Update spaCy pipeline

.gitattributes

@@ -37,3 +37,4 @@ en_extract_spanfinder-any-py3-none-any.whl filter=lfs diff=lfs merge=lfs -text
 tok2vec/model filter=lfs diff=lfs merge=lfs -text
 vocab/key2row filter=lfs diff=lfs merge=lfs -text
 vocab/vectors filter=lfs diff=lfs merge=lfs -text
+en_extract_definitions-any-py3-none-any.whl filter=lfs diff=lfs merge=lfs -text

README.md

@@ -4,12 +4,12 @@ tags:
 language:
 - en
 model-index:
-- name: en_extract_spanfinder
+- name: en_extract_definitions
   results: []
 ---
 | Feature | Description |
 | --- | --- |
-| **Name** | `en_extract_spanfinder` |
+| **Name** | `en_extract_definitions` |
 | **Version** | `0.0.1` |
 | **spaCy** | `>=3.7.5,<3.8.0` |
 | **Default Pipeline** | `tok2vec`, `span_finder` |
@@ -23,8 +23,8 @@ model-index:
 
 | Type | Score |
 | --- | --- |
-| `SPANS_SC_F` | 80.00 |
-| `SPANS_SC_P` | 80.00 |
-| `SPANS_SC_R` | 80.00 |
-| `TOK2VEC_LOSS` | 0.00 |
-| `SPAN_FINDER_LOSS` | 0.76 |
+| `SPANS_SC_F` | 67.80 |
+| `SPANS_SC_P` | 83.33 |
+| `SPANS_SC_R` | 57.14 |
+| `TOK2VEC_LOSS` | 543.30 |
+| `SPAN_FINDER_LOSS` | 7154.03 |

config.cfg

@@ -1,6 +1,6 @@
 [paths]
-train = "data/train-definitions.spacy"
-dev = "data/train-definitions.spacy"
+train = "data/bin/train-spanfinder.spacy"
+dev = "data/bin/train-spanfinder.spacy"
 vectors = "en_core_web_lg"
 init_tok2vec = null
 

en_extract_definitions-any-py3-none-any.whl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:35548c4108d255a960099ebece3e2118ab19647b089d3043dbb4e12858ef3529
+size 580767399

meta.json

@@ -1,6 +1,6 @@
 {
   "lang":"en",
-  "name":"extract_spanfinder",
+  "name":"extract_definitions",
   "version":"0.0.1",
   "description":"",
   "author":"",
@@ -35,11 +35,11 @@
 
   ],
   "performance":{
-    "spans_sc_f":0.8,
-    "spans_sc_p":0.8,
-    "spans_sc_r":0.8,
-    "tok2vec_loss":0.000000241,
-    "span_finder_loss":0.0075788116
+    "spans_sc_f":0.6779661017,
+    "spans_sc_p":0.8333333333,
+    "spans_sc_r":0.5714285714,
+    "tok2vec_loss":5.4330211096,
+    "span_finder_loss":71.5402888331
   },
   "requirements":[
 

tok2vec/model

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f8b40bf78edde3d73e8ffe8d9167f60803e94e6eb87994f078d90a6981f8a980
+oid sha256:994c729d86422b01c00fca6e9790a4d582373ed6932f3af750aa771e8da258d3
 size 6495793

vocab/strings.json

@@ -17,6 +17,8 @@
   "\t\u001e",
   "\t\u001f",
   "\t ",
+  "\t    ",
+  "\t           ",
   "\t\u0085",
   "\t\u0085\u1680",
   "\t\u00a0",
@@ -104,6 +106,8 @@
   "\r\t",
   "\r\n",
   "\r\n\r\n",
+  "\r\n\r\n\r\n",
+  "\r\n \r\n",
   "\r\u000b",
   "\r\f",
   "\r\u001c",
@@ -214,11 +218,15 @@
   " \u000b",
   " \f",
   " \r",
+  " \r\n",
+  " \r\n\r\n",
   " \u001c",
   " \u001d",
   " \u001e",
   " \u001f",
   "  ",
+  "  \r\n",
+  "   ",
   " \u00a0",
   " \u1680",
   " \u2000",
@@ -726,6 +734,8 @@
   "(._.)",
   "(02)",
   "(10",
+  "(3A",
+  "(3a",
   "(:",
   "(;",
   "(=",
@@ -6188,6 +6198,7 @@
   "/You",
   "/You're",
   "/Your",
+  "/[a",
   "/\\",
   "/a",
   "/a/",
@@ -12791,6 +12802,7 @@
   "14.68",
   "14.69",
   "14.7",
+  "14.7.2",
   "14.70",
   "14.71",
   "14.72",
@@ -13285,6 +13297,7 @@
   "15.28",
   "15.29",
   "15.3",
+  "15.3.4",
   "15.30",
   "15.31",
   "15.32",
@@ -14951,11 +14964,13 @@
   "1809",
   "1809\u20131810",
   "1809\u20131882",
+  "180A",
   "180C",
   "180C.",
   "180C/160C",
   "180C/350F",
   "180W",
+  "180a",
   "180c",
   "180c.",
   "180c/160c",
@@ -21382,6 +21397,11 @@
   "2.8l",
   "2.8million",
   "2.9",
+  "2.9.1",
+  "2.9.2",
+  "2.9.3",
+  "2.9.4",
+  "2.9.5",
   "2.90",
   "2.91",
   "2.92",
@@ -24057,6 +24077,8 @@
   "24.68",
   "24.7",
   "24.70",
+  "24.709",
+  "24.720",
   "24.74",
   "24.75",
   "24.8",
@@ -24965,6 +24987,8 @@
   "27.01",
   "27.1",
   "27.10",
+  "27.1202",
+  "27.1202(b",
   "27.14",
   "27.2",
   "27.20",
@@ -25322,6 +25346,8 @@
   "2849",
   "284th",
   "285",
+  "285(3A",
+  "285(3a",
   "285,000",
   "2850",
   "2851",
@@ -27038,6 +27064,8 @@
   "3027",
   "3028",
   "3029",
+  "302A",
+  "302a",
   "302d",
   "302nd",
   "302s",
@@ -31948,6 +31976,7 @@
   "5(b",
   "5(c",
   "5(d",
+  "5(g",
   "5,0",
   "5,00",
   "5,00,000",
@@ -32056,6 +32085,7 @@
   "5.1.2",
   "5.1.3",
   "5.1.4",
+  "5.1.4.1",
   "5.10",
   "5.11",
   "5.12",
@@ -33964,6 +33994,7 @@
   "6.6L",
   "6.6l",
   "6.7",
+  "6.7.3",
   "6.7/10",
   "6.70",
   "6.71",
@@ -35380,6 +35411,9 @@
   "7.10",
   "7.10pm",
   "7.11",
+  "7.11(a",
+  "7.11(b",
+  "7.11(c",
   "7.12",
   "7.13",
   "7.14",
@@ -35851,6 +35885,7 @@
   "7230",
   "7233",
   "724",
+  "724(5",
   "7240",
   "7242",
   "7243",
@@ -35904,9 +35939,11 @@
   "73.1",
   "73.2",
   "73.3",
+  "73.3555",
   "73.4",
   "73.5",
   "73.6",
+  "73.658",
   "73.7",
   "73.8",
   "73.805",
@@ -36150,12 +36187,18 @@
   "76.0",
   "76.00",
   "76.1",
+  "76.1501",
   "76.171",
   "76.171.",
   "76.2",
   "76.3",
   "76.4",
   "76.5",
+  "76.501",
+  "76.503",
+  "76.504",
+  "76.505",
+  "76.505(g",
   "76.6",
   "76.66",
   "76.7",
@@ -37511,6 +37554,7 @@
   "87sr/86sr",
   "87th",
   "88",
+  "88(3",
   "88,000",
   "88,500",
   "88-",
@@ -46980,6 +47024,7 @@
   "Agreed",
   "Agreeing",
   "Agreement",
+  "Agreement]]/[insert",
   "Agreements",
   "Agrees",
   "Agrell",
@@ -103883,6 +103928,7 @@
   "Consensual",
   "Consensus",
   "Consent",
+  "Consent]/[the",
   "Consenting",
   "Consents",
   "Consequence",
@@ -110157,6 +110203,7 @@
   "DATA",
   "DATABASE",
   "DATED",
+  "DATED]/[AS",
   "DATEM",
   "DATES",
   "DATING",
@@ -119750,6 +119797,7 @@
   "Directo",
   "Directoire",
   "Director",
+  "Director(s",
   "Director-",
   "Director/",
   "Directorate",
@@ -131450,6 +131498,7 @@
   "Encuentra",
   "Encuentro",
   "Encuentros",
+  "Encumbrance",
   "Encumbrances",
   "Encyclia",
   "Encyclical",
@@ -135313,6 +135362,7 @@
   "Excruciating",
   "Excu",
   "Excubitors",
+  "Exculpation",
   "Excursion",
   "Excursions",
   "Excuse",
@@ -178429,6 +178479,7 @@
   "IT-",
   "IT--",
   "ITA",
+  "ITA).]The",
   "ITA)Created",
   "ITAA",
   "ITAC",
@@ -182944,6 +182995,7 @@
   "Investments",
   "Investopedia",
   "Investor",
+  "Investor(s",
   "InvestorPlace",
   "Investoren",
   "Investors",
@@ -206685,6 +206737,7 @@
   "LIMITATION",
   "LIMITATIONS",
   "LIMITED",
+  "LIMITED]/[LTD",
   "LIMITS",
   "LIMK1",
   "LIMO",
@@ -285772,6 +285825,7 @@
   "RESSLER",
   "REST",
   "RESTART",
+  "RESTATED",
   "RESTAURANT",
   "RESTAURANTS",
   "RESTING",
@@ -346064,6 +346118,7 @@
   "Transferability",
   "Transferable",
   "Transferee",
+  "Transferees",
   "Transference",
   "Transfermarkt",
   "Transferology",
@@ -372229,6 +372284,7 @@
   "XXX),xxxx",
   "XXX)-",
   "XXX).-",
+  "XXX).]Xxx",
   "XXX)XXX",
   "XXX)Xxxxx",
   "XXX)d",
@@ -372335,6 +372391,8 @@
   "XXXXXX",
   "XXXXXXX",
   "XXXXXXXXXX",
+  "XXXX]/[XX",
+  "XXXX]/[XXX",
   "XXXX]/xxxx",
   "XXXX_XXXX",
   "XXXX_XXXX_XXXX_XXXX",
@@ -373907,8 +373965,10 @@
   "Xxxxx\\Xxxxx\\",
   "Xxxxx\\Xxxxx\\Xxx",
   "Xxxxx\\Xxxxx\\Xxxxx",
+  "Xxxxx]/[xxx",
   "Xxxxx]Xxx",
   "Xxxxx][Xxxxx",
+  "Xxxxx]]/[xxxx",
   "Xxxxx^d",
   "Xxxxx_Xxxx",
   "Xxxxx_XxxxXxxxx",
@@ -380829,6 +380889,7 @@
   "[:",
   "[=",
   "[A14062]",
+  "[AS",
   "[Aleusis]",
   "[Andreas]",
   "[BlackKnife]",
@@ -380853,6 +380914,8 @@
   "[alvarado]",
   "[and",
   "[andreas]",
+  "[as",
+  "[at",
   "[bingladen]",
   "[blackknife]",
   "[buabua]",
@@ -387353,6 +387416,8 @@
   "agreeing",
   "agreement",
   "agreement(s",
+  "agreement]/[the",
+  "agreement]]/[insert",
   "agreements",
   "agrees",
   "agreethis",
@@ -390499,6 +390564,7 @@
   "all3media",
   "all4",
   "allAfrica.com",
+  "all]/[at",
   "alla",
   "allaah",
   "allach",
@@ -451091,6 +451157,7 @@
   "consensus",
   "consensys",
   "consent",
+  "consent]/[the",
   "consented",
   "consentement",
   "consenting",
@@ -460423,6 +460490,7 @@
   "d.dXxx",
   "d.dd",
   "d.dd&xxxx;d",
+  "d.dd(x",
   "d.dd.d",
   "d.dd.dd",
   "d.dd:d",
@@ -462909,6 +462977,7 @@
   "datebook",
   "datecredit",
   "dated",
+  "dated]/[as",
   "datei",
   "datejust",
   "dateless",
@@ -463629,8 +463698,10 @@
   "dd.dd.dd",
   "dd.dd/",
   "dd.ddd",
+  "dd.ddd(x",
   "dd.ddd.",
   "dd.dddd",
+  "dd.dddd(x",
   "dd.dddd/ddd",
   "dd.ddxx",
   "dd.d\u00d7ddd",
@@ -463707,6 +463778,9 @@
   "ddd'x",
   "ddd(X",
   "ddd(X)(d",
+  "ddd(d",
+  "ddd(dX",
+  "ddd(dx",
   "ddd(x",
   "ddd(x)(d",
   "ddd(x)d",
@@ -471221,6 +471295,7 @@
   "directo",
   "directoire",
   "director",
+  "director(s",
   "director-",
   "director/",
   "directora",
@@ -545114,6 +545189,7 @@
   "investments",
   "investopedia",
   "investor",
+  "investor(s",
   "investoren",
   "investorplace",
   "investors",
@@ -547001,6 +547077,7 @@
   "it`ll",
   "it`s",
   "ita",
+  "ita).]the",
   "ita)created",
   "itaa",
   "itabashi",
@@ -577612,6 +577689,7 @@
   "limitations",
   "limite",
   "limited",
+  "limited]/[ltd",
   "limitedly",
   "limiter",
   "limiters",
@@ -605915,6 +605993,7 @@
   "month.[1",
   "month.[3",
   "monthCredit",
+  "month]/[quarter",
   "monthan",
   "monthcredit",
   "montherlant",
@@ -623114,6 +623193,7 @@
   "number(s",
   "number-",
   "number--",
+  "number]]/[a",
   "numbered",
   "numbering",
   "numberings",
@@ -629258,6 +629338,7 @@
   "optionally",
   "optionals",
   "optioned",
+  "optionholder",
   "optioning",
   "options",
   "options-",
@@ -667265,6 +667346,7 @@
   "remunerated",
   "remunerating",
   "remuneration",
+  "remuneration]/[audit",
   "remunerations",
   "remunerative",
   "remus",
@@ -677323,6 +677405,7 @@
   "s.",
   "s.1",
   "s.2",
+  "s.204",
   "s.3",
   "s.4",
   "s.5",
@@ -724236,6 +724319,7 @@
   "time4learning",
   "timeCredit",
   "timeI",
+  "time]/[each",
   "timea",
   "timeanddate",
   "timebase",
@@ -760183,6 +760267,7 @@
   "xxx(d",
   "xxx(x",
   "xxx(xxx",
+  "xxx).]xxx",
   "xxx).d.dd",
   "xxx)||",
   "xxx**",
@@ -760308,6 +760393,7 @@
   "xxx\\'xx",
   "xxx\\\u2019x",
   "xxx\\\u2019xx",
+  "xxx]/[xx",
   "xxx_dddd.xxx",
   "xxx_xx_XX",
   "xxx_xxxx",
@@ -760680,7 +760766,12 @@
   "xxxx[xxxx[x",
   "xxxx\\",
   "xxxx\\\u2019x",
+  "xxxx]/[xx",
+  "xxxx]/[xxx",
+  "xxxx]/[xxxx",
   "xxxx][x]Xxxxx",
+  "xxxx]]/[x",
+  "xxxx]]/[xxxx",
   "xxxx]]Xxxxx",
   "xxxx]xxx",
   "xxxx_XXX",