
198,
3976,
358,
1676,
1325,
31031,
374,
10783,
315,
279,
5426,
31209,
26166,
39193,
612,
11050,
17916,
520,
10170,
369,
9601,
32184,
320,
91792,
26833,
30146,
423,
4195,
570,
1283,
374,
1101,
69309,
3495,
14561,
520,
279,
9304,
315,
48562,
10170,
11,
6560,
45947,
11,
4987,
10384,
13,
13538,
18667,
41472,
26833,
30146,
612,
13528,
337,
1354,
4657,
96562,
295,
304,
220,
679,
18,
311,
743,
709,
279,
41472,
26833,
30146,
423,
4195,
5452,
11,
8626,
13,
1676,
1325,
31031,
5762,
5370,
13073,
520,
279,
93643,
612,
35567,
-100
] |
Western Colorado – Colorado Mesa University President Tim Foster announced the Montrose County School District and School District 51 concurrent enrollment nurse aide programs received approval from the State Board of Nursing. The announcement represents the latest work force training milestone for Western Colorado Community College, a division of CMU.
On January 23, 2019 the Colorado Department of Regulatory agencies informed WCCC that the Montrose program had received approval and sanctioning. The news was welcomed from the program administer and CMU Montrose Campus Director Gary Ratcliff, Ph.D.
The same program administered at the WCCC Grand Junction campus is for both high school students and adults and also received state board reaffirmation. The Grand Junction program enrolled 50 students from Mesa County School District 51 in 2018.
WCCC Vice President of Community College Affairs, Brigitte Sundermann believes the Montrose and Grand Junction programs prepare high school students for employment in long-term health care facilities, hospitals, skilled nursing facilities, home health care, assisted living, hospice and other health care related professions.
"Students who complete the courses receive the necessary knowledge, skills and abilities to perform as a competent nurse aide and prepare to test in the National Nurse Aide Assessment Program to be certified," said Sundermann.
The regulatory validations are a part of a larger effort at CMU to create expanded learning around health care. CMU offers The Master of Physician Assistant Studies as well as other advanced health care degrees, including the Master of Science in Athletic Training and the Doctor of Occupational Therapy. WCCC offers complementary programs like nurse aide and medical assisting that offer the local health care community a top-to-bottom skill-ready workforce in western Colorado.
"CMU and WCCC have really created an eco-system of health care training that matches the needs of local health care providers. From health clubs and rehabilitation centers to clinics and hospitals, WCCC and CMU students will be the future providers of health care in a community that has historically experienced shortages in many related fields," said Foster.
Do you have a story you would like to share with Colorado Mesa University? Please share your good news by contacting Katlin Birdsall at [email protected].
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 3,574 |
[
128000,
55924,
15745,
1389,
15745,
55965,
3907,
4900,
9538,
39274,
7376,
279,
9995,
25888,
6406,
6150,
11182,
323,
6150,
11182,
220,
3971,
35135,
39148,
29198,
41675,
7620,
4036,
14765,
505,
279,
3314,
8925,
315,
46353,
13,
578,
17480,
11105,
279,
5652,
990,
5457,
4967,
43291,
369,
11104,
15745,
12332,
9304,
11,
264,
13096,
315,
18582,
52,
627,
1966,
6186,
220,
1419,
11,
220,
679,
24,
279,
15745,
6011,
315,
69822,
13607,
16369,
468,
54973,
430,
279,
9995,
25888,
2068,
1047,
4036,
14765,
323,
45361,
287,
13,
578,
3754,
574,
32069,
505,
279,
2068,
27185,
323,
18582,
52,
9995,
25888,
39680,
10783,
24765,
41677,
566,
3168,
11,
2405,
920,
627,
791,
1890,
2068,
38018,
520,
279,
468,
54973,
10517,
69056,
15679,
374,
369,
2225,
1579,
2978,
4236,
323,
12884
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
55924,
15745,
1389,
15745,
55965,
3907,
4900,
9538,
39274,
7376,
279,
9995,
25888,
6406,
6150,
11182,
323,
6150,
11182,
220,
3971,
35135,
39148,
29198,
41675,
7620,
4036,
14765,
505,
279,
3314,
8925,
315,
46353,
13,
578,
17480,
11105,
279,
5652,
990,
5457,
4967,
43291,
369,
11104,
15745,
12332,
9304,
11,
264,
13096,
315,
18582,
52,
627,
1966,
6186,
220,
1419,
11,
220,
679,
24,
279,
15745,
6011,
315,
69822,
13607,
16369,
468,
54973,
430,
279,
9995,
25888,
2068,
1047,
4036,
14765,
323,
45361,
287,
13,
578,
3754,
574,
32069,
505,
279,
2068,
27185,
323,
18582,
52,
9995,
25888,
39680,
10783,
24765,
41677,
566,
3168,
11,
2405,
920,
627,
791,
1890,
2068,
38018,
520,
279,
468,
54973,
10517,
69056,
15679,
374,
369,
2225,
1579,
2978,
4236,
323,
12884,
-100
] |
On Saturday night, Olympic gold medalist Gabby Douglas will get the cheesy Lifetime movie treatment with The Gabby Douglas Story. The trailer just hit this Internet and while it looks camp, I'm excited to see Imani Hakim from Everybody Hates Chris in the lead role. Regina King and S. Epatha Merkerson are also in the film as Gabby's mother and grandmother.
The Douglas Story looks like the perfect family movie to watch with grandma — old ladies love Gabby — except some who have seen the made-for-TV film say the script leaves out one big piece, race.
Douglas overcame a lot—that's not up for debate. But she overcame quite a bit more that's not in the movie. The most glaring omission from her story is seemingly the most obvious one: It's impossible to ignore that Douglas is a black woman competing in a sport that is predominantly white. Even while Douglas was a favored presence at the games, her race was an issue of discussion, ranging from NBC's botched coverage of her time in London to arguments about her hair. Douglas has not been silent on the issue, either. In a 2012 interview with Oprah Winfrey, Douglas explains she felt so isolated because of the racism she experienced at her original Virginia Beach gym, it was one of the reasons for her move to Iowa.
Douglas' ethnicity is reportedly mentioned at least once, when her mom drops her off with an Iowa host family so she can train with coach Liang Chow. It is then that Gabby realizes that she'll probably be the only chocolate drop at the gym. But outside of that scene, the idea that Gabby is a woman of color from a poor family in a predominantly caucasian sport, which has made a habit of ostracizing competitors of color, is odd. But I guess Lifetime wasn't trying to rile up grandma.
Simone Biles Vs. The Racists: Are Black Gymnasts The New Black QBs?
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 9,334 |
[
128000,
1966,
7884,
3814,
11,
25944,
6761,
37712,
380,
24664,
1729,
31164,
690,
636,
279,
87847,
65452,
5818,
6514,
449,
578,
24664,
1729,
31164,
15457,
13,
578,
19809,
1120,
4295,
420,
8191,
323,
1418,
433,
5992,
3190,
11,
358,
2846,
12304,
311,
1518,
358,
78298,
72398,
318,
505,
51898,
473,
988,
11517,
304,
279,
3063,
3560,
13,
70780,
6342,
323,
328,
13,
469,
2398,
64,
108227,
1293,
527,
1101,
304,
279,
4632,
439,
24664,
1729,
596,
6691,
323,
39284,
627,
791,
31164,
15457,
5992,
1093,
279,
4832,
3070,
5818,
311,
3821,
449,
83777,
2001,
2362,
23628,
3021,
24664,
1729,
2001,
3734,
1063,
889,
617,
3970,
279,
1903,
15548,
53392,
4632,
2019,
279,
5429,
11141,
704,
832,
2466,
6710,
11,
7102,
627,
92741,
27481,
927,
6142,
264,
2763,
41128
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
1966,
7884,
3814,
11,
25944,
6761,
37712,
380,
24664,
1729,
31164,
690,
636,
279,
87847,
65452,
5818,
6514,
449,
578,
24664,
1729,
31164,
15457,
13,
578,
19809,
1120,
4295,
420,
8191,
323,
1418,
433,
5992,
3190,
11,
358,
2846,
12304,
311,
1518,
358,
78298,
72398,
318,
505,
51898,
473,
988,
11517,
304,
279,
3063,
3560,
13,
70780,
6342,
323,
328,
13,
469,
2398,
64,
108227,
1293,
527,
1101,
304,
279,
4632,
439,
24664,
1729,
596,
6691,
323,
39284,
627,
791,
31164,
15457,
5992,
1093,
279,
4832,
3070,
5818,
311,
3821,
449,
83777,
2001,
2362,
23628,
3021,
24664,
1729,
2001,
3734,
1063,
889,
617,
3970,
279,
1903,
15548,
53392,
4632,
2019,
279,
5429,
11141,
704,
832,
2466,
6710,
11,
7102,
627,
92741,
27481,
927,
6142,
264,
2763,
41128,
-100
] |
3rd Annual Bluegrass Anonymous Pick Inn November 24th | Louisville, Kentucky | Joe Hayden Real Estate Team - Your Real Estate Experts!
The Louisville Bluegrass Music Association will be picking and grinning for a good cause at their yearly benefit show on November 24th. The 3rd Annual Bluegrass Anonymous Holiday Pick Inn will be held at the Holiday Inn Hurstbourne Lane location in Louisville. Proceeds from the show benefit the Home of the Innocents and the Center for Women and Families. The festivities and raffles will begin at 3 p.m. with an affordable admission of only one new toy or a donation.
The spirit of giving comes to life at this event as the musicians gather in hopes of entertaining the crowds and giving less fortunate children holiday gifts. The music will begin at 5 p.m. with Mike Cleveland and Friends, Whiskey Bent Valley Boys, Brushfire and Murphy's Echo, and musicians will continue to gather and play all night long. Discounted rooms are available in advance for those wishing to participate or just enjoy the performances.
Bluegrass Anonymous is Louisville's Bluegrass Music Association and being a not-for-profit organization, they take music to heart. This collective of talented musicians perform and promote the bluegrass music that is deeply rooted in Kentucky's history. The Bluegrass Anonymous organizes yearly events to support local charities such as the Pick Inn this November 24th at the Hurstbourne Lane Holiday Inn. Families and friends can come out and enjoy some picking that will have you grinning for a good cause.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 7,204 |
[
128000,
18,
6634,
25992,
8868,
54454,
31537,
20305,
17382,
6841,
220,
1187,
339,
765,
46134,
11,
26036,
765,
13142,
77285,
8976,
25447,
8068,
482,
4718,
8976,
25447,
51859,
4999,
791,
46134,
8868,
54454,
10948,
10229,
690,
387,
21816,
323,
1099,
20353,
369,
264,
1695,
5353,
520,
872,
45370,
8935,
1501,
389,
6841,
220,
1187,
339,
13,
578,
220,
18,
6634,
25992,
8868,
54454,
31537,
32769,
20305,
17382,
690,
387,
5762,
520,
279,
32769,
17382,
21670,
267,
22152,
27109,
3813,
304,
46134,
13,
41870,
82,
505,
279,
1501,
8935,
279,
5492,
315,
279,
84593,
812,
323,
279,
5955,
369,
11215,
323,
50556,
13,
578,
80459,
323,
436,
81635,
690,
3240,
520,
220,
18,
281,
749,
13,
449,
459,
17049,
26360,
315,
1193,
832,
502,
22068,
477,
264,
25968,
627
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
18,
6634,
25992,
8868,
54454,
31537,
20305,
17382,
6841,
220,
1187,
339,
765,
46134,
11,
26036,
765,
13142,
77285,
8976,
25447,
8068,
482,
4718,
8976,
25447,
51859,
4999,
791,
46134,
8868,
54454,
10948,
10229,
690,
387,
21816,
323,
1099,
20353,
369,
264,
1695,
5353,
520,
872,
45370,
8935,
1501,
389,
6841,
220,
1187,
339,
13,
578,
220,
18,
6634,
25992,
8868,
54454,
31537,
32769,
20305,
17382,
690,
387,
5762,
520,
279,
32769,
17382,
21670,
267,
22152,
27109,
3813,
304,
46134,
13,
41870,
82,
505,
279,
1501,
8935,
279,
5492,
315,
279,
84593,
812,
323,
279,
5955,
369,
11215,
323,
50556,
13,
578,
80459,
323,
436,
81635,
690,
3240,
520,
220,
18,
281,
749,
13,
449,
459,
17049,
26360,
315,
1193,
832,
502,
22068,
477,
264,
25968,
627,
-100
] |
Children's International School Moss
Sign In Register Now
About CIS
Copyright © 2006-2019 Children's International School Moss
The principal on site is responsible for enrolling students in the school as well as placing students in the appropriate class.
The schools office manager coordinates all procedures, paperwork and inquiries related to admissions.
We do our best to include all students, with no discrimination. Students do not have to be able to speak English in order to enroll at CIS. While attending students will be expected to learn it and will be given help until they become independent. We wish to promote effective learning, interest in other cultures and global perspectives, and international mindedness.
The school has no deadline for applications. Students may be admitted at any time during the school year, apart from the month of June.
Children`s International School is a private school and school fees will be charged according to state regulations.
Moss Verk 1
Admissions Contacts
Fredrik Bjørge Hansen
[email protected]
Sangeetha Bysheim
[email protected]
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 4,977 |
[
128000,
11800,
596,
7327,
6150,
44610,
198,
7412,
763,
8618,
4800,
198,
10714,
75880,
198,
13857,
7388,
220,
1049,
21,
12,
679,
24,
15394,
596,
7327,
6150,
44610,
198,
791,
12717,
389,
2816,
374,
8647,
369,
665,
16608,
4236,
304,
279,
2978,
439,
1664,
439,
25012,
4236,
304,
279,
8475,
538,
627,
791,
8853,
5274,
6783,
14259,
682,
16346,
11,
45731,
323,
44983,
5552,
311,
48911,
627,
1687,
656,
1057,
1888,
311,
2997,
682,
4236,
11,
449,
912,
21934,
13,
20783,
656,
539,
617,
311,
387,
3025,
311,
6604,
6498,
304,
2015,
311,
52880,
520,
75880,
13,
6104,
24096,
4236,
690,
387,
3685,
311,
4048,
433,
323,
690,
387,
2728,
1520,
3156,
814,
3719,
9678,
13,
1226,
6562,
311,
12192,
7524,
6975,
11,
2802,
304,
1023,
27833,
323
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
11800,
596,
7327,
6150,
44610,
198,
7412,
763,
8618,
4800,
198,
10714,
75880,
198,
13857,
7388,
220,
1049,
21,
12,
679,
24,
15394,
596,
7327,
6150,
44610,
198,
791,
12717,
389,
2816,
374,
8647,
369,
665,
16608,
4236,
304,
279,
2978,
439,
1664,
439,
25012,
4236,
304,
279,
8475,
538,
627,
791,
8853,
5274,
6783,
14259,
682,
16346,
11,
45731,
323,
44983,
5552,
311,
48911,
627,
1687,
656,
1057,
1888,
311,
2997,
682,
4236,
11,
449,
912,
21934,
13,
20783,
656,
539,
617,
311,
387,
3025,
311,
6604,
6498,
304,
2015,
311,
52880,
520,
75880,
13,
6104,
24096,
4236,
690,
387,
3685,
311,
4048,
433,
323,
690,
387,
2728,
1520,
3156,
814,
3719,
9678,
13,
1226,
6562,
311,
12192,
7524,
6975,
11,
2802,
304,
1023,
27833,
323,
-100
] |
AGEPI became part of the EURAXESS-Moldova Services Network as EURAXESS Local Contact Point
EURAXESS-Moldova Services Network in is part of the European network that connects more than 200 service centers for researchers, who joined the EURAXESS Services Commitment. EURAXESS Services Network is an initiative of the European Commission and the countries participating in the research programs of the European Union. Through the EURAXESS Network is provided information and assistance to foreign researchers who wish to come to the Republic of Moldova, and to those looking for jobs in research in other European countries throughEURAXESS portal and with the support of the National EURAXESS Services Centres and the EURAXESS Local Contact Points. EURAXESS objective is to contribute to removing barriers to the mobility of researchers.EURAXESS-Moldova Services Network in provides free counseling services to researchers from abroad who wish to engage in universities, research institutes or among other companies of the Republic of Moldova.For more information on services provided by the EURAXESS Services Network, available jobs in the research sector, and information about living in another European country, you can contact the national EURAXESS portals of the 40 partner countries and EURAXESS Services Centres.
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 3,204 |
[
128000,
3669,
1932,
6244,
961,
315,
279,
38188,
3027,
10201,
5364,
820,
12949,
8471,
8304,
439,
38188,
3027,
10201,
8949,
9359,
5236,
198,
55838,
3027,
10201,
5364,
820,
12949,
8471,
8304,
304,
374,
961,
315,
279,
7665,
4009,
430,
34161,
810,
1109,
220,
1049,
2532,
19169,
369,
12074,
11,
889,
11096,
279,
38188,
3027,
10201,
8471,
9386,
479,
13,
38188,
3027,
10201,
8471,
8304,
374,
459,
20770,
315,
279,
7665,
9849,
323,
279,
5961,
24435,
304,
279,
3495,
7620,
315,
279,
7665,
9323,
13,
17331,
279,
38188,
3027,
10201,
8304,
374,
3984,
2038,
323,
13291,
311,
7362,
12074,
889,
6562,
311,
2586,
311,
279,
5545,
315,
55263,
12949,
11,
323,
311,
1884,
3411,
369,
7032,
304,
3495,
304,
1023,
7665,
5961,
1555,
55838,
3027,
10201,
24007,
323,
449
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
3669,
1932,
6244,
961,
315,
279,
38188,
3027,
10201,
5364,
820,
12949,
8471,
8304,
439,
38188,
3027,
10201,
8949,
9359,
5236,
198,
55838,
3027,
10201,
5364,
820,
12949,
8471,
8304,
304,
374,
961,
315,
279,
7665,
4009,
430,
34161,
810,
1109,
220,
1049,
2532,
19169,
369,
12074,
11,
889,
11096,
279,
38188,
3027,
10201,
8471,
9386,
479,
13,
38188,
3027,
10201,
8471,
8304,
374,
459,
20770,
315,
279,
7665,
9849,
323,
279,
5961,
24435,
304,
279,
3495,
7620,
315,
279,
7665,
9323,
13,
17331,
279,
38188,
3027,
10201,
8304,
374,
3984,
2038,
323,
13291,
311,
7362,
12074,
889,
6562,
311,
2586,
311,
279,
5545,
315,
55263,
12949,
11,
323,
311,
1884,
3411,
369,
7032,
304,
3495,
304,
1023,
7665,
5961,
1555,
55838,
3027,
10201,
24007,
323,
449,
-100
] |
This was my 2nd night at Ding Dong in a row reviewing bands. Ding Dong Lounge is such a cool little community of creative people! I know it's been a while since the renovations happened upstairs but I couldn't help but remember the first time I saw Skinny Hobos at Ding Dong a few years ago when the stage was in the corner. I mention this because Alex brought it up last night. I love the new and improved band room and with Paul behind the desk you are guaranteed an excellent sound! After reviewing Black Metal and ambient metal bands the night before and then these three Rock n Roll, Blues, Grungy Rock bands all with Paul behind the desk, all of the bands sounded amazing! All the levels were spot on, the drums were clear and the vocals all combined to make all seven bands I saw this weekend sound international quality!
The Saturday night at Ding Dong was opened by West Auckland band AnimalHead. Right from before they took the stage the room was packed with people cheering and calling for the band. Everyone was straight into it, the crowd and the band all feeding off each other in a high energy frenzy. It was one of the most active crowds I have ever seen for an opening band which made me take even more notice of this powerful 3 piece. It was my first time seeing AnimalHead but I have heard their name mentioned a few times. Before I got upstairs to the band room I was feeling very fatigued from the previous night's show but by the end of the first song from AnimalHead my fatigue was gone and I was just as excited and ready to dance like the crowd around me.
AnimalHead are inspired by a lot of different influences and this comes across in their music with not one song being able to be pigeonholed as one particular genre. Watching an AnimalHead set is almost a trip in history of Heavy Rock music with Grunge tones, Jazz, Blues, Heavy Rock boarding on Metal with the heavy drum beats and soaring guitar solos. I loved the drum tones and guitar solos that AnimalHead played, these guys know how to play their instruments and their love and passion shines through. AnimalHead will appeal to anyone who likes any of the sub genres of Rock and Metal and they have the stage presence and attitude to go along with it. Check them out, you won't be disappointed. Their next show is Winterslam back at Ding Dong on the 16th June, get along!
Auckland band Coridian were up next and I can see why this line-up is playing together straight away as Skinny Hobos are the only band I have seen previously. Again, like AnimalHead, I had heard the name Coridian but last night was my first time watching them. Dity is one of the best Rock frontmen I've seen live in a long time – he knows how to work a crowd and is jumping all over the stage right from the first note. I couldn't help but think to myself that I would love to see what he could do on a bigger stage and I would love to see them perform at The Powerstation or similar venue. Dity had the crowd eating out of his hands and by the end of their set the entire front half of the room is jumping up and down.
The set that Coridian played last night was a mix of fast Rock, slower Bluesy tunes with Jazz elements and heavier grunge tones. I don't want to keep drawing similarities with AnimalHead but this is another band that lovers of all heavy music can get into. This show was to celebrate their release of their newest single Reflections which is available on iTunes, Spotify, Google Music, Bandcamp, Soundcloud, Tidal and Deezer with all of these links available on their Facebook page. Reflections is a song that takes the listener on a journey and I loved where it took me. I could relate to most of the songs Coridian performed and my favourites were Reflections, Milk, Brave Disguise and their last song Blind Faith. Brave Disguise spoke to me the most and while writing this review I've already listened to it three times. By the end of their set I was regretting not having checked out their music prior to coming to the show but it is definitely providing me with some amazing music to check already.
Coridian have a few shows coming up get out and support them! They are touring with Wellington band Curlys Jewels in May and June. 26th May they are playing at Lovelands in Wellington (which has already SOLD OUT!) then the 10th of June at the Cabana in Napier and on the 24th June they are playing at Backbeat Bar in Auckland, so get along!
Last, but most definitely not least, was Auckland duo Skinny Hobos. When you think of a musical duo it is very easy to think of acoustic or low key but these are terms that do not apply at all to Skinny Hobos. They experimented with their set last night, playing a set they mentioned they had never played before. They played some of their crowd favourites Jokers and Fools and Merchant of Tirau and but they also mixed it up and the set overall was of a heavier tone. This could have been a risk with the other bands and crowd being more Rockier but I think it was a risk that worked out, I loved it and I wasn't the only one! By the end of their set the room was absolutely jam packed and there were people standing out the door. I looked around during the last song and even the guys standing around with their arms crossed (which seems to be an Auckland thing) were at least either moving their heads or swaying in time to the music. There wasn't one person standing still in that room and I've only seen this a handful of times this year with the 60 or so bands I've seen this year.
Right throughout their set Skinny Hobos had the crowd eating out of their hands and during their slower songs the people right up the front got out their lighters and started swaying their arms back and forwards, during the faster dancier tunes the crowd were all swaying as one and during the heavier bits everyone was head banging as one. It was as if everyone in that room was part of the performance instead of just Alex and Sam on stage. In-between the songs they kept up a steady stream of banter with the crowd and even tried raffling off one of the crowd members, Rodrigo who they did a shout out too who is playing drums in their next video, which I can't wait to view!
There won't be anyone that loves rock and grunge music that doesn't know what happened to Chris Cornell and I've read on social media and heard vocalised by a lot of people that there aren't many people who come close to his range of vocals. Skinny Hobos did a tribute to the great man which started off as a solo by Alex and I got chills and goosebumps while he was singing, it was the closest I've heard anyone come to, to emulating such an iconic voice and Alex nailed it! Talking of covers Skinny Hobos ended their set on a cover of Steriogram's Walkie Talkie Man if you have never seen a drummer rap and play drums at the same time it is something that blew my mind! Sam you are a machine to be able to pull that off! I wouldn't normally talk about bands doing covers and try to focus on their original music as much as possible but these two songs showed their wide range of musical abilities and were both widely different to their usual sound and they absolutely nailed them!
I've seen over 60 bands live this year and these three bands, collectively, put on one of the best live shows I've attended. They blew a lot of bigger bands I've seen out of the water and while I almost felt like I had to drag myself out of the house into the cold dark night, today I am still buzzing with how amazing these bands were. All three put on powerful, high energy performances. Their love for their music shone through and the crowd lapped all of it up. By the end of the three bands I should have been exhausted but I found myself wishing there was a fourth band to play after Skinny Hobos. If you haven't checked out any of these bands live yet, you won't be disappointed! Get out and check them out when you get the chance!
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 3,160 |
[
128000,
2028,
574,
856,
220,
17,
303,
3814,
520,
80473,
53483,
304,
264,
2872,
34988,
21562,
13,
80473,
53483,
50767,
374,
1778,
264,
7155,
2697,
4029,
315,
11782,
1274,
0,
358,
1440,
433,
596,
1027,
264,
1418,
2533,
279,
78664,
7077,
50007,
719,
358,
7846,
956,
1520,
719,
6227,
279,
1176,
892,
358,
5602,
95324,
37876,
437,
520,
80473,
53483,
264,
2478,
1667,
4227,
994,
279,
6566,
574,
304,
279,
9309,
13,
358,
6420,
420,
1606,
8683,
7263,
433,
709,
1566,
3814,
13,
358,
3021,
279,
502,
323,
13241,
7200,
3130,
323,
449,
7043,
4920,
279,
18496,
499,
527,
19883,
459,
9250,
5222,
0,
4740,
34988,
5348,
19757,
323,
35288,
9501,
21562,
279,
3814,
1603,
323,
1243,
1521,
2380,
9305,
308,
15028,
11,
33425,
11,
2895,
2234,
88
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
2028,
574,
856,
220,
17,
303,
3814,
520,
80473,
53483,
304,
264,
2872,
34988,
21562,
13,
80473,
53483,
50767,
374,
1778,
264,
7155,
2697,
4029,
315,
11782,
1274,
0,
358,
1440,
433,
596,
1027,
264,
1418,
2533,
279,
78664,
7077,
50007,
719,
358,
7846,
956,
1520,
719,
6227,
279,
1176,
892,
358,
5602,
95324,
37876,
437,
520,
80473,
53483,
264,
2478,
1667,
4227,
994,
279,
6566,
574,
304,
279,
9309,
13,
358,
6420,
420,
1606,
8683,
7263,
433,
709,
1566,
3814,
13,
358,
3021,
279,
502,
323,
13241,
7200,
3130,
323,
449,
7043,
4920,
279,
18496,
499,
527,
19883,
459,
9250,
5222,
0,
4740,
34988,
5348,
19757,
323,
35288,
9501,
21562,
279,
3814,
1603,
323,
1243,
1521,
2380,
9305,
308,
15028,
11,
33425,
11,
2895,
2234,
88,
-100
] |
Joe Biden to nominate judge Merrick Garland as attorney general
By Eric Tucker and Michael Balsamo
Updated 9:53AM
Biden warns of damage to national security by Trump administration
President-elect Joe Biden gave remarks Dec. 28, 2020 and warned of damage done to the national security apparatus by the Trump administration and roadblocks in communication that could undermine American security.
WASHINGTON (AP) - President-elect Joe Biden has selected Merrick Garland, a federal appeals court judge who in 2016 was snubbed by Republicans for a seat on the Supreme Court, as his attorney general, two people familiar with the selection process said Wednesday.
Biden is expected to announce Garland's appointment on Thursday, along with other senior leaders of the department, including former homeland security adviser Lisa Monaco as deputy attorney general and former Justice Department civil rights chief Vanita Gupta as associate attorney general.
In picking Garland, Biden is turning to an experienced judge who held senior positions at the Justice Department decades ago, including as a supervisor of the prosecution of the 1995 Oklahoma City bombing.
RELATED: Biden's team urges Americans to stay home on Inauguration Day and watch ceremony online
The pick will force Senate Republicans to contend with the nomination of someone they spurned in 2016 — refusing even to hold hearings when a Supreme Court vacancy arose — but Biden may be banking on Garland's credentials and reputation for moderation to ensure confirmation.
Garland was selected over other finalists including Alabama Sen. Doug Jones and former Deputy Attorney General Sally Yates. The people familiar with the process spoke on condition of anonymity.
If confirmed, Garland would confront immediate challenges, including an ongoing criminal tax investigation into Biden's son, Hunter, as well as calls from many Democrats to pursue inquiries into Trump after he leaves office. A special counsel investigation into the origins of the Russia probe also remains open, forcing a new attorney general to decide how to handle it and what to make public.
Garland would also inherit a Justice Department that has endured a tumultuous four years and would likely need to focus on not only civil rights issues and an overhaul of national policing policies after months of mass protests over the deaths of Black Americans at the hand of law enforcement.
U.S. Vice President Joe Biden congratulates Judge Merrick Garland after he was nominated by U.S. President Barack Obama to the Supreme Court in the Rose Garden at the White House, March 16, 2016 in Washington, DC. (Photo by Chip Somodevilla/Getty Ima
It was unclear how Garland's selection would be received by Black and Latino advocates who had advocated for a Black attorney general or for someone with a background in civil rights causes and criminal justice reform.
Garland would also return to a Justice Department radically different than the one he left. The Sept. 11 attacks was years away, the department's national security division had not yet been created and a proliferation of aggressive cyber and counterintelligence threats from foreign adversaries have made counties like China, Russia and North Korea top priorities for federal law enforcement.
But some of the issues from Garland's first stint at the department persist. Tensions between police and minorities, an issue that flared following the 1992 beating of Rodney King in Los Angeles, remain an urgent concern particularly following a summer of racial unrest that roiled American cities after the May killing of George Floyd in Minneapolis.
RELATED: Biden to take oath outside Capitol amid COVID-19 restrictions
And the FBI has confronted a surge in violent from anti-government and racially motivated extremists. That is a familiar threat to Garland, who as a senior Justice Department official in 1995 helped manage the federal government's response to the bombing of a government building in Oklahoma City that killed 168 people. The bomber, Timothy McVeigh, who was later executed.
Garland has called the work the "most important thing I have done" and was known for keeping a framed photo of Oklahoma City's Alfred P. Murrah Federal Building in his courthouse office in Washington.
At the time of the bombing, Garland was 42 and principal associate deputy attorney general, a top lieutenant to Attorney General Janet Reno. He was chosen to go to Oklahoma City, the highest-ranking Justice Department official there, and led the prosecution for a month until a permanent lead prosecutor was named.
Garland was selected over other contenders for the job including former Alabama senator Doug Jones, who lost his Senate seat last month, and former Deputy Attorney General Sally Yates.
It is rare but not unprecedented for attorneys general to have previously served as judges. It happened in 2007 when President George W. Bush picked Michael Mukasey, a former federal judge in Manhattan, for the job. Eric Holder, President Barack Obama's first attorney general, had also previously been a Superior Court judge.
RELATED: Biden taps Susan Rice as domestic policy adviser, Denis McDonough for VA
Garland was put forward by former President Barack Obama for a seat on the Supreme Court in 2016 following the death of Justice Antonin Scalia, but Republicans refused to hold hearings in the final year of Obama's term. The vacancy was later filled by Justice Neil Gorsuch during the Trump administration.
Republican Senate Majority Leader Mitch McConnell refused to let the nomination move forward in the Senate in the final months of Obama's tenure. He was criticized by Democrats this fall when he took the opposite approach toward confirming President Donald Trump's Supreme Court pick, Amy Coney Barrett. He said the difference this time around was that the White House and Senate were controlled by the same political parties.
One year later, after the firing of FBI Director James Comey, McConnell actually floated Garland's name as a replacement for that position, though Garland was said to be not interested.
Garland has been on the federal appeals court in Washington since 1997. Before that, he had worked in private practice, as well as a federal prosecutor, a senior official in the Justice Department's criminal division and as the principle associate deputy attorney general.
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 7,719 |
[
128000,
41444,
38180,
311,
88400,
11913,
48961,
875,
80074,
439,
14065,
4689,
198,
1383,
16645,
56256,
323,
8096,
426,
1147,
21781,
198,
16593,
220,
24,
25,
4331,
1428,
198,
33,
12770,
49140,
315,
5674,
311,
5426,
4868,
555,
3420,
8735,
198,
27229,
37081,
13142,
38180,
6688,
24208,
3799,
13,
220,
1591,
11,
220,
2366,
15,
323,
19201,
315,
5674,
2884,
311,
279,
5426,
4868,
41705,
555,
279,
3420,
8735,
323,
5754,
22692,
304,
10758,
430,
1436,
43047,
3778,
4868,
627,
51577,
320,
2599,
8,
482,
4900,
37081,
13142,
38180,
706,
4183,
48961,
875,
80074,
11,
264,
6918,
35730,
5590,
11913,
889,
304,
220,
679,
21,
574,
4224,
392,
2788,
555,
13063,
369,
264,
10954,
389,
279,
13814,
7301,
11,
439,
813,
14065,
4689,
11,
1403,
1274,
11537,
449
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
41444,
38180,
311,
88400,
11913,
48961,
875,
80074,
439,
14065,
4689,
198,
1383,
16645,
56256,
323,
8096,
426,
1147,
21781,
198,
16593,
220,
24,
25,
4331,
1428,
198,
33,
12770,
49140,
315,
5674,
311,
5426,
4868,
555,
3420,
8735,
198,
27229,
37081,
13142,
38180,
6688,
24208,
3799,
13,
220,
1591,
11,
220,
2366,
15,
323,
19201,
315,
5674,
2884,
311,
279,
5426,
4868,
41705,
555,
279,
3420,
8735,
323,
5754,
22692,
304,
10758,
430,
1436,
43047,
3778,
4868,
627,
51577,
320,
2599,
8,
482,
4900,
37081,
13142,
38180,
706,
4183,
48961,
875,
80074,
11,
264,
6918,
35730,
5590,
11913,
889,
304,
220,
679,
21,
574,
4224,
392,
2788,
555,
13063,
369,
264,
10954,
389,
279,
13814,
7301,
11,
439,
813,
14065,
4689,
11,
1403,
1274,
11537,
449,
-100
] |
NATS' project team discovered Christie MicroTiles during their quest for a flexible system that would display multiple sources in high definition. Having contacted the manufacturer, Christie hosted a demo at its EMEA HQ in Berks, where the NATS representatives were presented with various options.
In addition, he says, the walls enable slick display of multiple sources and data as required. With similarly configured 8 x 6 curved video walls in each Centre, operatives are able to display 2K x 2K data onto the wall from the ATC controller positions, as well as take 1080p and 800 x 600 data feeds.
In fact this versatile video processor acts as a universal routing switcher, with the capability of simultaneous, real-time, full frame rate monitoring of all inputs.
Each of the three facilities allows up to 20 people to interact in a simulated ATC environment — connected together over a WAN network, with each space viewable from the other facilities, while the ability to host WebEx meetings and video conferencing allows external participants, not present in the room, to take part.
By working with stakeholders, requirements can now be developed more quickly, and this can drive lower development costs from suppliers. "It will also allow 'plug and play' of new technology to enable suppliers to market their new tools," states Rob Langan.
While MicroTiles has provided the three facilities with a high degree of flexibility and visual acuity, NATS is also reassured by the system's ease of service, low cost maintenance — and reconfigurability, should it be required.
Summarizes Mark Race, "Above all, the speed with which we can replicate what the controller sees, and overlay bits of data to show what it would be like even before it's integrated into the main simulation software, is paramount — and it can all be done quickly.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 3,105 |
[
128000,
45,
50207,
6,
2447,
2128,
11352,
44406,
18654,
59465,
2391,
872,
2271,
369,
264,
19303,
1887,
430,
1053,
3113,
5361,
8336,
304,
1579,
7419,
13,
20636,
25559,
279,
14290,
11,
44406,
21685,
264,
17074,
520,
1202,
469,
91467,
46355,
304,
9084,
2857,
11,
1405,
279,
18749,
50,
24005,
1051,
10666,
449,
5370,
2671,
627,
644,
5369,
11,
568,
2795,
11,
279,
14620,
7431,
50738,
3113,
315,
5361,
8336,
323,
828,
439,
2631,
13,
3161,
30293,
20336,
220,
23,
865,
220,
21,
50264,
2835,
14620,
304,
1855,
14821,
11,
73290,
527,
3025,
311,
3113,
220,
17,
42,
865,
220,
17,
42,
828,
8800,
279,
7147,
505,
279,
7520,
34,
6597,
10093,
11,
439,
1664,
439,
1935,
220,
6640,
15,
79,
323,
220,
4728,
865,
220,
5067,
828,
35496
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
45,
50207,
6,
2447,
2128,
11352,
44406,
18654,
59465,
2391,
872,
2271,
369,
264,
19303,
1887,
430,
1053,
3113,
5361,
8336,
304,
1579,
7419,
13,
20636,
25559,
279,
14290,
11,
44406,
21685,
264,
17074,
520,
1202,
469,
91467,
46355,
304,
9084,
2857,
11,
1405,
279,
18749,
50,
24005,
1051,
10666,
449,
5370,
2671,
627,
644,
5369,
11,
568,
2795,
11,
279,
14620,
7431,
50738,
3113,
315,
5361,
8336,
323,
828,
439,
2631,
13,
3161,
30293,
20336,
220,
23,
865,
220,
21,
50264,
2835,
14620,
304,
1855,
14821,
11,
73290,
527,
3025,
311,
3113,
220,
17,
42,
865,
220,
17,
42,
828,
8800,
279,
7147,
505,
279,
7520,
34,
6597,
10093,
11,
439,
1664,
439,
1935,
220,
6640,
15,
79,
323,
220,
4728,
865,
220,
5067,
828,
35496,
-100
] |
House of Cards Season 3, Chapter 28 – Frank Underwood
February 27, 2015 by FullCustodyDad / Fred Campos
I have taken the day off to watch Netflix's House of Cards Season 3. Spoiler alert, please don't read any further if you haven't see the episode or watched the show. This post is dedicated to my House of Cards Marathon, Season 3, Chapter 28. Enjoy and share in the discussion!
House of Cards Season 3, Chapter 28 Summary – Frank Underwood Decides No Re-Election
President Underwood (Keven Spacey) tries hard to rally his party to back his American Works bill. Instead of hearing him, they propose that he not seek re-election in 2016 in an effort to unite the party. Things aren't going well for Claire (Robin Wright) either. Claire's U.N. ambassador bid runs into trouble when she makes a miss quote during her confirmation hearing.
Frank and Claire work the phones in an effort to get support for both their causes. By evening Frank has gotten no supporters and wrecks his desk lamp and sits exhausted on the floor. Claire shows support by engaging him in encouragement sex. They both start their next day confident but Claire loses the confirmation by two votes. She shows great courage and speaks well to the press.
Frank then comes up with an idea to address his plan directly to the public. Frank finally agrees with his Democratic party, of not seeking re-election, but does an excellent job of pushing his American Works agenda in a bold address to the American people.
Doug Stamper (Michael Kelly) still feels lost not working for the President and continues to turn to drugs and alcohol. The President asks Seth Grayson (Derick Cicil) to visit and check in on Doug. Seth appears to show up at the right time to keep Doug from taking his addiction too far. Jackie Sharp (Molly Parker) is unhappy with her position as Whip. She is working Remy Danton (Mahershala Ali) to consider her for a Vice-President nominee.
I've always said, that 'Power is more important that money.' But when it comes to elections, money gives power… well a run for it's money. … We have to reverse our thinking. When the wind is blowing at Gail forces, there is no point in sailing against it." – Frances Underwood
Fred Campos' Thoughts on the Episode
This chapter has good character development and a very realistic twist in that Claire doesn't automatically win her confirmation as an U.N. Ambassador. The show makes me feel sorry for Claire loosing only by two vote (52-48), but I was impressed she was able to multitask and stayed calm. She did great without prep in her immediate press conference, showing she learned her lesson from the conformation hearing and this time kept her cool. The episode certainly shows how tough politics and "the process" really can be.
I loved the split screen view of both Claire and Francis during the scene of them rounding up support. In the previous episode, as well as others, they mostly exhibited a business relationship and sleep in separate beds. Excellent use of jazz music as Frank sat on the floor exhausted and distraught after obviously having it out and destroyed his desk lamp. I am not sure that sex solves everything, but I was pleased that Claire tried to connect and encourage her husband. It appears Frank and Claire are trying to work together and may perhaps love each other. I would like to see their relationship grow deeper.
Once pushed to a corner, Frank was up to his old tricks of appearing to be working with people who don't support him. He continues to twists situations to his advantage and was certainly doing this with his American Works program. Jackie is becoming more dark and cunning as a character. I think she'll be the new "Zoe Barnes" this season, as she'll do anything to get what she wants. I am intrigued as to where to writers will go with her in future episodes.
The show ends with a pregnant pause and close up of Claire's frying pan eggs. Season three is certainly off to a sizzling good start. I love it!
Four Stars for Chapter 28!
What are your thoughts on the series or this episode?
Images from Netflix.com and Wikimedia. Some licensed under CC BY 2.0 via Wikimedia Commons.
Previous Post: House of Cards Season 3, Chapter 27 – Doug Stamper Next Post: Family Photos are Priceless in Custody
House of Cards Season 3 Chapter 35 – Innocence Lost
House of Cards Season 3 Chapter 34 – Hurricane Frank
The Main Reason to Vote Early in March Primaries…House of Cards Season 4
Five Tips for a Florida Fun In The Sun Vacation
Key Facts You Need to Know About Custody
Is a Flexible Schedule a Possible Parenting Solution?
Dad's Letter to His 16 Year Old Daughter
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 2,487 |
[
128000,
29707,
315,
27864,
14598,
220,
18,
11,
15957,
220,
1591,
1389,
9454,
9636,
6798,
198,
33877,
220,
1544,
11,
220,
679,
20,
555,
8797,
34,
592,
1094,
35,
329,
611,
28588,
89565,
198,
40,
617,
4529,
279,
1938,
1022,
311,
3821,
23469,
596,
4783,
315,
27864,
14598,
220,
18,
13,
57049,
5888,
5225,
11,
4587,
1541,
956,
1373,
904,
4726,
422,
499,
9167,
956,
1518,
279,
9417,
477,
15746,
279,
1501,
13,
1115,
1772,
374,
12514,
311,
856,
4783,
315,
27864,
51273,
11,
14598,
220,
18,
11,
15957,
220,
1591,
13,
23485,
323,
4430,
304,
279,
10430,
4999,
29707,
315,
27864,
14598,
220,
18,
11,
15957,
220,
1591,
22241,
1389,
9454,
9636,
6798,
3799,
3422,
2360,
1050,
13737,
1191,
198,
27229,
9636,
6798,
320,
6746,
1055,
11746
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
29707,
315,
27864,
14598,
220,
18,
11,
15957,
220,
1591,
1389,
9454,
9636,
6798,
198,
33877,
220,
1544,
11,
220,
679,
20,
555,
8797,
34,
592,
1094,
35,
329,
611,
28588,
89565,
198,
40,
617,
4529,
279,
1938,
1022,
311,
3821,
23469,
596,
4783,
315,
27864,
14598,
220,
18,
13,
57049,
5888,
5225,
11,
4587,
1541,
956,
1373,
904,
4726,
422,
499,
9167,
956,
1518,
279,
9417,
477,
15746,
279,
1501,
13,
1115,
1772,
374,
12514,
311,
856,
4783,
315,
27864,
51273,
11,
14598,
220,
18,
11,
15957,
220,
1591,
13,
23485,
323,
4430,
304,
279,
10430,
4999,
29707,
315,
27864,
14598,
220,
18,
11,
15957,
220,
1591,
22241,
1389,
9454,
9636,
6798,
3799,
3422,
2360,
1050,
13737,
1191,
198,
27229,
9636,
6798,
320,
6746,
1055,
11746,
-100
] |
Great Book covers by Jeff Brown at jeffbrowngraphics.com.
To concoct things that never existed and never will; that are beyond existence. A future world, a dystopia, inhabited planets, civilizations at the center of the earth or 20,000 leagues under the sea. We do not go to The Martian Chronicles to tell us what Mars is like, but for the power of Bradbury's imagination and his eloquence. Those series of stories exist quite comfortably alongside the science-based, factual The Martian. But at the deepest part of The Martian, it is about a struggle of a lone man to survive.
Same as 'The Weaponsmaker'. A rich young man thrown out into the hot and dangerous street without the slightest idea of what kind of people lived down there. Archetypes, deep plot lines scored in our heads.
I ran into this image through Bing, and I knew it was exactly the right one to use for Bude Rundel, but running down the source of the image was a bit of work. However, this got me in contact with Nuclear Snail studios in Germany and I have had a long and interesting e-mail correspondence with Dimitri Zaitsev, the designer. He designs post-apocalyptic/dystopian costuming for various film studios and CosPlay. He's very good. Thought I would give him a plug here on the blog.
In the 88.3 series it was a challenge to show the sudden climate change from a hot-dry world to catastrophic storms and flooding. This picture is interesting; what appears to be natural formations in the foreground are actually habitations. Kansas City is at the meeting-point of three major rivers; the Kaw, the Blue and the Missouri. The big Platte, draining an enormous expanse of prairie country in Nebraska, Kansas and even Wyoming, empties into the Missouri just north of Kansas City. In this over-urbanized world of the future, human beings in general and the controlling class in particular would have forgotten entirely about things like drainages and riverbeds. They are all covered over in slum apartments, shebangs, unauthorized shacks, shanty-towns and streets.
Even the maps of the rivers and their drainage patterns have been lost. Except to a few. Thus the devastation is unmeasureable.
He used to live among the elite, in the high places, where there were fountains and cooled air and the days went by smoothly, at ease, one after another.
Then the ban order came and he was thrown out, into the lower slum streets.
That's when the story begins.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 321 |
[
128000,
22111,
6017,
14861,
555,
12149,
10690,
520,
4864,
544,
65561,
32785,
916,
627,
1271,
76639,
302,
2574,
430,
2646,
25281,
323,
2646,
690,
26,
430,
527,
7953,
14209,
13,
362,
3938,
1917,
11,
264,
69110,
30651,
11,
70598,
33975,
11,
86569,
520,
279,
4219,
315,
279,
9578,
477,
220,
508,
11,
931,
39459,
1234,
279,
9581,
13,
1226,
656,
539,
733,
311,
578,
82128,
66416,
311,
3371,
603,
1148,
21725,
374,
1093,
11,
719,
369,
279,
2410,
315,
17478,
20176,
596,
28899,
323,
813,
64235,
4474,
13,
13266,
4101,
315,
7493,
3073,
5115,
45769,
16662,
279,
8198,
6108,
11,
61001,
578,
82128,
13,
2030,
520,
279,
51621,
961,
315,
578,
82128,
11,
433,
374,
922,
264,
14993,
315,
264,
47766,
893,
311,
18167,
627,
19749,
439,
364
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
22111,
6017,
14861,
555,
12149,
10690,
520,
4864,
544,
65561,
32785,
916,
627,
1271,
76639,
302,
2574,
430,
2646,
25281,
323,
2646,
690,
26,
430,
527,
7953,
14209,
13,
362,
3938,
1917,
11,
264,
69110,
30651,
11,
70598,
33975,
11,
86569,
520,
279,
4219,
315,
279,
9578,
477,
220,
508,
11,
931,
39459,
1234,
279,
9581,
13,
1226,
656,
539,
733,
311,
578,
82128,
66416,
311,
3371,
603,
1148,
21725,
374,
1093,
11,
719,
369,
279,
2410,
315,
17478,
20176,
596,
28899,
323,
813,
64235,
4474,
13,
13266,
4101,
315,
7493,
3073,
5115,
45769,
16662,
279,
8198,
6108,
11,
61001,
578,
82128,
13,
2030,
520,
279,
51621,
961,
315,
578,
82128,
11,
433,
374,
922,
264,
14993,
315,
264,
47766,
893,
311,
18167,
627,
19749,
439,
364,
-100
] |
Marina District Beauty! This South facing top floor corner unit is filled with light and faces Pantoja Park. Approx. 17 ft vaulted ceiling in living/dining & bedroom add to spacious feel. Completely remodeled kitchen has new Martha Stewart cabinets, white Kashmir granite and Viking appliances. Upgrades include new fireplace/mantle, refinished original wood floors & new plank white oak floors on second floor & bathroom upgraded tile. Large upstairs loft overlooks living room and can be bedroom or office.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 7,354 |
[
128000,
12331,
2259,
11182,
31685,
0,
1115,
4987,
13176,
1948,
6558,
9309,
5089,
374,
10409,
449,
3177,
323,
12580,
393,
13873,
5697,
5657,
13,
44919,
13,
220,
1114,
10702,
35684,
291,
22959,
304,
5496,
3529,
5859,
612,
14150,
923,
311,
33236,
2733,
13,
86214,
1323,
97579,
9979,
706,
502,
51970,
29868,
37328,
11,
4251,
50280,
38390,
323,
65450,
34802,
13,
3216,
23142,
2997,
502,
40511,
3262,
519,
273,
11,
2098,
25682,
4113,
7732,
27035,
612,
502,
73187,
4251,
38971,
27035,
389,
2132,
6558,
612,
15197,
33624,
10228,
13,
20902,
50007,
59527,
31805,
82,
5496,
3130,
323,
649,
387,
14150,
477,
5274,
13,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256,
128256
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0
] |
[
12331,
2259,
11182,
31685,
0,
1115,
4987,
13176,
1948,
6558,
9309,
5089,
374,
10409,
449,
3177,
323,
12580,
393,
13873,
5697,
5657,
13,
44919,
13,
220,
1114,
10702,
35684,
291,
22959,
304,
5496,
3529,
5859,
612,
14150,
923,
311,
33236,
2733,
13,
86214,
1323,
97579,
9979,
706,
502,
51970,
29868,
37328,
11,
4251,
50280,
38390,
323,
65450,
34802,
13,
3216,
23142,
2997,
502,
40511,
3262,
519,
273,
11,
2098,
25682,
4113,
7732,
27035,
612,
502,
73187,
4251,
38971,
27035,
389,
2132,
6558,
612,
15197,
33624,
10228,
13,
20902,
50007,
59527,
31805,
82,
5496,
3130,
323,
649,
387,
14150,
477,
5274,
13,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100,
-100
] |
\section{ALGORITHM}
\label{sec:algorithm}
\begin{algorithm}[t]
\caption{Hierarchical Thompson sampling.}
\label{alg:ts}
\begin{algorithmic}[1]
\State \textbf{Input:} Hyper-prior $Q$
\State Initialize $Q_1 \gets Q$
\For{$t = 1, 2, \dots$}
\State Observe tasks $\mathcal{S}_t \subseteq [m]$
\State Sample $\mu_t \sim Q_t$
\For{$s \in \mathcal{S}_t$}
\State Compute $P_{s, t}(\theta \mid \mu_t) \propto \mathcal{L}_{s, t}(\theta) P(\theta \mid \mu_t)$
\State Sample $\theta_{s, t} \sim P_{s, t}(\cdot \mid \mu_t)$
\State Take action $A_{s, t} \gets \argmax_{a \in \mathcal{A}} r(a; \theta_{s, t})$
\State Observe reward $Y_{s, t}$
\EndFor
\State Update $Q_{t + 1}$
\EndFor
\end{algorithmic}
\end{algorithm}
We take a Bayesian view and use hierarchical \emph{Thompson sampling (TS)}, which we call \ensuremath{\tt HierTS}\xspace, to solve our problem class. \ensuremath{\tt HierTS}\xspace samples task parameters from their posterior conditioned on history. Specifically, let $H_{s, t} = ((A_{s, \ell}, Y_{s, \ell}))_{\ell < t, \, s \in \mathcal{S}_\ell}$ denote the history of all interactions of \ensuremath{\tt HierTS}\xspace with task $s$ until round $t$, and $H_t = (H_{s, t})_{s \in [m]}$ be the concatenation of all histories up to round $t$. For each task $s \in \mathcal{S}_t$ in round $t$, \ensuremath{\tt HierTS}\xspace samples $\theta_{s, t} \sim \prob{\theta_{s, *} = \cdot \mid H_t}$ and then takes action $A_{s, t} = \argmax_{a \in \mathcal{A}} r(a; \theta_{s, t})$. The key difference from classical Thompson sampling is that the history $H_t$ includes observations of multiple tasks.
To sample $\theta_{s, t}$, we must address how the uncertainty over the unknown hyper-parameter $\mu_*$ and task parameters $\theta_{s, *}$ is modeled. The key idea is to maintain a \emph{hyper-posterior} $Q_t$ over $\mu_*$, given by
\begin{align*}
Q_t(\mu)
= \prob{\mu_* = \mu \mid H_t}\,,
\end{align*}
and then perform two-stage sampling. In particular, in round $t$, we first sample hyper-parameter $\mu_t \sim Q_t$. Next, for any task $s \in \mathcal{S}_t$, we sample the task parameter $\theta_{s, t} \sim P_{s, t}(\cdot \mid \mu_t)$, where
\begin{align*}
P_{s, t}(\theta \mid \mu)
= \prob{\theta_{s, *} = \theta \mid \mu_* = \mu, H_{s, t}}\,.
\end{align*}
In $P_{s, t}(\cdot \mid \mu)$, we only condition on the history of task $s$, since $\theta_{s, *}$ is independent of the other task histories given $\mu_* = \mu$ (\cref{fig:setting}). This process clearly samples from the true posterior, which is given by
\begin{align}
\condprob{\theta_{s, *} = \theta}{H_t}
& = \int_\mu \condprob{\theta_{s, *} = \theta, \mu_* = \mu}{H_t} \dif \mu
\label{eq:task posterior} \\
& = \int_\mu P_{s, t}(\theta \mid \mu)
Q_t(\mu) \dif \mu\,,
\nonumber
\end{align}
where $P_{s, t}(\theta \mid \mu) \propto \mathcal{L}_{s, t}(\theta) P(\theta \mid \mu)$ and
\begin{align*}
\textstyle
\mathcal{L}_{s, t}(\theta)
= \prod_{(a, y) \in H_{s, t}} P(y \mid a; \theta)
\end{align*}
denotes the likelihood of rewards in task $s$ given task parameter $\theta$.
The pseudo-code of \ensuremath{\tt HierTS}\xspace is shown in \cref{alg:ts}. Sampling in \eqref{eq:task posterior} can be implemented exactly in Gaussian graphical models (\cref{sec:models}). These models have interpretable closed-form posteriors, which permit the regret analysis of \ensuremath{\tt HierTS}\xspace. In practice, \ensuremath{\tt HierTS}\xspace can be implemented for any posterior distributions, but may require approximate inference \citep{doucet01sequential} to tractably sample from the posterior.
\section{Proof of \cref{lem:bayes regret}}
\label{sec:bayes regret proof}
The first claim is proved as follows. Fix round $t$ and task $s \in \mathcal{S}_t$. Since $\hat{\mu}_{s, t}$ is a deterministic function of $H_t$, and $A_{s, *}$ and $A_{s, t}$ are i.i.d.\ given $H_t$, we have
\begin{align*}
\E{}{A_{s, *}^\top \theta_{s, *} - A_{s, t}^\top \theta_{s, *}}
= \E{}{\condE{A_{s, *}^\top (\theta_{s, *} - \hat{\mu}_{s, t})}{H_t}} +
\E{}{\condE{A_{s, t}^\top (\hat{\mu}_{s, t} - \theta_{s, *})}{H_t}}\,.
\end{align*}
Moreover, $\theta_{s, *} - \hat{\mu}_{s, t}$ is a zero-mean random vector independent of $A_{s, t}$, and thus $\condE{A_{s, t}^\top (\hat{\mu}_{s, t} - \theta_{s, *})}{H_t} = 0$. So we only need to bound the first term above. Let
\begin{align*}
E_{s, t} =
\set{\normw{\theta_{s, *} - \hat{\mu}_{s, t}}{\hat{\Sigma}_{s, t}^{-1}}
\leq \sqrt{2 d \log(1 / \delta)}}
\end{align*}
be the event that a high-probability confidence interval for the task parameter $\theta_{s, *}$ holds. Fix history $H_t$. Then by the Cauchy-Schwarz inequality,
\begin{align*}
\condE{A_{s, *}^\top (\theta_{s, *} - \hat{\mu}_{s, t})}{H_t}
& \leq \condE{\normw{A_{s, *}}{\hat{\Sigma}_{s, t}}
\normw{\theta_{s, *} - \hat{\mu}_{s, t}}{\hat{\Sigma}_{s, t}^{-1}}}{H_t} \\
& \leq \sqrt{2 d \log(1 / \delta)} \, \condE{\normw{A_{s, *}}{\hat{\Sigma}_{s, t}}}{H_t} +
\underbrace{\max_{a \in \mathcal{A}} \normw{a}{\hat{\Sigma}_{s, t}}}_{\leq \sigma_{\max}}
\condE{\normw{\theta_{s, *} - \hat{\mu}_{s, t}}{\hat{\Sigma}_{s, t}^{-1}}
\I{\bar{E}_{s, t}}}{H_t} \\
& = \sqrt{2 d \log(1 / \delta)} \, \condE{\normw{A_{s, t}}{\hat{\Sigma}_{s, t}}}{H_t} +
\sigma_{\max} \, \condE{\normw{\theta_{s, *} - \hat{\mu}_{s, t}}{\hat{\Sigma}_{s, t}^{-1}}
\I{\bar{E}_{s, t}}}{H_t}\,.
\end{align*}
The equality follows from the fact that $\hat{\Sigma}_{s, t}$ is a deterministic function of $H_t$, and that $A_{s, *}$ and $A_{s, t}$ are i.i.d.\ given $H_t$. Now we focus on the second term above. First, note that
\begin{align*}
\normw{\theta_{s, *} - \hat{\mu}_{s, t}}{\hat{\Sigma}_{s, t}^{-1}}
= \normw{\hat{\Sigma}^{- \frac{1}{2}}_{s, t} (\theta_{s, *} - \hat{\mu}_{s, t})}{2}
\leq \sqrt{d} \maxnorm{\hat{\Sigma}^{- \frac{1}{2}}_{s, t} (\theta_{s, *} - \hat{\mu}_{s, t})}\,.
\end{align*}
By definition, $\theta_{s, *} - \hat{\mu}_{s, t} \mid H_t \sim \mathcal{N}(\mathbf{0}, \hat{\Sigma}_{s, t})$, and hence $\hat{\Sigma}^{- \frac{1}{2}}_{s, t} (\theta_{s, *} - \hat{\mu}_{s, t}) \mid H_t$ is a $d$-dimensional standard normal variable. Moreover, note that $\bar{E}_{s, t}$ implies $\maxnorm{\hat{\Sigma}^{- \frac{1}{2}}_{s, t} (\theta_{s, *} - \hat{\mu}_{s, t})} \geq \sqrt{2 \log(1 / \delta)}$. Finally, we combine these facts with a union bound over all entries of $\hat{\Sigma}^{- \frac{1}{2}}_{s, t} (\theta_{s, *} - \hat{\mu}_{s, t}) \mid H_t$, which are standard normal variables, and get
\begin{align*}
\condE{\maxnorm{\hat{\Sigma}^{- \frac{1}{2}}_{s, t} (\theta_{s, *} - \hat{\mu}_{s, t})}
\I{\bar{E}_{s, t}}}{H_t}
\leq 2 \sum_{i = 1}^d \frac{1}{\sqrt{2 \pi}} \int_{u = \sqrt{2 \log(1 / \delta)}}^\infty
u \exp\left[- \frac{u^2}{2}\right] \dif u
\leq \sqrt{\frac{2}{\pi}} d \delta\,.
\end{align*}
Now we combine all inequalities and have
\begin{align*}
\condE{A_{s, *}^\top (\theta_{s, *} - \hat{\mu}_{s, t})}{H_t}
\leq \sqrt{2 d \log(1 / \delta)} \, \condE{\normw{A_{s, t}}{\hat{\Sigma}_{s, t}}}{H_t} +
\sqrt{\frac{2}{\pi}} \sigma_{\max} d^\frac{3}{2} \delta\,.
\end{align*}
Since the above bound holds for any history $H_t$, we combine everything and get
\begin{align*}
\E{}{\sum_{t \geq 1} \sum_{s \in \mathcal{S}_t} A_{s, *}^\top \theta_{s, *} - A_{s, t}^\top \theta_{s, *}}
& \leq \sqrt{2 d \log(1 / \delta)} \,
\E{}{\sum_{t \geq 1} \sum_{s \in \mathcal{S}_t} \normw{A_{s, t}}{\hat{\Sigma}_{s, t}}} +
\sqrt{\frac{2}{\pi}} \sigma_{\max} d^\frac{3}{2} m n \delta \\
& \leq \sqrt{2 d m n \log(1 / \delta)}
\sqrt{\E{}{\sum_{t \geq 1} \sum_{s \in \mathcal{S}_t} \normw{A_{s, t}}{\hat{\Sigma}_{s, t}}^2}} +
\sqrt{\frac{2}{\pi}} \sigma_{\max} d^\frac{3}{2} m n \delta\,.
\end{align*}
The last step uses the Cauchy-Schwarz inequality and the concavity of the square root.
To bound $\sigma_{\max}$, we use Weyl's inequalities together with \eqref{eq:covariance decomposition}, the second claim in \cref{lem:covariance decomposition}, and \eqref{eq:linear hyperposterior}. Specifically, under the assumption that $\normw{a}{2} \leq 1$ for all $a \in \mathcal{A}$, we have
\begin{align*}
\max_{a \in \mathcal{A}} \normw{a}{\hat{\Sigma}_{s, t}}^2
& \leq \lambda_1(\hat{\Sigma}_{s, t})
\leq \lambda_1((\Sigma_0^{-1} + G_{s, t})^{-1}) +
\lambda_1((\Sigma_0^{-1} + G_{s, t})^{-1} \Sigma_0^{-1} \bar{\Sigma}_t
\Sigma_0^{-1} (\Sigma_0^{-1} + G_{s, t})^{-1}) \\
& \leq \lambda_1(\Sigma_0) +
\frac{\lambda_1^2(\Sigma_0) \lambda_1(\Sigma_q)}{\lambda_d^2(\Sigma_0)}
= \sigma_{\max}^2\,.
\end{align*}
This concludes the proof of the first claim.
The second claim is proved by modifying the first proof as follows. Fix round $t$ and task $s \in \mathcal{S}_t$. Let
\begin{align*}
E_{s, t} =
\set{\forall a \in \mathcal{A}: |a^\top (\theta_{s, *} - \hat{\mu}_{s, t})|
\leq \sqrt{2 \log(1 / \delta)} \normw{a}{\hat{\Sigma}_{s, t}}}
\end{align*}
be the event that all high-probability confidence intervals hold. Then we have
\begin{align*}
\condE{A_{s, *}^\top (\theta_{s, *} - \hat{\mu}_{s, t})}{H_t}
\leq \sqrt{2 \log(1 / \delta)} \, \condE{\normw{A_{s, t}}{\hat{\Sigma}_{s, t}}}{H_t} +
\condE{A_{s, *}^\top (\theta_{s, *} - \hat{\mu}_{s, t}) \I{\bar{E}_{s, t}}}{H_t}\,.
\end{align*}
Now note that for any action $a$, $a^\top (\theta_{s, *} - \hat{\mu}_{s, t}) / \normw{a}{\hat{\Sigma}_{s, t}}$ is a standard normal variable. It follows that
\begin{align*}
\condE{A_{s, *}^\top (\theta_{s, *} - \hat{\mu}_{s, t}) \I{\bar{E}_{s, t}}}{H_t}
\leq 2 \sum_{a \in \mathcal{A}} \normw{a}{\hat{\Sigma}_{s, t}} \frac{1}{\sqrt{2 \pi}}
\int_{u = \sqrt{2 \log(1 / \delta)}}^\infty
u \exp\left[- \frac{u^2}{2}\right] \dif u
\leq \sqrt{\frac{2}{\pi}} \sigma_{\max} K \delta\,.
\end{align*}
The rest of the proof proceeds as in the first claim, yielding
\begin{align*}
\condE{A_{s, *}^\top (\theta_{s, *} - \hat{\mu}_{s, t})}{H_t}
\leq \sqrt{2 \log(1 / \delta)} \, \condE{\normw{A_{s, t}}{\hat{\Sigma}_{s, t}}}{H_t} +
\sqrt{\frac{2}{\pi}} \sigma_{\max} K \delta\,.
\end{align*}
This completes the proof.
\section{Proof of \cref{thm:sequential regret}}
\label{sec:sequential proof}
\cref{lem:bayes regret} says that the Bayes regret $\mathcal{BR}(m, n)$ can be bounded by bounding the sum of posterior variances $\mathcal{V}(m, n)$. Since $|\mathcal{S}_t| = 1$, we make two simplifications. First, we replace the set of tasks $\mathcal{S}_t$ by a single task $S_t \in [m]$. Second, there are exactly $m n$ rounds.
Fix round $t$ and task $s = S_t$. To reduce clutter, let $M = \Sigma_0^{-1} + G_{s, t}$. By the total covariance decomposition in \cref{lem:covariance decomposition}, we have that
\begin{align}
\normw{A_{s, t}}{\hat{\Sigma}_{s, t}}^2
& = \sigma^2 \frac{A_{s, t}^\top \hat{\Sigma}_{s, t} A_{s, t}}{\sigma^2}
= \sigma^2 \left(\sigma^{-2} A_{s, t}^\top \tilde{\Sigma}_{s, t} A_{s, t} +
\sigma^{-2} A_{s, t}^\top M^{-1} \Sigma_0^{-1} \bar{\Sigma}_t
\Sigma_0^{-1} M^{-1} A_{s, t}\right)
\nonumber \\
& \leq c_1 \log(1 + \sigma^{-2} A_{s, t}^\top \tilde{\Sigma}_{s, t} A_{s, t}) +
c_2 \log(1 + \sigma^{-2} A_{s, t}^\top M^{-1} \Sigma_0^{-1} \bar{\Sigma}_t
\Sigma_0^{-1} M^{-1} A_{s, t})
\nonumber \\
& = c_1 \log\det(I_d + \sigma^{-2}
\tilde{\Sigma}_{s, t}^\frac{1}{2} A_{s, t} A_{s, t}^\top \tilde{\Sigma}_{s, t}^\frac{1}{2}) +
c_2 \log\det(I_d + \sigma^{-2}
\bar{\Sigma}^\frac{1}{2}_t \Sigma_0^{-1} M^{-1} A_{s, t} A_{s, t}^\top
M^{-1} \Sigma_0^{-1} \bar{\Sigma}^\frac{1}{2}_t)\,.
\label{eq:sequential proof decomposition}
\end{align}
The logarithmic terms are introduced using
\begin{align*}
x
= \frac{x}{\log(1 + x)} \log(1 + x)
\leq \left(\max_{x \in [0, u]} \frac{x}{\log(1 + x)}\right) \log(1 + x)
= \frac{u}{\log(1 + u)} \log(1 + x)\,,
\end{align*}
which holds for any $x \in [0, u]$. The resulting constants are
\begin{align*}
c_1
= \frac{\lambda_1(\Sigma_0)}{\log(1 + \sigma^{-2} \lambda_1(\Sigma_0))}\,, \quad
c_2
= \frac{c_q}{\log(1 + \sigma^{-2} c_q)}\,, \quad
c_q
= \frac{\lambda_1^2(\Sigma_0) \lambda_1(\Sigma_q)}{\lambda_d^2(\Sigma_0)}\,.
\end{align*}
The derivation of $c_1$ uses that
\begin{align*}
A_{s, t}^\top \tilde{\Sigma}_{s, t} A_{s, t}
= \lambda_1(\tilde{\Sigma}_{s, t})
= \lambda_d^{-1}(\Sigma_0^{-1} + G_{s, t})
\leq \lambda_d^{-1}(\Sigma_0^{-1})
= \lambda_1(\Sigma_0)\,.
\end{align*}
The derivation of $c_2$ follows from
\begin{align*}
A_{s, t}^\top M^{-1} \Sigma_0^{-1} \bar{\Sigma}_t \Sigma_0^{-1} M^{-1} A_{s, t}
\leq \lambda_1^2(M^{-1}) \lambda_1^2(\Sigma_0^{-1}) \lambda_1(\bar{\Sigma}_t)
\leq \frac{\lambda_1^2(\Sigma_0) \lambda_1(\Sigma_q)}{\lambda_d^2(\Sigma_0)}\,.
\end{align*}
This is also proved as the second claim in \cref{lem:covariance decomposition}. Now we focus on bounding the logarithmic terms in \eqref{eq:sequential proof decomposition}.
\subsection{First Term in \eqref{eq:sequential proof decomposition}}
\label{sec:sequential proof 1}
This is a per-instance term and can be rewritten as
\begin{align*}
\log\det(I_d + \sigma^{-2}
\tilde{\Sigma}_{s, t}^\frac{1}{2} A_{s, t} A_{s, t}^\top \tilde{\Sigma}_{s, t}^\frac{1}{2})
= \log\det(\tilde{\Sigma}_{s, t}^{-1} + \sigma^{-2} A_{s, t} A_{s, t}^\top) - \log\det(\tilde{\Sigma}_{s, t}^{-1})\,.
\end{align*}
When we sum over all rounds with task $s$, we get telescoping and the contribution of this term is at most
\begin{align*}
\sum_{t = 1}^{m n} \I{S_t = s}
\log\det(I_d + \sigma^{-2} \tilde{\Sigma}_{s, t}^\frac{1}{2} A_{s, t}
A_{s, t}^\top \tilde{\Sigma}_{s, t}^\frac{1}{2})
& = \log\det(\tilde{\Sigma}_{s, m n + 1}^{-1}) - \log\det(\tilde{\Sigma}_{s, 1}^{-1})
= \log\det(\Sigma_0^\frac{1}{2} \tilde{\Sigma}_{s, m n + 1}^{-1} \Sigma_0^\frac{1}{2}) \\
& \leq d \log\left(\frac{1}{d} \trace(\Sigma_0^\frac{1}{2} \tilde{\Sigma}_{s, m n + 1}^{-1}
\Sigma_0^\frac{1}{2})\right)
\leq d \log\left(1 + \frac{\lambda_1(\Sigma_0) n}{\sigma^2 d}\right)\,,
\end{align*}
where we use that task $s$ appears at most $n$ times. Now we sum over all $m$ tasks and get
\begin{align*}
\sum_{t = 1}^{m n}
\log\det(I_d + \sigma^{-2} \tilde{\Sigma}_{S_t, t}^\frac{1}{2} A_{S_t, t}
A_{S_t, t}^\top \tilde{\Sigma}_{S_t, t}^\frac{1}{2})
\leq d m \log\left(1 + \frac{\lambda_1(\Sigma_0) n}{\sigma^2 d}\right)\,.
\end{align*}
\subsection{Second Term in \eqref{eq:sequential proof decomposition}}
\label{sec:sequential proof 2}
This is a hyper-parameter term. Before we analyze it, let $v = \sigma^{-1} M^{- \frac{1}{2}} A_{s, t}$ and note that
\begin{align}
\bar{\Sigma}_{t + 1}^{-1} - \bar{\Sigma}_t^{-1}
& = (\Sigma_0 + (G_{s, t} + \sigma^{-2} A_{s, t} A_{s, t}^\top)^{-1})^{-1} -
(\Sigma_0 + G_{s, t}^{-1})^{-1}
\nonumber \\
& = \Sigma_0^{-1} - \Sigma_0^{-1} (M + \sigma^{-2} A_{s, t} A_{s, t}^\top)^{-1} \Sigma_0^{-1} -
(\Sigma_0^{-1} - \Sigma_0^{-1} M^{-1} \Sigma_0^{-1})
\nonumber \\
& = \Sigma_0^{-1} (M^{-1} - (M + \sigma^{-2} A_{s, t} A_{s, t}^\top)^{-1}) \Sigma_0^{-1}
\nonumber \\
& = \Sigma_0^{-1} M^{- \frac{1}{2}}
(I_d - (I_d + \sigma^{-2} M^{- \frac{1}{2}} A_{s, t} A_{s, t}^\top M^{- \frac{1}{2}})^{-1})
M^{- \frac{1}{2}} \Sigma_0^{-1}
\nonumber \\
& = \Sigma_0^{-1} M^{- \frac{1}{2}}
(I_d - (I_d + v v^\top)^{-1})
M^{- \frac{1}{2}} \Sigma_0^{-1}
\nonumber \\
& = \Sigma_0^{-1} M^{- \frac{1}{2}}
\frac{v v^\top}{1 + v^\top v}
M^{- \frac{1}{2}} \Sigma_0^{-1}
\nonumber \\
& = \sigma^{-2} \Sigma_0^{-1} M^{-1}
\frac{A_{s, t} A_{s, t}^\top}{1 + v^\top v}
M^{-1} \Sigma_0^{-1}\,,
\label{eq:linear telescoping}
\end{align}
where we first use the Woodbury matrix identity and then the Sherman-Morrison formula. Since $\normw{A_{s, t}}{2} \leq 1$,
\begin{align*}
1 + v^\top v
= 1 + \sigma^{-2} A_{s, t}^\top M^{-1} A_{s, t}
\leq 1 + \sigma^{-2} \lambda_1(\Sigma_0) = c\,.
\end{align*}
Based on the above derivations, we bound the second logarithmic term in \eqref{eq:sequential proof decomposition} as
\begin{align*}
& \log\det(I_d +
\sigma^{-2} \bar{\Sigma}_t^\frac{1}{2} \Sigma_0^{-1} M^{-1} A_{s, t} A_{s, t}^\top
M^{-1} \Sigma_0^{-1} \bar{\Sigma}_t^\frac{1}{2}) \\
& \quad \leq c \log\det(I_d +
\sigma^{-2} \bar{\Sigma}^\frac{1}{2}_t \Sigma_0^{-1} M^{-1} A_{s, t} A_{s, t}^\top
M^{-1} \Sigma_0^{-1} \bar{\Sigma}^\frac{1}{2}_t / c) \\
& \quad = c \left[\log\det(\bar{\Sigma}_t^{-1} +
\sigma^{-2} \Sigma_0^{-1} M^{-1} A_{s, t} A_{s, t}^\top M^{-1} \Sigma_0^{-1} / c) -
\log\det(\bar{\Sigma}_t^{-1})\right] \\
& \quad \leq c \left[\log\det(\bar{\Sigma}_{t + 1}^{-1}) -
\log\det(\bar{\Sigma}_t^{-1})\right]\,.
\end{align*}
The first inequality holds because $\log(1 + x) \leq c \log(1 + x / c)$ for any $x \geq 0$ and $c \geq 1$. The second inequality follows from the fact that we have a rank-$1$ update of $\bar{\Sigma}_t^{-1}$. Now we sum over all rounds and get telescoping
\begin{align*}
& \sum_{t = 1}^{m n}
\log\det(I_d + \sigma^{-2} \bar{\Sigma}_t^\frac{1}{2} \Sigma_0^{-1}
(\Sigma_0^{-1} + G_{S_t, t})^{-1} A_{S_t, t} A_{S_t, t}^\top (\Sigma_0^{-1} + G_{S_t, t})^{-1}
\Sigma_0^{-1} \bar{\Sigma}_t^\frac{1}{2}) \\
& \quad \leq c \left[\log\det(\bar{\Sigma}_{m n + 1}^{-1}) -
\log\det(\bar{\Sigma}_1^{-1})\right]
= c \log\det(\Sigma_q^\frac{1}{2} \bar{\Sigma}_{m n + 1}^{-1} \Sigma_q^\frac{1}{2})
\leq c d \log\left(\frac{1}{d} \trace(\Sigma_q^\frac{1}{2} \bar{\Sigma}_{m n + 1}^{-1}
\Sigma_q^\frac{1}{2})\right) \\
& \quad \leq c d
\log(\lambda_1(\Sigma_q^\frac{1}{2} \bar{\Sigma}_{m n + 1}^{-1} \Sigma_q^\frac{1}{2}))
\leq c d \log\left(1 + \frac{\lambda_1(\Sigma_q) m}{\lambda_d(\Sigma_0)}\right)\,.
\end{align*}
Finally, we combine the upper bounds for both logarithmic terms and get
\begin{align*}
\mathcal{V}(m, n)
= \E{}{\sum_{t = 1}^{m n} \normw{A_{S_t, t}}{\hat{\Sigma}_{S_t, t}}^2}
\leq d \left[c_1 m \log\left(1 + \frac{\lambda_1(\Sigma_0) n}{\sigma^2 d}\right) +
c_2 c \log\left(1 + \frac{\lambda_1(\Sigma_q) m}{\lambda_d(\Sigma_0)}\right)\right]\,,
\end{align*}
which yields the desired result after we substitute this bound into \cref{lem:bayes regret}. To simplify presentation in the main paper, $c_1$ and $c_2$ in \cref{thm:sequential regret} include the above logarithmic terms that multiply them.
\section{Proof of \cref{thm:concurrent regret}}
\label{sec:concurrent proof}
From \cref{ass:basis}, there exists a basis of $d$ actions such that if all actions in the basis are taken in task $s$ by round $t$, it is guaranteed that $\lambda_d(G_{s, t}) \geq \eta / \sigma^2$. We modify \ensuremath{\tt HierTS}\xspace to takes these actions first in any task $s$. Let $\mathcal{C}_t = \{s \in \mathcal{S}_t: \lambda_d(G_{s, t}) \geq \eta / \sigma^2\}$ be the set of \emph{sufficiently-explored tasks} by round $t$.
Using $\mathcal{C}_t$, we decompose the Bayes regret as
\begin{align*}
\mathcal{BR}(m, n)
\leq
\E{}{\sum_{t \geq 1} \sum_{s \in \mathcal{S}_t} \I{s \in \mathcal{C}_t}
(A_{s, *}^\top \theta_{s, *} - A_{s, t}^\top \theta_{s, *})} +
\E{}{\sum_{t \geq 1} \sum_{s \in \mathcal{S}_t} \I{s \not\in \mathcal{C}_t}
(A_{s, *}^\top \theta_{s, *} - A_{s, t}^\top \theta_{s, *})}\,.
\end{align*}
For any task $s$ and round $t$, we can trivially bound
\begin{align*}
\E{}{(A_{s, *} - A_{s, t})^\top \theta_{s, *}}
\leq \E{}{\normw{A_{s, *} - A_{s, t}}{\hat{\Sigma}_{s, 1}}
\normw{\theta_{s, *}}{\hat{\Sigma}_{s, 1}^{-1}}}
\leq 2 \sigma_{\max} \left(\normw{\mu_q}{\hat{\Sigma}_{s, 1}^{-1}} +
\E{}{\normw{\theta_{s, *} - \mu_q}{\hat{\Sigma}_{s, 1}^{-1}}}\right)\,,
\end{align*}
where $\sigma_{\max} = \sqrt{\lambda_1(\Sigma_q + \Sigma_0)}$ as in \cref{sec:sequential proof}. Here we use that $\normw{A_{s, *} - A_{s, t}}{2} \leq 2$ and that the prior covariance of $\theta_{s, *}$ is $\hat{\Sigma}_{s, 1} = \Sigma_q + \Sigma_0$. We know from \eqref{eq:gaussian hierarchical} that $\theta_{s, *} - \mu_q \sim \mathcal{N}(\mathbf{0}, \Sigma_q + \Sigma_0)$. This means that $\hat{\Sigma}^{- \frac{1}{2}}_{s, 1} (\theta_{s, *} - \mu_q)$ is a vector of $d$ independent standard normal variables. It follows that
\begin{align*}
\E{}{\normw{\theta_{s, *} - \mu_q}{\hat{\Sigma}_{s, 1}^{-1}}}
= \E{}{\normw{\hat{\Sigma}^{- \frac{1}{2}}_{s, 1} (\theta_{s, *} - \mu_q)}{2}}
\leq \sqrt{\E{}{\normw{\hat{\Sigma}^{- \frac{1}{2}}_{s, 1}
(\theta_{s, *} - \mu_q)}{2}^2}}
= \sqrt{d}\,.
\end{align*}
Since $s \not\in \mathcal{C}_t$ occurs at most $d$ times for any task $s$, the total regret due to forced exploration is bounded as
\begin{align*}
\E{}{\sum_{t \geq 1} \sum_{s \in \mathcal{S}_t} \I{s \not\in \mathcal{C}_t}
(A_{s, *}^\top \theta_{s, *} - A_{s, t}^\top \theta_{s, *})}
\leq 2 \sigma_{\max} \left(\normw{\mu_q}{\hat{\Sigma}_{s, 1}^{-1}} + \sqrt{d}\right) d m
= c_3\,.
\end{align*}
It remains to bound the first term in $\mathcal{BR}(m, n)$. On event $s \in \mathcal{C}_t$, \ensuremath{\tt HierTS}\xspace samples from the posterior and behaves exactly as \cref{alg:ts}. Therefore, we only need to bound $\mathcal{V}(m, n) = \E{}{\sum_{t \geq 1} \sum_{s \in \mathcal{S}_t} \I{s \in \mathcal{C}_t} \normw{A_{s, t}}{\hat{\Sigma}_{s, t}}^2}$ and then substitute the bound into \cref{lem:bayes regret}. By the total covariance decomposition in \cref{lem:covariance decomposition}, we have
\begin{align}
\normw{A_{s, t}}{\hat{\Sigma}_{s, t}}^2
= A_{s, t}^\top \tilde{\Sigma}_{s, t} A_{s, t} +
A_{s, t}^\top M^{-1} \Sigma_0^{-1} \bar{\Sigma}_t \Sigma_0^{-1} M^{-1} A_{s, t}\,,
\label{eq:concurrent proof decomposition}
\end{align}
where $M = \Sigma_0^{-1} + G_{s, t}$ to reduce clutter. As in \cref{sec:sequential proof}, we bound the contribution of each term separately.
\subsection{First Term in \eqref{eq:concurrent proof decomposition}}
This term depends only on $\tilde{\Sigma}_{s, t}$, which does not depend on interactions with other tasks than task $s$. Therefore, the bound is the same as in the sequential case in \cref{sec:sequential proof 1},
\begin{align*}
\sum_{t \geq 1} \sum_{s \in \mathcal{S}_t} \I{s \in \mathcal{C}_t} A_{s, t}^\top \tilde{\Sigma}_{s, t} A_{s, t}
\leq c_1 d m \log\left(1 + \frac{\lambda_1(\Sigma_0) n}{\sigma^2 d}\right)\,,
\end{align*}
where $c_1$ is defined in \cref{sec:sequential proof}.
\subsection{Second Term in \eqref{eq:concurrent proof decomposition}}
The difference from the sequential setting is in how we bound the second term in \eqref{eq:concurrent proof decomposition}. Before we had $|\mathcal{S}_t| = 1$, while now we have $|\mathcal{S}_t| \leq L \leq m$ for some $L$. Since more than one task is acted upon per round, the telescoping identity in \eqref{eq:linear telescoping} no longer holds. To remedy this, we reduce the concurrent case to the sequential one. Specifically, suppose that task $s \in \mathcal{S}_t$ in round $t$ has access to the concurrent observations of prior tasks in round $t$, for some order of tasks $\mathcal{S}_t = \{S_{t, i}\}_{i = 1}^L$. As \cref{thm:sequential regret} holds for any order, we choose the order where sufficiently-explored tasks $s \in \mathcal{C}_t$ appear first.
Let $\mathcal{S}_{t, i} = \{S_{t, j}\}_{j = 1}^{i - 1}$ be the first $i - 1$ tasks in $\mathcal{S}_t$ according to our chosen order. For $s = S_{t, i}$, let
\begin{align*}
\bar{\Sigma}_{s, t}^{-1}
= \Sigma_q^{-1} + \sum_{z \in \mathcal{S}_{t, i}} (\Sigma_0 + G_{z, t + 1}^{-1})^{-1} +
\sum_{z \in [m] \setminus \mathcal{S}_{t, i}} (\Sigma_0 + G_{z, t}^{-1})^{-1}
\end{align*}
be the reciprocal of the hyper-posterior covariance updated with concurrent observations in tasks $\mathcal{S}_{t, i}$. Next we show that $\bar{\Sigma}_t$ and $\bar{\Sigma}_{s, t}$ are similar.
\begin{lemma}
\label{lem:sequential concurrent ratio} Fix round $t$ and $i \in [L]$. Let $s = S_{t, i}$ and $\lambda_d(G_{s, t}) \geq \eta / \sigma^2$. Then
\begin{align*}
\lambda_1(\bar{\Sigma}_{s, t}^{-1} \bar{\Sigma}_t)
\leq 1 + \frac{\sigma^{-2} \lambda_1(\Sigma_q) (\lambda_1(\Sigma_0) + \sigma^2 / \eta)}
{\lambda_1(\Sigma_q) + (\lambda_1(\Sigma_0) + \sigma^2 / \eta) / L}\,.
\end{align*}
\end{lemma}
\begin{proof}
Using standard eigenvalue inequalities, we have
\begin{align}
\lambda_1(\bar{\Sigma}_{s, t}^{-1} \bar{\Sigma}_t)
= \lambda_1((\bar{\Sigma}_t^{-1} + \bar{\Sigma}_{s, t}^{-1} - \bar{\Sigma}_t^{-1})
\bar{\Sigma}_t)
\leq 1 + \lambda_1((\bar{\Sigma}_{s, t}^{-1} - \bar{\Sigma}_t^{-1}) \bar{\Sigma}_t)
\leq 1 + \frac{\lambda_1(\bar{\Sigma}_{s, t}^{-1} - \bar{\Sigma}_t^{-1})}
{\lambda_d(\bar{\Sigma}_t^{-1})}\,.
\label{eq:ratio decomposition}
\end{align}
By Weyl's inequalities, and from the definition of $\bar{\Sigma}_t$, we have
\begin{align*}
\lambda_d(\bar{\Sigma}_t^{-1})
& \geq \lambda_d(\Sigma_q^{-1}) + \sum_{z \in [m]} \lambda_d((\Sigma_0 + G_{z, t}^{-1})^{-1})
= \lambda_d(\Sigma_q^{-1}) + \sum_{z \in [m]} \lambda_1^{-1}(\Sigma_0 + G_{z, t}^{-1}) \\
& \geq \lambda_d(\Sigma_q^{-1}) +
\sum_{z \in [m]} (\lambda_1(\Sigma_0) + \lambda_1(G_{z, t}^{-1}))^{-1}
\geq \lambda_d(\Sigma_q^{-1}) + (i - 1) (\lambda_1(\Sigma_0) + \sigma^2 / \eta)^{-1}\,.
\end{align*}
In the last inequality, we use that the previous $i - 1$ tasks $\mathcal{S}_{t, i}$ are sufficiently explored. Analogously to \eqref{eq:linear telescoping},
\begin{align*}
\bar{\Sigma}_{s, t}^{-1} - \bar{\Sigma}_t^{-1}
& = \sum_{z \in \mathcal{S}_{t, i}}
(\Sigma_0 + (G_{z, t} + \sigma^{-2} A_{z, t} A_{z, t}^\top)^{-1})^{-1} -
(\Sigma_0 + G_{z, t}^{-1})^{-1} \\
& = \sigma^{-2} \sum_{z \in \mathcal{S}_{t, i}} \Sigma_0^{-1} M_{z, t}^{-1}
\frac{A_{z, t} A_{z, t}^\top}{1 + \sigma^{-2} A_{z, t}^\top M_{z, t}^{-1} A_{z, t}}
M_{z, t}^{-1} \Sigma_0^{-1}\,,
\end{align*}
where $M_{z, t} = \Sigma_0^{-1} + G_{z, t}$ to reduce clutter. Moreover, since $\normw{A_{z, t}}{2} \leq 1$ and $\sigma^{-2} A_{z, t}^\top M_{z, t}^{-1} A_{z, t} \geq 0$, we have
\begin{align*}
\lambda_1(\bar{\Sigma}_{s, t}^{-1} - \bar{\Sigma}_t^{-1})
\leq (i - 1) \sigma^{-2}\,.
\end{align*}
Finally, we substitute our upper bounds to the right-hand side of \eqref{eq:ratio decomposition} and get
\begin{align*}
\frac{\lambda_1(\bar{\Sigma}_{s, t}^{-1} - \bar{\Sigma}_t^{-1})}
{\lambda_d(\bar{\Sigma}_t^{-1})} \leq
\frac{(i - 1) \sigma^{-2}}{\lambda_1^{-1}(\Sigma_q) +
(i - 1) (\lambda_1(\Sigma_0) + \sigma^2 / \eta)^{-1}}
\leq \frac{\sigma^{-2} \lambda_1(\Sigma_q) (\lambda_1(\Sigma_0) + \sigma^2 / \eta)}
{\lambda_1(\Sigma_q) + (\lambda_1(\Sigma_0) + \sigma^2 / \eta) / L}\,,
\end{align*}
where we use that the ratio is maximized when $i - 1 = L$. This completes the proof.
\end{proof}
Now we return to \eqref{eq:concurrent proof decomposition}. First, we have that
\begin{align*}
A_{s, t}^\top M^{-1} \Sigma_0^{-1} \bar{\Sigma}_t \Sigma_0^{-1} M^{-1} A_{s, t}
& = A_{s, t}^\top M^{-1} \Sigma_0^{-1} \bar{\Sigma}_{s, t}^\frac{1}{2}
\left(\bar{\Sigma}_{s, t}^{- \frac{1}{2}} \bar{\Sigma}_t^\frac{1}{2}
\bar{\Sigma}_t^\frac{1}{2} \bar{\Sigma}_{s, t}^{- \frac{1}{2}}\right)
\bar{\Sigma}_{s, t}^\frac{1}{2} \Sigma_0^{-1} M^{-1} A_{s, t} \\
& \leq \lambda_1(\bar{\Sigma}_{s, t}^{- \frac{1}{2}} \bar{\Sigma}_t^\frac{1}{2}
\bar{\Sigma}_t^\frac{1}{2} \bar{\Sigma}_{s, t}^{- \frac{1}{2}})
A_{s, t}^\top M^{-1} \Sigma_0^{-1} \bar{\Sigma}_{s, t} \Sigma_0^{-1} M^{-1} A_{s, t} \\
& \leq \lambda_1(\bar{\Sigma}_{s, t}^{-1} \bar{\Sigma}_t)
A_{s, t}^\top M^{-1} \Sigma_0^{-1} \bar{\Sigma}_{s, t} \Sigma_0^{-1} M^{-1} A_{s, t}\,,
\end{align*}
where we use that the above expression is a quadratic form. Next we apply \cref{lem:sequential concurrent ratio} and get
\begin{align*}
\lambda_1(\bar{\Sigma}_{s, t}^{-1} \bar{\Sigma}_t)
\leq 1 + \frac{\sigma^{-2} \lambda_1(\Sigma_q) (\lambda_1(\Sigma_0) + \sigma^2 / \eta)}
{\lambda_1(\Sigma_q) + (\lambda_1(\Sigma_0) + \sigma^2 / \eta) / L}
= c_4\,.
\end{align*}
After $\bar{\Sigma}_t$ is turned into $\bar{\Sigma}_{s, t}$, we follow \cref{sec:sequential proof 2} and get that the hyper-parameter regret is
\begin{align*}
c_2 c_4 c d \log\left(1 + \frac{\lambda_1(\Sigma_q) m}{\lambda_d(\Sigma_0)}\right)\,,
\end{align*}
where the only difference is the extra factor of $c_4$. Finally, we combine all upper bounds and get
\begin{align*}
\mathcal{V}(m, n)
= \E{}{\sum_{t \geq 1} \sum_{s \in \mathcal{S}_t} \I{s \in \mathcal{C}_t}
\normw{A_{s, t}}{\hat{\Sigma}_{s, t}}^2}
\leq d \left[c_1 m \log\left(1 + \frac{\lambda_1(\Sigma_0) n}{\sigma^2 d}\right) +
c_2 c_4 c \log\left(1 + \frac{\lambda_1(\Sigma_q) m}{\lambda_d(\Sigma_0)}\right)\right]\,,
\end{align*}
which yields the desired result after we substitute it into \cref{lem:bayes regret}. To simplify presentation in the main paper, $c_1$ and $c_2$ in \cref{thm:concurrent regret} include the above logarithmic terms that multiply them.
\section{Gaussian Bandit Regret Bounds}
\label{sec:mab bounds}
Our regret bounds in \cref{sec:regret bounds} can be specialized to $K$-armed Gaussian bandits (\cref{sec:gaussian bandit}). Specifically, when the action set $\mathcal{A} = \set{e_i}_{i \in [K]}$ is the standard Euclidean basis in $\mathbb{R}^K$, \cref{thm:sequential regret,thm:concurrent regret} can be restated as follows.
\begin{theorem}[Sequential Gaussian bandit regret]
\label{thm:sequential mab regret} Let $|\mathcal{S}_t| = 1$ for all rounds $t$. Let $\delta = 1 / (m n)$. Then the Bayes regret of \ensuremath{\tt HierTS}\xspace is
\begin{align*}
\mathcal{BR}(m, n)
\leq \sqrt{2 K m n [c_1 m + c_2] \log(m n)} + c_3\,,
\end{align*}
where $c_3 = O(K)$,
\begin{align*}
c_1
= \frac{\sigma_0^2}{\log(1 + \sigma^{-2} \sigma_0^2)}
\log\left(1 + \frac{\sigma_0^2 n}{\sigma^2 K}\right)\,, \quad
c_2
= \frac{\sigma_q^2 c}{\log(1 + \sigma^{-2} \sigma_q^2)}
\log\left(1 + \frac{\sigma_q^2 m}{\sigma_0^2}\right)\,, \quad
c
= 1 + \frac{\sigma_0^2}{\sigma^2}\,.
\end{align*}
\end{theorem}
The main difference from the proof of \cref{thm:sequential regret} is that we start with the finite-action bound in \cref{lem:bayes regret}. Other than that, we use the facts that $\lambda_1(\Sigma_0) = \lambda_d(\Sigma_0) = \sigma_0^2$ and $\lambda_1(\Sigma_q) = \sigma_q^2$.
\begin{theorem}[Concurrent Gaussian bandit regret]
\label{thm:concurrent mab regret} Let $|\mathcal{S}_t| \leq L \leq m$. Let $\delta = 1 / (m n)$. Then the Bayes regret of \ensuremath{\tt HierTS}\xspace is
\begin{align*}
\mathcal{BR}(m, n)
\leq \sqrt{2 K m n [c_1 m + c_2] \log(m n)} + c_3\,,
\end{align*}
where $c_1$ and $c$ are defined as in \cref{thm:sequential mab regret},
\begin{align*}
c_2
= \frac{\sigma_q^2 c_4 c}{\log(1 + \sigma^{-2} \sigma_q^2)}
\log\left(1 + \frac{\sigma_q^2 m}{\sigma_0^2}\right)\,, \quad
c_4
= 1 + \frac{\sigma^{-2} \sigma_q^2 (\sigma_0^2 + \sigma^2)}
{\sigma_q^2 + (\sigma_0^2 + \sigma^2) / L}\,,
\end{align*}
and $c_3 = O(K m)$.
\end{theorem}
When we specialize \cref{thm:concurrent regret}, we note that $\eta = 1$, since the action set $\mathcal{A}$ is the standard Euclidean basis.
\section{Image Classification Experiment}
\label{sec:classification experiments}
\begin{figure*}[t]
\centering
\begin{minipage}{0.45\textwidth}
\includegraphics[width=\linewidth]{figures/mnist_pos0.pdf}
\end{minipage}
\begin{minipage}{0.45\textwidth}
\includegraphics[width=\linewidth]{figures/mnist_pos1.pdf}
\end{minipage}
\begin{minipage}{0.45\textwidth}
\includegraphics[width=\linewidth]{figures/mnist_mu_bar_pos0.pdf}
\end{minipage}
\begin{minipage}{0.45\textwidth}
\includegraphics[width=\linewidth]{figures/mnist_mu_bar_pos1.pdf}
\end{minipage}
\caption{Evaluation of \ensuremath{\tt HierTS}\xspace on multi-task digit classification using MNIST with different positive image classes. On the top, we plot the cumulative Bayes regret at each round. On the bottom, we visualize the most-rewarding image according to the learned hyper-parameter at evenly-spaced intervals.}
\label{fig:mnist}
\end{figure*}
We conduct an additional experiment that considers online classification using a real-world image dataset. The problem is cast as a multi-task linear bandit with Bernoulli rewards. Specifically, we construct a set of tasks where one image class is selected randomly to have high reward. In each task, at every round, $K$ images are uniformly sampled at random as actions, and the aim of the learning agent is to select an image from the unknown positive image class. The reward of an image from the positive class is $\mathsf{Ber}(0.9)$ and for all other classes is $\mathsf{Ber}(0.1)$.
We use the MNIST dataset \citep{mnist}, which consists of $60, 000$ images of handwritten digits, which we split equally into a training and test set. We down-sample each image to $d = 49$ pixels, which become the feature vector for the corresponding action in the bandit problem.
For each digit, the training set is used to estimate $\mu_*, \Sigma_0$, where all three algorithms use $\Sigma_0$ but only \ensuremath{\tt OracleTS}\xspace can use $\mu_*$. The algorithms are evaluated on the test set.
Given a positive digit class, we construct a different task $s$ by sub-sampling from the test set, and computing $\theta_{s, *}$ using positive images from the sub-sampled data.
For each digit as the positive image class, we evaluate our three algorithms on a multi-task linear bandit with $m = 10$ tasks, $n = 400$ interactions per task, and $K = 30$ actions, uniformly sampled from the test images. We chose $L = 5$ tasks per round, leading to $800$ rounds total. We assume a hyper-prior of $Q = \mathcal{N}(\mathbf{0}, I_d)$ with reward noise $\sigma = 0.5$ because the rewards are Bernoulli.
\cref{fig:mnist} shows the performance of all algorithms for two digits across $20$ independent runs. We see that \ensuremath{\tt HierTS}\xspace performs very well compared to standard \ensuremath{\tt TS}\xspace. In addition to regret, we also visualized the learned hyper-parameter $\bar{\mu}_t$ every $80$ rounds. We see that \ensuremath{\tt HierTS}\xspace very quickly learns the correct hyper-parameter, showing that it effectively leverages the shared structure across task. Overall, this experiment shows that even if \ensuremath{\tt HierTS}\xspace assumes a misspecified model of the environment, with non-Gaussian rewards and not knowing the true hyper-prior $Q$ and covariance $\Sigma_0$, \ensuremath{\tt HierTS}\xspace still performs very well.
\section{CONCLUSIONS}
\label{sec:conclusions}
We study \emph{hierarchical Bayesian bandits}, a general setting for solving similar bandit tasks. Instances of our setting recover meta-, multi-task, and federated bandits in prior works. We propose a natural hierarchical Thompson sampling algorithm, which can be implemented exactly and analyzed in Gaussian models. We analyze it using a novel total variance decomposition, which leads to interpretable regret bounds that scale with the hyper-prior and task prior widths. The benefit of hierarchical models is shown in both synthetic and real-world domains.
While we view our work as solving an extremely general problem, there are multiple directions for future work. For instance, we only study a specific hierarchical Gaussian structure in \cref{sec:models}. However, based on the discussion in \cref{sec:extensions}, we believe that our tools would apply to arbitrary graphical models with general sub-Gaussian distributions. Another direction for future work are frequentist upper bounds and matching lower bounds, in both the frequentist and Bayesian settings.
\section{EXPERIMENTS}
\label{sec:experiments}
\begin{figure*}[t!]
\centering
\begin{minipage}{0.32\textwidth}
\includegraphics[width=\linewidth]{figures/synthetic_d2_sigma_q0.500.pdf}
\end{minipage}
\begin{minipage}{0.32\textwidth}
\includegraphics[width=\linewidth]{figures/synthetic_d2_sigma_q1.000.pdf}
\end{minipage}
\begin{minipage}{0.32\textwidth}
\includegraphics[width=\linewidth]{figures/synthetic_tasks_per_round.pdf}
\end{minipage}
\vspace{-0.1in}
\caption{Evaluation of \ensuremath{\tt HierTS}\xspace on synthetic bandit problems. From left to right, we report the Bayes regret (a) for smaller $\sigma_q$, (b) for larger $\sigma_q$, (c) and as a function of the number of concurrent tasks $L$.}
\label{fig:synthetic}
\end{figure*}
We compare \ensuremath{\tt HierTS}\xspace to two TS baselines (\cref{sec:tightness}) that do not learn the hyper-parameter $\mu_*$. The first baseline is an idealized algorithm that knows $\mu_*$ and uses the true prior $\mathcal{N}(\mu_*, \Sigma_0)$. We call it \ensuremath{\tt OracleTS}\xspace. As \ensuremath{\tt OracleTS}\xspace has more information than \ensuremath{\tt HierTS}\xspace, we expect it to outperform \ensuremath{\tt HierTS}\xspace. The second baseline, which we call \ensuremath{\tt TS}\xspace, ignores that $\mu_*$ is shared among the tasks and uses the marginal prior of $\theta_{s, *}$, $\mathcal{N}(\mu_q, \Sigma_q + \Sigma_0)$, in each task.
We experiment with two linear bandit problems with $m = 10$ tasks: a synthetic problem with Gaussian rewards and an online image classification problem. The former is used to validate our regret bounds. The latter has non-Gaussian rewards and demonstrates that \ensuremath{\tt HierTS}\xspace is robust to prior misspecification. Our setup closely follows \citet{basu21noregrets}. However, our tasks can arrive in an arbitrary order and in parallel. Due to space constraints, we only report the synthetic experiment here, and defer the rest to \cref{sec:classification experiments}.
The synthethic problem is defined as follows: $d = 2$, $\abs{\mathcal{A}} = 10$, and each action is sampled uniformly from $[-0.5, 0.5]^d$. Initially, the number of concurrent tasks is $L = 5$; but we vary it later to measure its impact on regret. The number of rounds is $n = 200 m / L$ and $\mathcal{S}_t$ is defined as follows. First, we take a random permutation of the list of tasks where each task appears exactly $200$ times. Then we batch every $L$ consecutive elements of the list and set $\mathcal{S}_t$ to the $t$-th batch. The hyper-prior is $\mathcal{N}(\mathbf{0}, \Sigma_q)$ with $\Sigma_q = \sigma_q^2 I_d$, the task covariance is $\Sigma_0 = \sigma_0^2 I_d$, and the reward noise is $\sigma = 0.5$. We choose $\sigma_q \in \set{0.5, 1}$ and $\sigma_0 = 0.1$, where $\sigma_q \gg \sigma_0$ so that the effect of learning $\mu_*$ on faster learning of $\theta_{s, *}$ is easier to measure.
The regret of all compared algorithms is reported in \cref{fig:synthetic}. In plots (a) and (b), we show how the regret scales with the number of rounds for small ($\sigma_q = 0.5$) and large ($\sigma_q$ = 1) hyper-prior width. As suggested in \cref{sec:sequential regret}, \ensuremath{\tt HierTS}\xspace outperforms \ensuremath{\tt TS}\xspace that does not try to learn $\mu_*$. It is comparable to \ensuremath{\tt OracleTS}\xspace when $\sigma_q$ is small, but degrades as $\sigma_q$ increases. This matches the regret bound in \cref{thm:concurrent regret}, where $c_2$ grows with $\sigma_q$. In plot (c), we show how the regret of \ensuremath{\tt HierTS}\xspace varies with the number of concurrent tasks $L$. We observe that it increases with $L$, but the increase is sublinear, as suggested in \cref{sec:concurrent regret}.
\section{INTRODUCTION}
\label{sec:introduction}
A \emph{stochastic bandit} \citep{lai85asymptotically,auer02finitetime,lattimore19bandit} is an online learning problem where a \emph{learning agent} sequentially interacts with an environment over $n$ rounds. In each round, the agent takes an \emph{action} and receives a \emph{stochastic reward}. The agent aims to maximize its expected cumulative reward over $n$ rounds. It does not know the mean rewards of the actions \emph{a priori}, and must learn them by taking the actions. This induces the \emph{exploration-exploitation dilema}: \emph{explore}, and learn more about an action; or \emph{exploit}, and take the action with the highest estimated reward. In online advertising, an action could be showing an advertisement and its reward could be an indicator of a click.
\renewcommand{\thefootnote}{\fnsymbol{footnote}}
\footnotetext[1]{The work started while being at Google Research.}
\renewcommand{\thefootnote}{\arabic{footnote}}
More statistically-efficient exploration is the primary topic of bandit papers. This is attained by leveraging the structure of the problem, such as the form of the reward distribution \citep{garivier11klucb}, prior distribution over model parameters \citep{thompson33likelihood,agrawal12analysis,chapelle11empirical,russo18tutorial}, conditioning on known feature vectors \citep{dani08stochastic,abbasi-yadkori11improved,agrawal13thompson}, or modeling the process by which the total reward arises \citep{radlinski08learning,kveton15cascading,gai12combinatorial,chen16combinatorial,kveton15tight}. In this work, we solve multiple similar bandit tasks, and each task teaches the agent how to solve other tasks more efficiently.
We formulate the problem of learning to solve similar bandit tasks as regret minimization in a \emph{hierarchical Bayesian model} \citep{gelman13bayesian}. Each task is parameterized by a \emph{task parameter}, which is sampled i.i.d.\ from a distribution parameterized by a \emph{hyper-parameter}. The parameters are unknown and this relates all tasks, in the sense that each task teaches the agent about any other task. We derive Bayes regret bounds that reflect the structure of the problem and show that the price for learning the hyper-parameter is low. Our derivations use a novel \emph{total variance decomposition}, which decomposes the parameter uncertainty into per-task uncertainty conditioned on knowing the hyper-parameter and hyper-parameter uncertainty. After that, we individually bound each uncertainty source by elliptical lemmas \citep{dani08stochastic,abbasi-yadkori11improved}. Our approach can be exactly implemented and analyzed in hierarchical multi-armed and linear bandits with Gaussian rewards, but can be extended to other graphical model structures.
We build on numerous prior works that study a similar structure, under the names of collaborating filtering bandits \citep{gentile14online,kawale15efficient,li16collaborative}, bandit meta-learning and multi-task learning \citep{azar13sequential,deshmukh17multitask,bastani19meta,cella20metalearning,kveton21metathompson,moradipari21parameter}, and representation learning \citep{yang21impact}. Despite it, we make major novel contributions, both in terms of a more general setting and analysis techniques. Our setting relaxes the assumptions that the tasks are solved in a sequence and that exactly one task is solved per round. Moreover, while the design of our posterior sampling algorithm is standard, we make novel contributions in its analysis. In the sequential setting (\cref{sec:sequential regret}), we derive a Bayes regret bound by decomposing the posterior covariance, which is an alternative to prior derivations based on filtered mutual information \citep{russo16information,lu19informationtheoretic}. This technique is general, simple, and yields tighter regret bounds because it avoids marginal task parameter covariance; and so is of a broad interest. In the concurrent setting (\cref{sec:concurrent regret}), we bound the additional regret due to not updating the posterior after each interaction. This is non-trivial and a major departure from other bandit analyses. Our Bayes regret bound for this setting is the first of its kind.
The paper is organized as follows. In \cref{sec:setting}, we formalize our setting of \emph{hierarchical Bayesian bandits}. In \cref{sec:algorithm}, we introduce a natural Thompson sampling algorithm (\ensuremath{\tt HierTS}\xspace) for solving it. In \cref{sec:models}, we instantiate it in hierarchical Gaussian models. In \cref{sec:key ideas}, we review key ideas in our regret analyses, including a novel total covariance decomposition that allows us to analyze posteriors in hierarchical models. In \cref{sec:regret bounds}, we prove Bayes regret bounds for \ensuremath{\tt HierTS}\xspace in sequential and concurrent settings. Finally, in \cref{sec:experiments}, we evaluate \ensuremath{\tt HierTS}\xspace empirically to confirm our theoretical results.
\section{KEY IDEAS IN OUR ANALYSES}
\label{sec:key ideas}
This section reviews key ideas in our regret analyses, including a novel variance decomposition for the posterior of a hierarchical Gaussian model. Due to space constraints, we only discuss the linear bandit in \cref{sec:linear bandit}.
\subsection{Bayes Regret Bound}
Fix round $t$ and task $s \in \mathcal{S}_t$. Since \ensuremath{\tt HierTS}\xspace is a posterior sampling algorithm, both the posterior sample $\theta_{s, t}$ and the unknown task parameter $\theta_{s, *}$ are i.i.d.\ conditioned on $H_t$. Moreover, \eqref{eq:task posterior} is a marginalization and conditioning in a hierarchical Gaussian model given in \cref{fig:setting}. Therefore, although we never explicitly derive $\theta_{s, *} \mid H_t$, we know that it is a multivariate Gaussian distribution \citep{koller09probabilistic}; and we denote it by $\condprob{\theta_{s, *} = \theta}{H_t} = \mathcal{N}(\theta; \hat{\mu}_{s, t}, \hat{\Sigma}_{s, t})$.
Following existing Bayes regret analyses \citep{russo14learning}, we have that
\begin{align*}
& \E{}{A_{s, *}^\top \theta_{s, *} - A_{s, t}^\top \theta_{s, *} \mid H_t} = \\
& \E{}{A_{s, *}^\top (\theta_{s, *} - \hat{\mu}_{s, t}) \mid H_t} +
\E{}{A_{s, t}^\top (\hat{\mu}_{s, t}- \theta_{s, *}) \mid H_t}\,.
\end{align*}
Conditioned on history $H_t$, we observe that $\hat{\mu}_{s, t} - \theta_{s, *}$ is a zero-mean random vector and that $A_{s, t}$ is independent of it. Hence $\condE{A_{s, t}^\top (\hat{\mu}_{s, t} - \theta_{s, *})}{H_t} = 0$ and the Bayes regret is bounded as
\begin{align*}
\mathcal{BR}(m, n)
\leq \E{}{\sum_{t \geq 1}\sum_{s \in \mathcal{S}_t}
\E{}{A_{s, *}^\top (\theta_{s, *} - \hat{\mu}_{s, t}) \mid H_t}}\,.
\end{align*}
The following lemma provides an upper bound on the Bayes regret for $m$ tasks, with at most $n$ interactions with each, using the sum of posterior variances
\begin{align}
\mathcal{V}(m, n)
= \E{}{\sum_{t \geq 1} \sum_{s \in \mathcal{S}_t} \normw{A_{s, t}}{\hat{\Sigma}_{s, t}}^2}\,.
\label{eq:posterior variances}
\end{align}
The proof is deferred to \cref{sec:bayes regret proof}.
\begin{lemma}
\label{lem:bayes regret} For any $\delta > 0$, the Bayes regret $\mathcal{BR}(m, n)$ in a hierarchical linear bandit (\cref{sec:linear bandit}) is bounded by
\begin{align*}
\sqrt{2 d m n \mathcal{V}(m, n) \log(1 / \delta)} +
\sqrt{2 / \pi} \sigma_{\max} d^\frac{3}{2} m n \delta\,,
\end{align*}
where $\sigma_{\max}^2 = \lambda_1(\Sigma_0) + \lambda_1^2(\Sigma_0) \lambda_1(\Sigma_q) / \lambda_d^2(\Sigma_0)$. When the action space is finite, $|\mathcal{A}| = K$, we also get
\begin{align*}
\sqrt{2 m n \mathcal{V}(m, n) \log(1 / \delta)} +
\sqrt{2 / \pi} \sigma_{\max} K m n \delta\,.
\end{align*}
\end{lemma}
Therefore, to bound the regret, we only need to bound the posterior variances induced by the taken actions. The main challenge is that our posterior is over multiple variables. As can be seen in \eqref{eq:task posterior}, it comprises the hyper-posterior $Q_t$ over $\mu_*$ and the conditional $P_{s, t}$ over $\theta_{s, *}$. For any fixed $\mu_*$, $P_{s, t}(\cdot \mid \mu_*)$ should concentrate at $\theta_{s, *}$ as the agent gets more observations from task $s$. In addition, $Q_t$ should concentrate at $\mu_*$ as the agent learns more about $\mu_*$ from all tasks.
\subsection{Total Variance Decomposition}
\label{sec:total variance decomposition}
Due to the hierarchical structure of our problem, it is difficult to reason about the rate at which $\hat{\Sigma}_{s, t}$ \say{decreases}. In this work, we propose a novel variance decomposition that allows this. The decomposition uses the law of total variance \citep{weiss05probability}, which states that for any $X$ and $Y$,
\begin{align*}
\var{X}
= \E{}{\condvar{X}{Y}} + \var{\condE{X}{Y}}\,.
\end{align*}
If $X = \theta$ was a scalar task parameter and $Y = \mu$ was a scalar hyper-parameter, and we conditioned on $H$, the law would give
\begin{align*}
\condvar{\theta}{H}
= \condE{\condvar{\theta}{\mu, H}}{H} +
\condvar{\condE{\theta}{\mu, H}}{H}\,.
\end{align*}
This law extends to covariances \citep{weiss05probability}, where the conditional variance $\condvar{\cdot}{H}$ is substituted with the covariance $\condcov{\cdot}{H}$. We show the decomposition for a hierarchical Gaussian model below.
\subsection{Hierarchical Gaussian Models}
Recall that $\hat{\Sigma}_{s, t} = \condcov{\theta_{s, *}}{H_t}$. We derive a general formula for decomposing $\condcov{\theta_{s, *}}{H_t}$ below. To simplify notation, we consider a fixed task $s$ and round $t$, and drop subindexing by them.
\begin{lemma}
\label{lem:covariance decomposition} Let $\theta \mid \mu \sim \mathcal{N}(\mu, \Sigma_0)$ and $H = (x_t, Y_t)_{t = 1}^n$ be $n$ observations generated as $Y_t \mid \theta, x_t \sim \mathcal{N}(x_t^\top \theta, \sigma^2)$. Let $\condprob{\mu}{H} = \mathcal{N}(\mu; \bar{\mu}, \bar{\Sigma})$. Then
\begin{align*}
\condcov{\theta}{H}
= {} & (\Sigma_0^{-1} + G)^{-1} + {} \\
& (\Sigma_0^{-1} + G)^{-1} \Sigma_0^{-1} \bar{\Sigma}
\Sigma_0^{-1} (\Sigma_0^{-1} + G)^{-1}\,,
\end{align*}
where $G = \sigma^{-2} \sum_{t = 1}^n x_t x_t^\top$. Moreover, for any $x \in \mathbb{R}^d$,
\begin{align*}
& x^\top (\Sigma_0^{-1} + G)^{-1} \Sigma_0^{-1} \bar{\Sigma}
\Sigma_0^{-1} (\Sigma_0^{-1} + G)^{-1} x \\
& \quad \leq \frac{\lambda_1^2(\Sigma_0) \lambda_1(\bar{\Sigma})}{\lambda_d^2(\Sigma_0)}
\normw{x}{2}^2\,.
\end{align*}
\end{lemma}
\begin{proof}
By definition,
\begin{align*}
\condcov{\theta}{\mu, H}
& = (\Sigma_0^{-1} + G)^{-1}\,, \\
\condE{\theta}{\mu, H}
& = \condcov{\theta}{\mu, H}
(\Sigma_0^{-1} \mu + B)\,,
\end{align*}
where $B = \sigma^{-2}\sum_{t = 1}^n x_t Y_t$. Because $\condcov{\theta}{\mu, H}$ does not depend on $\mu$, $\condE{\condcov{\theta}{\mu, H}}{H} = \condcov{\theta}{\mu, H}$. In addition, since $B$ is a constant conditioned on $H$,
\begin{align*}
& \condcov{\condE{\theta}{\mu, H}}{H} \\
& \quad = \condcov{\condcov{\theta}{\mu, H} \Sigma_0^{-1} \mu}{H} \\
& \quad = (\Sigma_0^{-1} + G)^{-1} \Sigma_0^{-1} \bar{\Sigma}
\Sigma_0^{-1} (\Sigma_0^{-1} + G)^{-1}\,.
\end{align*}
This proves the first claim. The second claim follows from standard norm and eigenvalue inequalities.
\end{proof}
We use \cref{lem:covariance decomposition} as follows. For task $s$ and round $t$, the posterior covariance decomposes as
\begin{align}
\hat{\Sigma}_{s, t}
= {} & (\Sigma_0^{-1} + G_{s, t})^{-1} + {}
\label{eq:covariance decomposition} \\
& (\Sigma_0^{-1} + G_{s, t})^{-1} \Sigma_0^{-1} \bar{\Sigma}_t
\Sigma_0^{-1} (\Sigma_0^{-1} + G_{s, t})^{-1}\,.
\nonumber
\end{align}
The first term is $\condcov{\theta_{s, *}}{\mu_*, H_t}$ and captures uncertainty in $\theta_{s, *}$ conditioned on $\mu_*$. The second term depends on hyper-posterior covariance $\bar{\Sigma}_t$ and represents uncertainty in $\mu_*$. Since the first term is exactly $\tilde{\Sigma}_{s, t}$ in \eqref{eq:linear conditional}, while the second term is weighted by it, both are small when we get enough observations for task $s$. The above also says that $\normw{A_{s, t}}{\hat{\Sigma}_{s, t}}^2 = A_{s, t}^\top \hat{\Sigma}_{s, t} A_{s, t}$ in \eqref{eq:posterior variances} decompose into the two respective norms, which yields our regret decomposition.
\subsection{Extensions}
\label{sec:extensions}
So far, we only focused on hierarchical Gaussian models with known hyper-prior and task prior covariances. This is only because they have closed-form posteriors that are easy to interpret and manipulate, without resorting to approximations \citep{doucet01sequential}. This choice simplifies algebra and allows us to focus on the key hierarchical structure of our problem. We believe that the tools developed in this section can be applied more broadly. We discuss this next.
\cref{lem:bayes regret} decomposes the Bayes regret into posterior variances and upper bounds on the regret due to tail events. The posterior variance can be derived for any exponential-family posterior with a conjugate prior. On the other hand, the tail inequalities require sub-Gaussianity, which is a property of many exponential-family distributions.
\cref{lem:covariance decomposition} decomposes the posterior covariance in a hierarchical Gaussian model. It relies on the law of total covariance, which holds for any distribution, to obtain the task and hyper-parameter uncertainties. We expect that similar lemmas can be proved for other hierarchical models, so long as closed-form expressions for the respective uncertainties exist. Another notable property of our decomposition is that it does not require the marginal posterior of $\theta_{s, *}$. We view it as a strength. It means that our approach can be applied to complex graphical models where the marginal uncertainty may be hard to express, but the conditional and prior uncertainties are readily available.
One limitation of our analyses is that we bound the Bayes regret, instead of a stronger frequentist regret. This simplifies our proofs while they still capture our problem structure. Our analyses can be extended to the frequentist setting. This only requires a new proof of \cref{lem:bayes regret}, with martingale bounds for tail events and anti-concentration bounds for posterior sampling. The rest of the analysis, where our main contributions are, would not change.
\section{HIERARCHICAL GAUSSIAN BANDITS}
\label{sec:models}
Now we instantiate \ensuremath{\tt HierTS}\xspace in hierarchical Gaussian models. This yields closed-form posteriors, which permit regret analysis (\cref{sec:key ideas}). We discuss generalization to other distributions in \cref{sec:extensions}.
We assume that the environment is generated as
\begin{align}
\mu_*
& \sim \mathcal{N}(\mu_q, \Sigma_q)\,,
\label{eq:gaussian hierarchical} \\
\theta_{s, *} \mid \mu_*
& \sim \mathcal{N}(\mu_*, \Sigma_0)\,,
& \forall s \in [m] \,,
\nonumber \\
Y_{s, t} \mid A_{s, t}, \theta_{s, *}
& \sim \mathcal{N}(A_{s, t}^\top \theta_{s, *}, \sigma^2)\,,
& \forall t \geq 1, \, s \in \mathcal{S}_t\,,
\nonumber
\end{align}
where $\Sigma_q \in \mathbb{R}^{d \times d}$ and $\Sigma_0 \in \mathbb{R}^{d \times d}$ are covariance matrices; $\mu_q$, $\mu_*$, $\theta_{s, *}$ are $d$-dimensional vectors; the set of actions is $\mathcal{A} \subseteq \mathbb{R}^d$; and the mean reward of action $a \in \mathcal{A}$ is $r(a; \theta) = a^\top \theta$. The reward noise is $\mathcal{N}(0, \sigma^2)$. This formulation captures both the multi-armed and linear bandits, since the actions in the former can be viewed as vectors in a standard Euclidean basis. We assume that all of $\mu_q$, $\Sigma_q$, $\Sigma_0$, and $\sigma$ are known by the agent. This assumption is only needed in the analysis of \ensuremath{\tt HierTS}\xspace, where we require an analytically tractable posterior. We relax it in our experiments (\cref{sec:experiments}), where we learn these quantities from past data.
\subsection{Gaussian Bandit}
\label{sec:gaussian bandit}
We start with a $K$-armed Gaussian bandit, which we instantiate as \eqref{eq:gaussian hierarchical} as follows. The task parameter $\theta_{s, *}$ is a vector of mean rewards in task $s$, where $\theta_{s, *, i}$ is the mean reward of action $i$. The covariance matrices are diagonal, $\Sigma_q = \sigma_q^2 I_K$ and $\Sigma_0 = \sigma_0^2 I_K$. We assume that both $\sigma_q > 0$ and $\sigma_0 > 0$ are known. The reward distribution of action $i$ is $\mathcal{N}(\theta_{s, *, i}, \sigma^2)$, where $\sigma > 0$ is a known reward noise.
Because $\Sigma_q$ and $\Sigma_0$ are diagonal, the hyper-posterior in round $t$ factors across the actions. Specifically, it is $Q_t = \mathcal{N}(\bar{\mu}_t, \bar{\Sigma}_t)$, where $\bar{\Sigma}_t = \diag{(\bar{\sigma}_{t, i}^2)_{i \in [K]}}$ and
\begin{align}
\bar{\mu}_{t, i}
& = \bar{\sigma}_{t, i}^2 \left(\frac{\mu_{q, i}}{\sigma_q^2} +
\sum_{s \in [m]} \frac{N_{s, t, i}}{N_{s, t, i} \sigma_0^2 + \sigma^2}
\frac{B_{s, t, i}}{N_{s, t, i}}\right)\,,
\label{eq:mab hyperposterior} \\
\bar{\sigma}_{t, i}^{-2}
& = \sigma_q^{-2} + \sum_{s \in [m]} \frac{N_{s, t, i}}{N_{s, t, i} \sigma_0^2 + \sigma^2}\,.
\nonumber
\end{align}
Here $N_{s, t, i} = \sum_{\ell < t} \I{s \in \mathcal{S}_\ell, A_{s, \ell} = i}$ is the number of times that action $i$ is taken in task $s$ up to round $t$ and $B_{s, t, i} = \sum_{\ell < t} \I{s \in \mathcal{S}_\ell, A_{s, \ell} = i} Y_{s, \ell}$ is its total reward. The hyper-posterior is derived in Appendix D of \citet{kveton21metathompson}. To understand it, it is helpful to view it as a Gaussian posterior where each task is a single observation. The observation of task $s$ is the empirical mean reward estimate of action $i$ in task $s$, $B_{s, t, i} / N_{s, t, i}$, and its variance is $(N_{s, t, i} \sigma_0^2 + \sigma^2) / N_{s, t, i}$. The tasks with more observations affect the value of $\bar{\mu}_{t, i}$ more, because their mean reward estimates have lower variances. The variance never decreases below $\sigma_0^2$, because even the actual mean reward $\theta_{s, *, i}$ would be a noisy observation of $\mu_{*, i}$ with variance $\sigma_0^2$.
After the hyper-parameter is sampled, $\mu_t \sim Q_t$, the task parameter is sampled, $\theta_{s, t} \sim \mathcal{N}(\tilde{\mu}_{s, t}, \tilde{\Sigma}_{s, t})$, where $\tilde{\Sigma}_{s, t} = \diag{(\tilde{\sigma}_{s, t, i}^2)_{i \in [K]}}$ and
\begin{align}
\tilde{\mu}_{s, t, i}
& = \tilde{\sigma}_{s, t, i}^2 \left(\frac{\mu_t}{\sigma_0^2} +
\frac{B_{s, t, i}}{\sigma^2}\right)\,,
\label{eq:mab conditional} \\
\tilde{\sigma}_{s, t, i}^{-2}
& = \frac{1}{\sigma_0^2} + \frac{N_{s, t, i}}{\sigma^2}\,.
\nonumber
\end{align}
Note that the above is a Gaussian posterior with prior $\mathcal{N}(\mu_t, \sigma_0^2 I_K)$ and $N_{s, t, i}$ observations.
\subsection{Linear Bandit with Gaussian Rewards}
\label{sec:linear bandit}
Now we study a $d$-dimensional linear bandit, which is instantiated as \eqref{eq:gaussian hierarchical} as follows. The task parameter $\theta_{s, *}$ are coefficients in a linear model. The covariance matrices $\Sigma_q$ are $\Sigma_0$ are positive semi-definite and known. The reward distribution of action $a$ is $\mathcal{N}(a^\top \theta_{s, *}, \sigma^2)$, where $\sigma > 0$ is a known reward noise.
Similarly to \cref{sec:gaussian bandit}, we obtain closed-form posteriors using \citet{kveton21metathompson}. The hyper-posterior in round $t$ is $Q_t = \mathcal{N}(\bar{\mu}_t, \bar{\Sigma}_t)$, where
\begin{align}
\bar{\mu}_t
& = \bar{\Sigma}_t \Big(\Sigma_q^{-1} \mu_q +
\smashoperator{\sum_{s \in [m]}}
B_{s, t} - G_{s, t}(\Sigma_0^{-1} + G_{s, t})^{-1} B_{s, t}\Big)
\nonumber \\
& = \bar{\Sigma}_t \Big(\Sigma_q^{-1} \mu_q +
\sum_{s \in [m]} (\Sigma_0 + G_{s, t}^{-1})^{-1} G_{s, t}^{-1} B_{s, t}\Big)\,,
\nonumber \\
\bar{\Sigma}_t^{-1}
& = \Sigma_q^{-1} +
\sum_{s \in [m]} G_{s, t} - G_{s, t}(\Sigma_0^{-1} + G_{s, t})^{-1} G_{s, t}
\nonumber \\
& = \Sigma_q^{-1} +
\sum_{s \in [m]} (\Sigma_0 + G_{s, t}^{-1})^{-1}\,.
\label{eq:linear hyperposterior}
\end{align}
Here
\begin{align*}
G_{s, t}
= \sigma^{-2} \sum_{\ell < t} \I{s \in \mathcal{S}_\ell} A_{s, \ell} A_{s, \ell}^\top
\end{align*}
is the outer product of the features of taken actions in task $s$ up to round $t$ and
\begin{align*}
B_{s, t}
= \sigma^{-2} \sum_{\ell < t} \I{s \in \mathcal{S}_\ell} A_{s, \ell} Y_{s, \ell}
\end{align*}
is their sum weighted by the observed rewards. Similarly to \eqref{eq:mab hyperposterior}, it is helpful to view \eqref{eq:linear hyperposterior} as a multivariate Gaussian posterior where each task is a single observation. The observation of task $s$ is the least squares estimate of $\theta_{s, *}$ from task $s$, $G_{s, t}^{-1} B_{s, t}$, and its covariance is $\Sigma_0 + G_{s, t}^{-1}$. Again, the tasks with many observations affect the value of $\bar{\mu}_t$ more, because $G_{s, t}^{-1}$ approaches a zero matrix in these tasks. In this setting, the covariance approaches $\Sigma_0$, because even the unknown task parameter $\theta_{s, *}$ would be a noisy observation of $\mu_*$ with covariance $\Sigma_0$.
After the hyper-parameter is sampled, $\mu_t \sim Q_t$, the task parameter is sampled, $\theta_{s, t} \sim \mathcal{N}(\tilde{\mu}_{s, t}, \tilde{\Sigma}_{s, t})$, where
\begin{align}
\tilde{\mu}_{s, t}
& = \tilde{\Sigma}_{s, t} \left(\Sigma_0^{-1} \mu_t + B_{s, t}\right)\,,
\label{eq:linear conditional} \\
\tilde{\Sigma}_{s, t}^{-1}
& = \Sigma_0^{-1} + G_{s, t}\,.
\nonumber
\end{align}
The above is the posterior of a linear model with a Gaussian prior $\mathcal{N}(\mu_t, \Sigma_0)$ and Gaussian observations.
\subsubsection*{\bibname}}
\usepackage{algorithm}
\usepackage{algorithmicx}
\usepackage[noend]{algpseudocode}
\usepackage{amsmath}
\usepackage{amssymb}
\usepackage{amsthm}
\usepackage{bbm}
\usepackage{bm}
\usepackage{caption}
\usepackage{color}
\usepackage{dirtytalk}
\usepackage{dsfont}
\usepackage{enumerate}
\usepackage{graphicx}
\usepackage{listings}
\usepackage{mathtools}
\usepackage{subfigure}
\usepackage{xspace}
\usepackage[usenames,dvipsnames]{xcolor}
\usepackage[bookmarks=false]{hyperref}
\hypersetup{
pdffitwindow=true,
pdfstartview={FitH},
pdfnewwindow=true,
colorlinks,
linktocpage=true,
linkcolor=Green,
urlcolor=Green,
citecolor=Green
}
\usepackage[capitalize,noabbrev]{cleveref}
\usepackage[backgroundcolor=white]{todonotes}
\newcommand{\todob}[2][]{\todo[color=Red!20,size=\tiny,inline,#1]{B: #2}}
\newcommand{\todoj}[2][]{\todo[color=Blue!20,size=\tiny,inline,#1]{J: #2}}
\newcommand{\todomgh}[2][]{\todo[color=Cyan!20,size=\tiny,inline,#1]{MGH: #2}}
\newcommand{\mz}[2][]{\textcolor{red}{[MZ: #2]}}
\newcommand{\commentout}[1]{}
\newcommand{\junk}[1]{}
\usepackage{tikz}
\usetikzlibrary{bayesnet}
\Crefname{corollary}{Corollary}{Corollaries}
\Crefname{proposition}{Proposition}{Propositions}
\Crefname{theorem}{Theorem}{Theorems}
\Crefname{definition}{Definition}{Definitions}
\Crefname{assumption}{Assumption}{Assumptions}
\Crefname{example}{Example}{Examples}
\Crefname{remark}{Remark}{Remarks}
\Crefname{setting}{Setting}{Settings}
\Crefname{lemma}{Lemma}{Lemmas}
\usepackage{thmtools}
\declaretheorem[name=Theorem,refname={Theorem,Theorems},Refname={Theorem,Theorems}]{theorem}
\declaretheorem[name=Lemma,refname={Lemma,Lemmas},Refname={Lemma,Lemmas},sibling=theorem]{lemma}
\declaretheorem[name=Corollary,refname={Corollary,Corollaries},Refname={Corollary,Corollaries},sibling=theorem]{corollary}
\declaretheorem[name=Assumption,refname={Assumption,Assumptions},Refname={Assumption,Assumptions}]{assumption}
\declaretheorem[name=Proposition,refname={Proposition,Propositions},Refname={Proposition,Propositions},sibling=theorem]{proposition}
\declaretheorem[name=Definition,refname={Definition,Definitions},Refname={Definition,Definitions},sibling=theorem]{definition}
\declaretheorem[name=Example,refname={Example,Examples},Refname={Example,Examples}]{example}
\declaretheorem[name=Remark,refname={Remark,Remarks},Refname={Remark,Remarks}]{remark}
\newcommand{\mathcal{A}}{\mathcal{A}}
\newcommand{\mathcal{B}}{\mathcal{B}}
\newcommand{\mathcal{C}}{\mathcal{C}}
\newcommand{\mathcal{D}}{\mathcal{D}}
\newcommand{\mathcal{E}}{\mathcal{E}}
\newcommand{\mathcal{F}}{\mathcal{F}}
\newcommand{\mathcal{G}}{\mathcal{G}}
\newcommand{\mathcal{H}}{\mathcal{H}}
\newcommand{\mathcal{I}}{\mathcal{I}}
\newcommand{\mathcal{J}}{\mathcal{J}}
\newcommand{\mathcal{K}}{\mathcal{K}}
\newcommand{\mathcal{L}}{\mathcal{L}}
\newcommand{\mathcal{M}}{\mathcal{M}}
\newcommand{\mathcal{N}}{\mathcal{N}}
\newcommand{\mathcal{O}}{\mathcal{O}}
\newcommand{\mathcal{P}}{\mathcal{P}}
\newcommand{\mathcal{Q}}{\mathcal{Q}}
\newcommand{\mathcal{R}}{\mathcal{R}}
\newcommand{\mathcal{S}}{\mathcal{S}}
\newcommand{\mathcal{T}}{\mathcal{T}}
\newcommand{\mathcal{U}}{\mathcal{U}}
\newcommand{\mathcal{V}}{\mathcal{V}}
\newcommand{\mathcal{W}}{\mathcal{W}}
\newcommand{\mathcal{X}}{\mathcal{X}}
\newcommand{\mathcal{Y}}{\mathcal{Y}}
\newcommand{\mathcal{Z}}{\mathcal{Z}}
\newcommand{\varepsilon}{\varepsilon}
\newcommand{\mathbb{Z}}{\mathbb{Z}}
\newcommand{\mathbb{N}}{\mathbb{N}}
\newcommand{\mathbb{R}}{\mathbb{R}}
\newcommand{\diag}[1]{\mathrm{diag}\left(#1\right)}
\newcommand{\domain}[1]{\mathrm{dom}\left(#1\right)}
\newcommand{\mathsf{pa}}{\mathsf{pa}}
\newcommand{\range}[1]{\mathrm{rng}\left[#1\right]}
\newcommand{\E}[2]{\mathbb{E}_{#1} \left[#2\right]}
\newcommand{\condE}[2]{\mathbb{E} \left[#1 \,\middle|\, #2\right]}
\newcommand{\Et}[1]{\mathbb{E}_t \left[#1\right]}
\newcommand{\prob}[1]{\mathbb{P} \left(#1\right)}
\newcommand{\condprob}[2]{\mathbb{P} \left(#1 \,\middle|\, #2\right)}
\newcommand{\probt}[1]{\mathbb{P}_t \left(#1\right)}
\newcommand{\var}[1]{\mathrm{var} \left[#1\right]}
\newcommand{\condvar}[2]{\mathrm{var} \left[#1 \,\middle|\, #2\right]}
\newcommand{\std}[1]{\mathrm{std} \left[#1\right]}
\newcommand{\condstd}[2]{\mathrm{std} \left[#1 \,\middle|\, #2\right]}
\newcommand{\cov}[1]{\mathrm{cov} \left[#1\right]}
\newcommand{\condcov}[2]{\mathrm{cov} \left[#1 \,\middle|\, #2\right]}
\newcommand{\abs}[1]{\left|#1\right|}
\newcommand{\ceils}[1]{\left\lceil#1\right\rceil}
\newcommand{\dbar}[1]{\bar{\bar{#1}}}
\newcommand*\dif{\mathop{}\!\mathrm{d}}
\newcommand{\floors}[1]{\left\lfloor#1\right\rfloor}
\newcommand{\I}[1]{\mathds{1} \! \left\{#1\right\}}
\newcommand{\inner}[2]{\langle#1, #2\rangle}
\newcommand{\kl}[2]{D_\mathrm{KL}(#1 \,\|\, #2)}
\newcommand{\klplus}[2]{D_\mathrm{KL}^+(#1 \,\|\, #2)}
\newcommand{\maxnorm}[1]{\|#1\|_\infty}
\newcommand{\maxnormw}[2]{\|#1\|_{\infty, #2}}
\newcommand{\negpart}[1]{\left[#1\right]^-}
\newcommand{\norm}[1]{\|#1\|}
\newcommand{\normw}[2]{\|#1\|_{#2}}
\newcommand{\pospart}[1]{\left[#1\right]^+}
\newcommand{\rnd}[1]{\bm{#1}}
\newcommand{\set}[1]{\left\{#1\right\}}
\newcommand{\subreal}[0]{\preceq}
\newcommand{\supreal}[0]{\succeq}
\newcommand{^\top}{^\top}
\DeclareMathOperator*{\argmax}{arg\,max\,}
\DeclareMathOperator*{\argmin}{arg\,min\,}
\let\det\relax
\DeclareMathOperator{\det}{det}
\DeclareMathOperator{\poly}{poly}
\DeclareMathOperator{\rank}{rank}
\DeclareMathOperator{\sgn}{sgn}
\let\trace\relax
\DeclareMathOperator{\trace}{tr}
\mathchardef\mhyphen="2D
\newcommand{\ensuremath{\tt HierTS}\xspace}{\ensuremath{\tt HierTS}\xspace}
\newcommand{\ensuremath{\tt LinTS}\xspace}{\ensuremath{\tt LinTS}\xspace}
\newcommand{\ensuremath{\tt LinUCB}\xspace}{\ensuremath{\tt LinUCB}\xspace}
\newcommand{\ensuremath{\tt OracleTS}\xspace}{\ensuremath{\tt OracleTS}\xspace}
\newcommand{\ensuremath{\tt TS}\xspace}{\ensuremath{\tt TS}\xspace}
\newcommand{\ensuremath{\tt UCB}\xspace}{\ensuremath{\tt UCB}\xspace}
\def\mathcal{R}{\mathcal{R}}
\def\mathcal{BR}{\mathcal{BR}}
\begin{document}
\twocolumn[
\aistatstitle{Hierarchical Bayesian Bandits}
\aistatsauthor{Joey Hong \And Branislav Kveton \And Manzil Zaheer \And Mohammad Ghavamzadeh}
\aistatsaddress{UC Berkeley$^*$ \And Amazon$^*$ \And Google DeepMind \And Google Research}]
\begin{abstract}
Meta-, multi-task, and federated learning can be all viewed as solving similar tasks, drawn from a distribution that reflects task similarities. We provide a unified view of all these problems, as learning to act in a \emph{hierarchical Bayesian bandit}. We propose and analyze a natural hierarchical Thompson sampling algorithm (\ensuremath{\tt HierTS}\xspace) for this class of problems. Our regret bounds hold for many variants of the problems, including when the tasks are solved sequentially or in parallel; and show that the regret decreases with a more informative prior. Our proofs rely on a novel total variance decomposition that can be applied beyond our models. Our theory is complemented by experiments, which show that the hierarchy helps with knowledge sharing among the tasks. This confirms that hierarchical Bayesian bandits are a universal and statistically-efficient tool for learning to act with similar bandit tasks.
\end{abstract}
\input{introduction}
\input{setting}
\input{algorithm}
\input{models}
\input{key_ideas}
\input{regret}
\input{experiments}
\input{related_work}
\input{conclusions}
\bibliographystyle{abbrvnat}
\section{REGRET BOUNDS}
\label{sec:regret bounds}
This section bounds the Bayes regret of \ensuremath{\tt HierTS}\xspace in the linear bandit in \cref{sec:linear bandit}. Our bounds are specialized to multi-armed bandits in \cref{sec:mab bounds}. The key idea is to bound the posterior variances $\mathcal{V}(m, n)$ in \eqref{eq:posterior variances} and then substitute the bound into the infinite-action bound in \cref{lem:bayes regret}. We bound the variances using the total covariance decomposition in \cref{sec:total variance decomposition}. Without loss of generality, we assume that the action set $\mathcal{A}$ is a subset of a unit ball, that is $\max_{a \in \mathcal{A}} \normw{a}{2} \leq 1$ for any action $a \in \mathcal{A}$.
We make the following contributions in theory. First, we prove regret bounds using a novel variance decomposition (\cref{sec:total variance decomposition}), which improves in constant factors over classical information-theory bounds \citep{russo16information}. Second, we prove the first Bayes regret bound for the setting where an agent that interacts with multiple tasks simultaneously.
This section has two parts. In \cref{sec:sequential regret}, we assume that only one action is taken in any round $t$, $|\mathcal{S}_t| = 1$. We call this setting \emph{sequential}, and note that it is the primary setting studied by prior works \citep{kveton21metathompson,basu21noregrets}. In \cref{sec:concurrent regret}, we focus on a \emph{concurrent} setting, where a single action can be taken in up to $L$ tasks in any round $t$, $|\mathcal{S}_t| \leq L \leq m$. The challenge of this setting is that the task parameters are only updated after all actions are taken.
\subsection{Sequential Regret}
\label{sec:sequential regret}
The following theorem provides a regret bound for the sequential setting.
\begin{theorem}[Sequential regret]
\label{thm:sequential regret} Let $|\mathcal{S}_t| = 1$ for all rounds $t$ and $\mathcal{A} \subseteq \mathbb{R}^d$. Choose $\delta = 1 / (m n)$. Then the Bayes regret of \ensuremath{\tt HierTS}\xspace is
\begin{align*}
\mathcal{BR}(m, n)
\leq d \sqrt{2 m n [c_1 m + c_2] \log(m n)} + c_3\,,
\end{align*}
where $c_3 = O(d^\frac{3}{2})$,
\begin{align*}
c_1
& = \frac{\lambda_1(\Sigma_0)}{\log(1 + \sigma^{-2} \lambda_1(\Sigma_0))}
\log\left(1 + \frac{\lambda_1(\Sigma_0) n}{\sigma^2 d}\right)\,, \\
c_2
& = \frac{c_q c}{\log(1 + \sigma^{-2} c_q)}
\log\left(1 + \frac{\lambda_1(\Sigma_q) m}{\lambda_d(\Sigma_0)}\right)\,, \\
c_q
& = \frac{\lambda_1^2(\Sigma_0) \lambda_1(\Sigma_q)}{\lambda_d^2(\Sigma_0)}, \quad
c = 1 + \sigma^{-2} \lambda_1(\Sigma_0)\,.
\end{align*}
\end{theorem}
The proof of \cref{thm:sequential regret} is based on three steps. First, we use \cref{lem:bayes regret}. Second, we employ \cref{lem:covariance decomposition} to decompose the posterior variance in any round into that of the task parameters and hyper-parameter. Finally, we apply elliptical lemmas to bound each term separately. \cref{thm:sequential regret} has a nice interpretation: $m \sqrt{c_1 n}$ is the regret for learning task parameters and $\sqrt{c_2 m n}$ is the regret for learning the hyper-parameter $\mu_*$. We elaborate on both terms below.
The term $m \sqrt{c_1 n}$ represents the regret for solving $m$ bandit tasks, which are sampled i.i.d.\ from a known prior $\mathcal{N}(\mu_*, \Sigma_0)$. Under this assumption, no task provides information about any other task, and thus the term is linear in $m$. The constant $c_1$ is $O(\lambda_1(\Sigma_0))$ and reflects the dependence on the prior width $\sqrt{\lambda_1(\Sigma_0)}$. Roughly speaking, when the task prior is half as informative, $\sqrt{c_1}$ doubles and so does $m \sqrt{c_1 n}$. This is the expected scaling with conditional uncertainty of $\theta_{s, *}$ given $\mu_*$.
The term $\sqrt{c_2 m n}$ is the regret for learning the hyper-parameter. Asymptotically, it is $O(\sqrt{m})$ smaller than $m \sqrt{c_1 n}$. Therefore, for a large number of tasks $m$, its contribution to the total regret is negligible. This is why hierarchical Bayesian bandits perform so well in practice. The constant $c_2$ is $O(\lambda_1(\Sigma_q))$ and reflects the dependence on the hyper-prior width $\sqrt{\lambda_1(\Sigma_0)}$. When the hyper-prior is half as informative, $\sqrt{c_2}$ doubles and so does $\sqrt{c_2 m n}$. This is the expected scaling with the marginal uncertainty of $\mu_*$.
\subsection{Tightness of Regret Bounds}
\label{sec:tightness}
One shortcoming of our current analysis is that we do not provide a matching lower bound.
To the best of our knowledge, Bayes regret lower bounds are rare and do not match existing upper bounds. The only lower bound that we are aware of is $\Omega(\log^2 n)$ in Theorem 3 of \citet{lai87adaptive}. The bound is for $K$-armed bandits and it is unclear how to apply it to structured problems. Seminal works on Bayes regret minimization \citep{russo14learning,russo16information} do not match it. Therefore, to show that our problem structure is reflected in our bound, we compare the regret of \ensuremath{\tt HierTS}\xspace to baselines that have more information or use less structure.
Now we compare the regret of \ensuremath{\tt HierTS}\xspace to two \ensuremath{\tt LinTS}\xspace \citep{agrawal13thompson} baselines that do use our hierarchical model. This first is an oracle \ensuremath{\tt LinTS}\xspace that knows $\mu_*$, and so has more information than \ensuremath{\tt HierTS}\xspace. Its Bayes regret would be as in \cref{thm:sequential regret} with $c_2 = 0$. Not surprisingly, it is lower than that of \ensuremath{\tt HierTS}\xspace. The second baseline is \ensuremath{\tt LinTS}\xspace that knows that $\mu_* \sim \mathcal{N}(\mu_q, \Sigma_q)$, but does not model the structure that the tasks share $\mu_*$. In this case, each task parameter can be viewed as having prior $\mathcal{N}(\mu_q, \Sigma_q + \Sigma_0)$. The regret of this algorithm would be as in \cref{thm:sequential regret} with $c_2 = 0$, while $\lambda_1(\Sigma_0)$ in $c_1$ would be $\lambda_1(\Sigma_q + \Sigma_0)$. Since $c_1$ is multiplied by $m$ while $c_2$ is not, \ensuremath{\tt HierTS}\xspace would have lower regret as $m \to \infty$. This is a powerful testament to the benefit of learning the hyper-parameter.
Finally, we want to comment on linear dependence in $d$ and $m$ in \cref{thm:sequential regret}. The dependence on $d$ is standard in Bayes regret analyses for linear bandits with infinitely many arms \citep{russo14learning,lu19informationtheoretic}. As for the number of tasks $m$, since the tasks are drawn i.i.d.\ from the same hyper-prior, they do not provide any additional information about each other. So, even if the hyper-parameter is known, the regret for learning to act in $m$ tasks with $n$ rounds would be $O(m \sqrt{n})$. Our improvements are in constants due to better variance attribution. Other bandit meta-learning works \citep{kveton21metathompson,basu21noregrets} made similar observations. Also note that the frequentist regret of \ensuremath{\tt LinTS}\xspace applied to $m$ independent linear bandit tasks is $\tilde{O}(m d^\frac{3}{2} \sqrt{n})$ \citep{agrawal13thompson}. This is worse by a factor of $\sqrt{d}$ than the bound in \cref{thm:sequential regret}.
\subsection{Concurrent Regret}
\label{sec:concurrent regret}
Now we investigate the concurrent setting, where the agent acts in up to $L$ tasks per round. This setting is challenging because the hyper-posterior $Q_t$ is not updated until the end of the round. This is because the task posteriors are not refined with the observations from concurrent tasks. This delayed feedback should increase regret. Before we show it, we make the following assumption on the action space.
\begin{assumption}
\label{ass:basis} There exist actions $\{a_i\}_{i = 1}^d \subseteq \mathcal{A}$ and $\eta > 0$ such that $\lambda_d(\sum_{i = 1}^d a_i a_i^\top) \geq \eta$.
\end{assumption}
This assumption is without loss of generality. Specifically, if $\mathbb{R}^d$ was not spanned by actions in $\mathcal{A}$, we could project $\mathcal{A}$ into a subspace where the assumption would hold. Our regret bound is below.
\begin{theorem}[Concurrent regret]
\label{thm:concurrent regret} Let $|\mathcal{S}_t| \leq L \leq m$ and $\mathcal{A} \subseteq \mathbb{R}^d$. Let $\delta = 1 / (m n)$. Then the Bayes regret of \ensuremath{\tt HierTS}\xspace is
\begin{align*}
\mathcal{BR}(m, n)
\leq d \sqrt{2 m n [c_1 m + c_2] \log(m n)} + c_3\,,
\end{align*}
where $c_1$, $c_q$, and $c$ are defined as in \cref{thm:sequential regret},
\begin{align*}
c_2
& = \frac{c_q c_4 c}{\log(1 + \sigma^{-2} c_q)}
\log\left(1 + \frac{\lambda_1(\Sigma_q) m}{\lambda_d(\Sigma_0)}\right)\,, \\
c_4
& = 1 + \frac{\sigma^{-2} \lambda_1(\Sigma_q) (\lambda_1(\Sigma_0) + \sigma^2 / \eta)}
{\lambda_1(\Sigma_q) + (\lambda_1(\Sigma_0) + \sigma^2 / \eta) / L}\,,
\end{align*}
and $c_3 = O(d^\frac{3}{2} m)$.
\end{theorem}
The key step in the proof is to modify \ensuremath{\tt HierTS}\xspace as follows. For the first $d$ interactions with any task $s$, we take actions $\{a_i\}_{i = 1}^d$. This guarantees that we explore all directions within the task, and allows us to bound losses from not updating the task posterior with concurrent observations. This modification of \ensuremath{\tt HierTS}\xspace is trivial and analogous to popular initialization in bandits, where each arm is pulled once in the first rounds \citep{auer02finitetime}.
The regret bound in \cref{thm:concurrent regret} is similar to that in \cref{thm:sequential regret}. There are two key differences. First, the additional scaling factor $c_4$ in $c_2$ is the price for taking concurrent actions. It increases as more actions $L$ are taken concurrently, but is sublinear in $L$. Second, $c_3$ arises due to trivially bounding $dm$ rounds of forced exploration. To the best of our knowledge, \cref{thm:concurrent regret} is the first Bayes regret bound where multiple bandit tasks are solved concurrently. Prior works only proved frequentist regret bounds \citep{yang21impact}.
\section{RELATED WORK}
\label{sec:related work}
The most related works are recent papers on bandit meta-learning \citep{bastani19meta,ortega19metalearning,cella20metalearning,kveton21metathompson,basu21noregrets,peleg21metalearning,simchowitz21bayesian}, where a learning agent interacts with a single task at a time until completion. Both \citet{kveton21metathompson} and \citet{basu21noregrets} represent their problems using graphical models and apply Thompson sampling to solve them. The setting of these papers is less general than ours. \citet{wan21metadatabased} study a setting where the tasks can arrive in any order. We differ from this work in several aspects. First, they only consider a $K$-armed bandit. Second, their model is different. In our notation, \citet{wan21metadatabased} assume that the mean reward of action $a$ in task $s$ is $x_{s, a}^\top \mu_*$ plus i.i.d.\ noise, where $x_{s, a}$ is an observed feature vector. The i.i.d.\ noise prevents generalization to a large number of actions. In our work, the mean reward of action $a$ in task $s$ is $a^\top \theta_{s, *}$, where $\theta_{s, *} \sim \mathcal{N}(\mu_*, \Sigma_0)$. Third, \citet{wan21metadatabased} derive a frequentist regret bound, which matches \cref{thm:sequential regret} asymptotically, but does not explicitly depend on prior widths. Finally, \citet{wan21metadatabased} do not consider the concurrent setting. To the best of our knowledge, we are the first to study Bayesian bandits with arbitrarily ordered and concurrent tasks.
The novelty in our analysis is the total covariance decomposition, which leads to better variance attribution in structured models than information-theoretic bounds \citep{russo16information,lu19informationtheoretic,basu21noregrets}. For instance, take Theorem 5 of \citet{basu21noregrets}, which corresponds to our sequential meta-learning setting. Forced exploration is needed to make their task term $O(\lambda_1(\Sigma_0))$. This is because the upper bound on the regret with filtered mutual information depends on the maximum marginal task parameter covariance, which can be $\lambda_1(\Sigma_q + \Sigma_0)$. In our analysis, the comparable term $c_1$ (\cref{thm:sequential regret}) is $O(\lambda_1(\Sigma_0))$ without any forced exploration. We also improve upon related analysis of \citet{kveton21metathompson} in several aspects. First, \citet{kveton21metathompson} analyze only a $K$-armed bandit. Second, they derive that the additional regret for meta-learning is $\tilde{O}(\sqrt{m} n^2)$; while our bound shows $\tilde{O}(\sqrt{m n})$. Finally, our setting generalizes bandit meta-learning.
Meta- and multi-task bandits have also been studied in the frequentist setting \citep{azar13sequential,deshmukh17multitask}. \citet{cella20metalearning} propose a \ensuremath{\tt LinUCB}\xspace algorithm \citep{abbasi-yadkori11improved} that constructs an ellipsoid around the unknown hyper-parameter in a linear bandit. The concurrent setting has also been studied, but with a different shared structure of task parameters. \citet{dubey20kernel} use a kernel matrix, \citet{wang21multitask} utilize pairwise distances of task parameters, and \citet{yang21impact} use low-rank factorization. Our structure, where the task parameters are drawn from an unknown prior, is both novel and important to study because it differs significantly from the aforementioned works. Earlier works on bandits with similar instances rely on clustering \citep{gentile14online,gentile17clustering,li16collaborative} and low-rank factorization \citep{kawale15efficient,sen17contextual,katariya16dcm,katariya17stochastic}. They analyze the frequentist regret, which is a stronger metric than the Bayes regret. Except for one work, all algorithms are UCB-like and conservative in practice. In comparison, \ensuremath{\tt HierTS}\xspace uses a natural stochastic structure. This makes it practical, to the point that the analyzed algorithm performs well in practice without any additional tuning.
Another related line of work are latent bandits \citep{maillard14latent,hong20latent,hong22thompson}, where the bandit problem is parameterized by an unknown latent state. If known, the latent state could help the agent to identify the bandit instance that it interacts with. These works reason about latent variables; but the purpose is different from our work, where we introduce the unknown hyper-parameter $\mu_*$ to relate multiple similar tasks.
\section{SETTING}
\label{sec:setting}
We use the following notation. Random variables are capitalized, except for Greek letters like $\theta$ and $\mu$. For any positive integer $n$, we define $[n] = \set{1, \dots, n}$. The indicator function is denoted by $\I{\cdot}$. The $i$-th entry of vector $v$ is $v_i$. If the vector is already indexed, such as $v_j$, we write $v_{j, i}$. A matrix with diagonal entries $v$ is $\diag{v}$. For any matrix $M \in \mathbb{R}^{d \times d}$, the maximum eigenvalue is $\lambda_1(M)$ and the minimum is $\lambda_d(M)$. The big O notation up to logarithmic factors is $\tilde{O}$.
Now we present our setting for solving similar bandit tasks. Each task is a \emph{bandit instance} with actions $a \in \mathcal{A}$, where $\mathcal{A}$ denotes an \emph{action set}. Rewards of actions are generated by \emph{reward distribution} $P(\cdot \mid a; \theta)$, where $\theta \in \Theta$ is an unknown parameter shared by all actions. We assume that the rewards are $\sigma^2$-sub-Gaussian and denote by $r(a; \theta) = \E{Y \sim P(\cdot \mid a; \theta)}{Y}$ the mean reward of action $a$ under $\theta$. The learning agent interacts with $m$ tasks. In a recommender system, each task could be an individual user. The task $s \in [m]$ is parameterized by a \emph{task parameter} $\theta_{s, *} \in \Theta$, which is sampled i.i.d.\ from a \emph{task prior distribution} $\theta_{s, *} \sim P(\cdot \mid \mu_*)$; which is parameterized by an unknown \emph{hyper-parameter} $\mu_*$.
The agent acts at discrete decision points, which are integers and we call them \emph{rounds}. At round $t \geq 1$, the agent is asked to act in a set of tasks $\mathcal{S}_t \subseteq [m]$. It takes actions $A_t = (A_{s, t})_{s \in \mathcal{S}_t}$, where $A_{s, t} \in \mathcal{A}$ is the action in task $s$; and receives rewards $Y_t = (Y_{s, t})_{s \in \mathcal{S}_t} \in \mathbb{R}^{|\mathcal{S}_t|}$, where $Y_{s, t} \sim P(\cdot \mid A_{s, t}; \theta_{s, *})$ is a stochastic reward for taking action $A_{s, t}$ in task $s$. The rewards are drawn i.i.d.\ from their respective distributions. The set $\mathcal{S}_t$ can depend arbitrarily on the history. The assumption that the action set $\mathcal{A}$ is the same across all tasks and rounds is only to simplify exposition.
\begin{figure}[t]
\centering
\input{graphical_model}
\caption{Graphical model of our hierarchical Bayesian bandit.}
\label{fig:setting}
\end{figure}
In \emph{hierarchical Bayesian bandits}, the hyper-parameter $\mu_*$ is initially sampled from a \emph{hyper-prior} $Q$ known by the learning agent. Our full model is given by
\begin{align*}
\mu_* & \sim Q\,, \\
\theta_{s, *} \mid \mu_*
& \sim P(\cdot \mid \mu_*)\,,
& \forall s \in [m]\,, \\
Y_{s, t} \mid A_{s, t}, \theta_{s, *}
& \sim P(\cdot \mid A_{s, t}; \theta_{s, *})\,,
& \forall t \geq 1, \, s \in \mathcal{S}_t\,,
\end{align*}
and also visualized in \cref{fig:setting}. Note that $P$ denotes both the task prior and reward distribution; but they can be distinguished based on their parameters. Our setting is an instance of a hierarchical Bayesian model commonly used in supervised learning \citep{lindley72bayes,zhang17survey}, and has been studied in bandits in special cases where tasks appear sequentially \citep{kveton21metathompson,basu21noregrets}.
Our learning agent interacts with each of $m$ tasks for at most $n$ times. So the total number of rounds varies, as it depends on the number of tasks that the agent interacts with simultaneously in each round. However, the maximum number of interactions is $m n$. The goal is to minimize the \emph{Bayes regret} \citep{russo14learning} defined as
\begin{align*}
\mathcal{BR}(m, n)
= \E{}{\sum_{t \geq 1} \sum_{s \in \mathcal{S}_t}
r(A_{s, *}; \theta_{s, *}) - r(A_{s, t}; \theta_{s, *})}\,,
\end{align*}
where $A_{s, *} = \argmax_{a \in \mathcal{A}} r(a; \theta_{s, *})$ is the optimal action in task $s$. The expectation in $\mathcal{BR}(m, n)$ is over $\mu_*$, $\theta_{s, *}$ for $s \in [m]$, and actions $A_{s,t}$ of the agent. While weaker than a traditional frequentist regret, the Bayes regret is a practical performance metric, as we are often interested in an average performance \citep{hong20latent,kveton21metathompson}. For example, when recommending to a group of users, it is natural to optimize over the whole population rather than an individual. Our definition of $\mathcal{BR}(m, n)$ is also dictated by the fact that $m$ and $n$ are the primary quantities of interest in our regret analyses. Our goal is to minimize $\mathcal{BR}(m, n)$ without knowing $\mu_*$ and $\theta_{s, *}$ a priori.
Since the set of tasks $\mathcal{S}_t$ can be chosen arbitrarily, our setting is general and subsumes many prior settings. For instance, when the agent interacts with the same task for $n$ rounds before shifting to the next one, and $\mathcal{S}_t = \{\lceil t / n \rceil\}$, we get a \emph{meta-learning bandit} \citep{kveton21metathompson}. More generally, when the agent interacts with the tasks sequentially, $|\mathcal{S}_t|$ = 1, our setting can be viewed as multi-task learning where any task helps the agent to solve other tasks. Therefore, we recover a \emph{multi-task bandit} \citep{wan21metadatabased}. Finally, when the agent acts in multiple tasks concurrently, $|\mathcal{S}_t| > 1$, we recover the setting of \emph{collaborative filtering bandits} \citep{gentile14online,li16collaborative} or more recently \emph{federated bandits} \citep{shi21federated}. Our algorithm and its analysis apply to all of these settings.
|
{
"redpajama_set_name": "RedPajamaArXiv"
}
| 668 |
[
128000,
59,
2879,
90,
984,
75685,
534,
59,
1530,
90,
5132,
25,
20266,
633,
59,
7413,
90,
20266,
44489,
83,
933,
220,
1144,
24232,
90,
76009,
46334,
26224,
25936,
13,
534,
220,
1144,
1530,
90,
24823,
25,
2641,
534,
220,
1144,
7413,
90,
20266,
292,
44489,
16,
933,
262,
1144,
1423,
1144,
1342,
13536,
90,
2566,
25,
92,
33832,
2320,
3334,
400,
48,
26101,
262,
1144,
1423,
9185,
400,
48,
62,
16,
1144,
19209,
1229,
26101,
262,
1144,
2520,
7148,
83,
284,
220,
16,
11,
220,
17,
11,
1144,
68916,
3,
534,
415,
1144,
1423,
4761,
13570,
9256,
59060,
10590,
5531,
90,
50,
20009,
83,
1144,
53496,
510,
76,
60,
26101,
415,
1144,
1423,
19690,
59060,
15479,
530,
1144,
15124,
1229,
530,
26101,
415,
1144,
2520,
7148,
82
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
59,
2879,
90,
984,
75685,
534,
59,
1530,
90,
5132,
25,
20266,
633,
59,
7413,
90,
20266,
44489,
83,
933,
220,
1144,
24232,
90,
76009,
46334,
26224,
25936,
13,
534,
220,
1144,
1530,
90,
24823,
25,
2641,
534,
220,
1144,
7413,
90,
20266,
292,
44489,
16,
933,
262,
1144,
1423,
1144,
1342,
13536,
90,
2566,
25,
92,
33832,
2320,
3334,
400,
48,
26101,
262,
1144,
1423,
9185,
400,
48,
62,
16,
1144,
19209,
1229,
26101,
262,
1144,
2520,
7148,
83,
284,
220,
16,
11,
220,
17,
11,
1144,
68916,
3,
534,
415,
1144,
1423,
4761,
13570,
9256,
59060,
10590,
5531,
90,
50,
20009,
83,
1144,
53496,
510,
76,
60,
26101,
415,
1144,
1423,
19690,
59060,
15479,
530,
1144,
15124,
1229,
530,
26101,
415,
1144,
2520,
7148,
82,
-100
] |
Defraggler is a free product from the good guys who make CCleaner, which I use often. It has been in a beta program for months, and the improvements being made are impressive.
It's simple, but fast. I can't think of a single program that is able to analyse a hard drive quicker. And often the file defrag time is faster than I have come to expect. I always assumed that contig would remain my simple defrag utility of choice, but I am finding Defraggler indispensable.
I use an external USB hard drive for making backups, and once I have used SyncToy to make the backups, I do a quick check with Defraggler to make sure that nothing is fragmented. (See screen shot). A quick selection of the fragmented files, and a click on the "Defrag"button and the files are tidied up in no time.
Defraggler is not a full-featured defrag tool yet. It makes no attempt to optimise file placement, and relies on the availability of a block of free space big enough for the entire file being defragged. But when speed is required, or a quick check to see if there are a lot of fragmented files on a drive, this has become my first port-of-call.
It does not attempt to compete with fully-featured products like JkDefrag or PerfectDisk, especially with regard to file placement, but if you're used to using the built-in Windows defrag utility, then give Defraggler a try at www.defraggler.com. It's a small 445kb download, with no nagware or annoying ads: just a quick and simple way to tidy up your hard drive quickly.
Stephen Blewett, the MD of Autopage called me this morning to apologise about the Look4Me, Look4Help issues. He told me that Autopage does not give out its customer information to anyone, and that if he catches any Autopage employee doing so he will fire them personally.
He explained to me that the call centres guess the number ranges of his and other companies, and then make calls without their knowledge. The first time they hear of the billing is when it comes from the network provider (i.e. Vodacom) as a WASP billing record. This system is full of problems, and can be abused by the WASPs. As an example of call centres guessing cell phone numbers, he told me of his wife being called to be offered a "better deal" on her phone.
Stephen also personally apologised on the HelloPeter web site.
I guess the question is why Cellfind claims to have obtained my number from Autopage and why they say they have contacted Autopage to ensure I am on their DNC list. Is it more ducking and diving? Eugenie from Autopage was also extremely doubtful that Autopage had supplied my number to Cellfind. Yet both Carla Legatt and Neil Barnard from Cellfind say it was so. Perhaps they are getting confused with who they sent the billing info to.
Update 11:30am: Andre Geboers from Digicall phoned to accept responsibility for his call centre's part in the process. They didn't check the numbers against the DMA list at the time of making the call, because they received their first list from the DMA on 1st December. They are doing checks against the DMA's list now. Because he was willing to accept personal responsibility for the call, I have dropped my demand for an advert in the Sunday Times.
Not only did the Sunday Times fail to read David Bullard's column before publishing it, but they failed to uphold the press freedom they so ardently espouse. We are expected to become alarmed when a media buyout threatens their independence, or the editor is arrested, but agree with them when they fire a columnist without even the option of an apology. Clueless.
Imagine for a moment what life would be like in South Africa if the evil white man hadn't come to disturb the rustic idyll of the early black settlers.
Ignored by the Portuguese and Dutch, except as a convenient resting point en route to India. Shunned by the British, who had decided that their empire was already large enough and didn't need to include bits of Africa.
The vast mineral wealth lying undisturbed below the Highveld soil as simple tribesmen graze their cattle blissfully unaware that beneath them lies one of the richest gold seams in the world. But what would they want with gold?
There are no roads because no roads are needed because there are no cars. It's 2008 and no one has taken the slightest interest in South Africa, apart from a handful of botanists and zoologists who reckon that the country's flora and fauna rank as one of the largest unspoilt areas in a polluted world.
Because they have never been exposed to the sinful ways of the West, the various tribes of South Africa live healthy and peaceful lives, only occasionally indulging in a bit of ethnic cleansing.
Their children don't watch television because there is no television to watch. Instead they listen to their grandparents telling stories around a fire. They live in single-storey huts arranged to catch most of the day's sunshine and their animals are kept nearby.
Nobody has any more animals than his family needs and nobody grows more crops than he requires to feed his family and swap for other crops. Ostentation is unknown because what is the point of trying to impress your fellow citizens when they are not impressible?
The dreaded Internet doesn't exist in South Africa and cellphone companies have laughed off any hope of interesting the inhabitants in talking expensively into a piece of black plastic. There are no unsightly shopping malls selling expensive goods made by Asian slave workers and consequently there are no newspapers or magazines carrying articles comparing the relative merits of ladies' handbags.
Whisky, the curse of the white man, isn't known in this undeveloped land and neither are cigars. The locals brew a sort of beer out of vegetables and drink it out of shallow wooden bowls. Five-litre paint cans have yet to arrive in South Africa.
Every so often a child goes missing from the village, eaten either by a hungry lion or a crocodile. The family mourn for a week or so and then have another child. Life is, on the whole, pretty good but there is something vital missing. Being unaware of the temptations of the outside world, nobody knows what it is. Fire has been discovered and the development of the wheel is coming on nicely but the tribal elders are still aware of some essential happiness ingredient they still need to discover. Praying to the ancestors is no help because they are just as clueless.
Then something happens that will change this undisturbed South Africa forever. Huge metal ships land on the coast and big metal flying birds are sent to explore the sparsely populated hinterland. They are full of men from a place called China and they are looking for coal, metal, oil, platinum, farmland, fresh water and cheap labour and lots of it. Suddenly the indigenous population realise what they have been missing all along: someone to blame. At last their prayers have been answered.
At the outset, I must state that the column should never have been published in the first place.
The Sunday Times respects its readers and — while it is our mission to provoke, challenge and entertain — will not insult or abuse them.
The sad fact is that this idyllic pre-colonial existence is taught in schools and universities, but the irony is that this myth is the invention of a French colonialist. Once again, euro-centric political correctness is suckered by euro-centric colonial claptrap.
It seems to be a national pastime that whenever someone is accused of incompetence or stupidity, the race card is played. And if it isn't the race card, it's the colonial/imperial card.
Face it: Africa is a hard place with hard leaders who cling to power and amass vast personal fortunes while destroying their country. While some of the moderate leaders sing songs about machine guns, others buy them in bulk and use them against their rivals, when they are not torturing them or locking them up on Robben Island.
"David Bullard's weekly Out to Lunch column is the finest example of the benefits of a free and unshackled press. Let's hope it continues to be so."
Quote from the dust jacket of Bullard's book "Screw it, Let's Do Lunch", written by Sunday Times Editor Mondli Makhanya, the same person who fired him!?
Confirmation: Your wife's number was added to our internal Do Not Call when you spoke to Carla yesterday. Your number was added this morning after you responded to my initial e-mail.
Cellfind notified you/your wife on at least 12 occasions via SMS of the changes. [Earlier in the day you told me that you had only sent 4 messages on 30-11-2007 and 2 on 10-04-2008. I wonder when the other 6 were sent, if at all?] I cannot speak for them but Autopage as a rule notify every customer, every time a new service is added to their account (Also via SMS).
Let me ensure [sic] you that I take this very seriously and even though in our view, we are not in contravention of any law, we have made every effort possible to ensure we don't have a repeat occurrence and have provided you with as much assistance as humanly possible. So, please let me know what you see as adequate?
Since you ask, Mr Barnard, it would be helpful if you actually stopped ducking and diving and pushing responsibility onto someone else. Autopage did not make the call or send the first 4 SMS messages. Each one of those messages violates the ECT Act, even though you claim you didn't know that I should have been on your list. That was due to your negligence, not mine. Ignorance is no excuse. A thief who gets caught with the stolen goods is still a thief, even if he says "sorry" and gives the goods back.
Secondly, adding someone onto your database without their permission is dishonest, and expecting them to opt out is even more so. In fact it's a bloody cheek, and not the best way to treat your customers. The correct option is "Opt-in", where customers confirm that they WANT the service, rather than being expected to say if they don't, assuming they heard the question.
Thirdly, your call centre first told me that they had obtained the number from Vodacom, and then the story changed later to Autopage. They have yet to confirm that they did, in fact, supply the number. Who am I supposed to believe?
It's all very well to "fire" the person who called my wife, but when are you going to tell me that your company or Digicall made any attempt to contact ALL the people this person subscribed, to make sure they were aware of paying for the service. This could be done by sending an SMS to all the "subscribers", asking them to confirm that they still WANT the service. Clearly this did not happen because we did not receive any messages "several months ago" when the alleged bad apple was "fired".
Either your company is completely incompetent and cavalier in its approach to customer privacy; or you are systematically exploiting loopholes and using convenient excuses to pacify irate customers while you continue to deduct money on a monthly basis from thousands of unsuspecting consumers. I don't know which of the two it is, but neither is a good idea.
If you are genuinely concerned about people being subscribed without their knowledge or permission, then change your business model to an "opt-in" one, and send an SMS to all your existing subscribers asking them to confirm that they still want the service. Also, publish an apology on your web site for the inconvenience caused by your current dodgy practices. Then I'll believe you. Until then, CellFind is the company who got caught with its hands in the cookie jar.
Update Monday 14 th April: Autopage company policy is that they don't give out lists of numbers to anyone. What do you say to that, Mr Barnard? You were very vague about who you got the numbers from.
Update 21 November 2008: It seems that their practices have not changed. Here is an article from the Independent Online web site.
In November a call centre agent phoned my wife's cell phone and offered her services called "Look4Me" and "Look4Help". We don't like marketing calls and she did not agree to anything. Even so, we started getting billed for this service, in spite of the fact that no permission was obtained from the account holder, i.e. me, violating AutoPage's own policy on customer accounts.
So I called AutoPage (0860 23 2424) and they promised to cancel the service. I also asked for a refund. Judging by the complaints on HelloPeter, I am not the only one to be billed in this way. As far as I can tell this is a case of fraud.
Then I called the Look4Help call centre for on 086 010 5774, and was told that Vodacom had supplied them with my wife's number. Later I received an email from Carla Leggatt who claimed that AutoPage had supplied them with the number. Since AutoPage has done this before, I am not surprised, but now CellFind have added an ECT Act violation to their billing fraud. By phoning a number that is already marked as a Do Not Call (both by Autopage and the Direct Marketing Association), they violated the ECT Act, and the penalty for this is 1 year in jail, with no option of a fine.
What also gets me is that at no point was I, as the account holder, notified, either by mail, email or SMS, that we had been signed up for this service, and it was only because I checked my cell phone bill that I noticed the problem. Scumbags like this should rot in jail, and I am considering charges against them for both fraud and privacy violations, unless they make a public apology, preferably in the national newspapers, and request anyone else billed in this way to contact them for a refund.
Update Friday 10am: Niel Barnard from CellFind confirmed that the details were obtained from AutoPage, and admitted that they did not check the data against the DMA's list. He has promised to send me a copy of the recording of the original call. I spoke to the secretary for the MD for AutoPage, who put me through to customer services without really listening to my enquiry. I spoke to Andre Geboers at Digicall, the call centre company in Cape Town (021-405-1425), and he explained that they only get cell numbers from AutoPage, not ID numbers, so they can't really do any checks against the DMA list. It seems to me like a deliberate loophole to avoid prosecution, and the call verification process sounds a little vague. I have asked for a copy of the "script" used by the call centre agents, to see how well it complies with the DMA's Code of Conduct.
The person who spoke to your wife has in actual fact been fired a few months ago. I have an agreement with Digicall that any transgression of the script, etc will lead to an immediate dismissal of the offending agent on their side.
How convenient. Employ a temp worker for a few months (say November and December), sign up as many people as possible, legally or not, and then when anyone complains tell them that the person was fired. But they cannot tell me who this person was, or whether any of the people he/she called have been contacted to confirm that they did, in fact, agree to the service. So the firing does not absolve the company of the ECT Act violations or the billing fraud, for which the company as a whole is responsible.
So if you don't get the messages, ignore them, or just delete them as spam, you have effectively "agreed" to their service by not doing anything. Not only is this dishonest and unethical, it's also a violation of the DMA's Code of Conduct.
Update Monday 14 th April: Autopage company policy is that they don't give out lists of numbers to anyone.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 9,030 |
[
128000,
2685,
4193,
70,
1565,
374,
264,
1949,
2027,
505,
279,
1695,
7752,
889,
1304,
13844,
2742,
261,
11,
902,
358,
1005,
3629,
13,
1102,
706,
1027,
304,
264,
13746,
2068,
369,
4038,
11,
323,
279,
18637,
1694,
1903,
527,
16358,
627,
2181,
596,
4382,
11,
719,
5043,
13,
358,
649,
956,
1781,
315,
264,
3254,
2068,
430,
374,
3025,
311,
49586,
264,
2653,
6678,
39392,
13,
1628,
3629,
279,
1052,
711,
4193,
892,
374,
10819,
1109,
358,
617,
2586,
311,
1755,
13,
358,
2744,
19655,
430,
687,
343,
1053,
7293,
856,
4382,
711,
4193,
15919,
315,
5873,
11,
719,
358,
1097,
9455,
3979,
4193,
70,
1565,
64284,
627,
40,
1005,
459,
9434,
11602,
2653,
6678,
369,
3339,
60766,
11,
323,
3131,
358,
617,
1511,
30037,
69070,
311
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
2685,
4193,
70,
1565,
374,
264,
1949,
2027,
505,
279,
1695,
7752,
889,
1304,
13844,
2742,
261,
11,
902,
358,
1005,
3629,
13,
1102,
706,
1027,
304,
264,
13746,
2068,
369,
4038,
11,
323,
279,
18637,
1694,
1903,
527,
16358,
627,
2181,
596,
4382,
11,
719,
5043,
13,
358,
649,
956,
1781,
315,
264,
3254,
2068,
430,
374,
3025,
311,
49586,
264,
2653,
6678,
39392,
13,
1628,
3629,
279,
1052,
711,
4193,
892,
374,
10819,
1109,
358,
617,
2586,
311,
1755,
13,
358,
2744,
19655,
430,
687,
343,
1053,
7293,
856,
4382,
711,
4193,
15919,
315,
5873,
11,
719,
358,
1097,
9455,
3979,
4193,
70,
1565,
64284,
627,
40,
1005,
459,
9434,
11602,
2653,
6678,
369,
3339,
60766,
11,
323,
3131,
358,
617,
1511,
30037,
69070,
311,
-100
] |
Q: Hidden Youtube autoplay on click image Need some help please,
I am making an unordered list with once an image once an embed iframe from youtube,
when clicking on the image it replaces the image with the hidden iframe from above using the .replace and .prev function of jquery
I was wondering if after clicking on the image the youtube video could autoplay, when adding autoplay to the parameters of youtube itself, it starts on page load, even when being set to display none.
I am doing this with using only classes, because the list will get big and I don't want to add for every separate id some jquery. I am not sure if this is even possible by only using classes.
Could someone guide me a little.
this is what I came up so far:
javascript:
$('.coverimageforplayer').click(function () {
$(this).replaceWith($(this).prev('li.showme').show());
});
html:
<ul>
<li class="showme" style="display:none">
<iframe width="480" height="360" src="//www.youtube.com/embed/TZMoS2QBc8U?autoplay=0" frameborder="0" allowfullscreen></iframe>
</li>
<li class="coverimageforplayer">
<img src="http://www.nasa.gov/images/content/711375main_grail20121205_4x3_946-710.jpg" />
</li>
<li class="showme" style="display:none">
<iframe class="iframer_1" width="480" height="360" src="//www.youtube.com/embed/TZMoS2QBc8U" frameborder="0" allowfullscreen></iframe>
</li>
<li class="coverimageforplayer">
<img src="http://www.nasa.gov/images/content/711375main_grail20121205_4x3_946-710.jpg" />
</li>
</ul>
thank you very much
A: I would ditch the showme list item as this adds unnecerary html and makes your code too complex in the long run.
I've written a small fiddle to demonstrate a more elegant approach, stash the video url in a data attribute on the list image, then insert a iframe on click and voila!
Here's the fiddle: http://jsfiddle.net/TxbHx/1/
The html:
<ul>
<li>autoplays:</li>
<li class="coverimageforplayer" data-videoSRC="//www.youtube.com/embed/TZMoS2QBc8U?autoplay=1">
<img src="http://www.nasa.gov/images/content/711375main_grail20121205_4x3_946-710.jpg" />
</li>
<li>Does not autoplay:</li>
<li class="coverimageforplayer" data-videoSRC="//www.youtube.com/embed/TZMoS2QBc8U?autoplay=0">
<img src="http://www.nasa.gov/images/content/711375main_grail20121205_4x3_946-710.jpg" />
</li>
</ul>
The javascript:
$('.coverimageforplayer').on('click', function() {
var element = $(this); // reuse variables for best practice. http://code.tutsplus.com/tutorials/quick-tip-jquery-newbs-stop-jumping-in-the-pool--net-22142
var videoSRC = element.attr('data-videoSRC'); // Get the video URL from the data attribute
var iframe = '<iframe width="480" height="360" src="'+videoSRC+'" frameborder="0" allowfullscreen></iframe>'; // create the iframe string
element.html(iframe); // insert the iframe
});
|
{
"redpajama_set_name": "RedPajamaStackExchange"
}
| 5,955 |
[
128000,
48,
25,
35342,
38403,
58410,
389,
4299,
2217,
14998,
1063,
1520,
4587,
345,
40,
1097,
3339,
459,
56833,
1160,
449,
3131,
459,
2217,
3131,
459,
11840,
49513,
505,
28277,
11,
720,
9493,
18965,
389,
279,
2217,
433,
41800,
279,
2217,
449,
279,
8340,
49513,
505,
3485,
1701,
279,
662,
8319,
323,
662,
10084,
734,
315,
62801,
198,
40,
574,
20910,
422,
1306,
18965,
389,
279,
2217,
279,
28277,
2835,
1436,
58410,
11,
994,
7999,
58410,
311,
279,
5137,
315,
28277,
5196,
11,
433,
8638,
389,
2199,
2865,
11,
1524,
994,
1694,
743,
311,
3113,
7000,
627,
40,
1097,
3815,
420,
449,
1701,
1193,
6989,
11,
1606,
279,
1160,
690,
636,
2466,
323,
358,
1541,
956,
1390,
311,
923,
369,
1475,
8821,
887,
1063,
62801,
13,
358,
1097
] |
[
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
] |
[
48,
25,
35342,
38403,
58410,
389,
4299,
2217,
14998,
1063,
1520,
4587,
345,
40,
1097,
3339,
459,
56833,
1160,
449,
3131,
459,
2217,
3131,
459,
11840,
49513,
505,
28277,
11,
720,
9493,
18965,
389,
279,
2217,
433,
41800,
279,
2217,
449,
279,
8340,
49513,
505,
3485,
1701,
279,
662,
8319,
323,
662,
10084,
734,
315,
62801,
198,
40,
574,
20910,
422,
1306,
18965,
389,
279,
2217,
279,
28277,
2835,
1436,
58410,
11,
994,
7999,
58410,
311,
279,
5137,
315,
28277,
5196,
11,
433,
8638,
389,
2199,
2865,
11,
1524,
994,
1694,
743,
311,
3113,
7000,
627,
40,
1097,
3815,
420,
449,
1701,
1193,
6989,
11,
1606,
279,
1160,
690,
636,
2466,
323,
358,
1541,
956,
1390,
311,
923,
369,
1475,
8821,
887,
1063,
62801,
13,
358,
1097,
-100
] |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.