End of training
Browse files- README.md +6 -4
- all_results.json +14 -0
- runs/Aug03_04-21-22_1d636f5ddae9/events.out.tfevents.1722665036.1d636f5ddae9.53.1 +3 -0
- test_results.json +9 -0
- trainer_state.json +458 -0
- training_results.json +8 -0
README.md
CHANGED
@@ -2,6 +2,8 @@
|
|
2 |
license: apache-2.0
|
3 |
base_model: microsoft/swinv2-base-patch4-window8-256
|
4 |
tags:
|
|
|
|
|
5 |
- generated_from_trainer
|
6 |
datasets:
|
7 |
- generator
|
@@ -15,11 +17,11 @@ should probably proofread and complete it, then remove this comment. -->
|
|
15 |
|
16 |
# swinv2-base-panorama-IQA
|
17 |
|
18 |
-
This model is a fine-tuned version of [microsoft/swinv2-base-patch4-window8-256](https://huggingface.co/microsoft/swinv2-base-patch4-window8-256) on the
|
19 |
It achieves the following results on the evaluation set:
|
20 |
-
- Loss: 0.
|
21 |
-
- Srocc: 0.
|
22 |
-
- Lcc: 0.
|
23 |
|
24 |
## Model description
|
25 |
|
|
|
2 |
license: apache-2.0
|
3 |
base_model: microsoft/swinv2-base-patch4-window8-256
|
4 |
tags:
|
5 |
+
- image-classification
|
6 |
+
- vision
|
7 |
- generated_from_trainer
|
8 |
datasets:
|
9 |
- generator
|
|
|
17 |
|
18 |
# swinv2-base-panorama-IQA
|
19 |
|
20 |
+
This model is a fine-tuned version of [microsoft/swinv2-base-patch4-window8-256](https://huggingface.co/microsoft/swinv2-base-patch4-window8-256) on the isiqa-2019-hf dataset.
|
21 |
It achieves the following results on the evaluation set:
|
22 |
+
- Loss: 0.0254
|
23 |
+
- Srocc: 0.1552
|
24 |
+
- Lcc: 0.2308
|
25 |
|
26 |
## Model description
|
27 |
|
all_results.json
ADDED
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 32.857142857142854,
|
3 |
+
"eval_LCC": 0.23080971037643985,
|
4 |
+
"eval_SROCC": 0.15515006002400963,
|
5 |
+
"eval_loss": 0.025447649881243706,
|
6 |
+
"eval_runtime": 36.1026,
|
7 |
+
"eval_samples_per_second": 1.385,
|
8 |
+
"eval_steps_per_second": 0.055,
|
9 |
+
"total_flos": 2.895570431785304e+18,
|
10 |
+
"train_loss": 0.030544885701459388,
|
11 |
+
"train_runtime": 6073.1055,
|
12 |
+
"train_samples_per_second": 1.762,
|
13 |
+
"train_steps_per_second": 0.025
|
14 |
+
}
|
runs/Aug03_04-21-22_1d636f5ddae9/events.out.tfevents.1722665036.1d636f5ddae9.53.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:6313082c5730528854d232ccd0b33abd070a7579bf7b2c7943f78a2b1e1bb2a5
|
3 |
+
size 448
|
test_results.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 32.857142857142854,
|
3 |
+
"eval_LCC": 0.23080971037643985,
|
4 |
+
"eval_SROCC": 0.15515006002400963,
|
5 |
+
"eval_loss": 0.025447649881243706,
|
6 |
+
"eval_runtime": 36.1026,
|
7 |
+
"eval_samples_per_second": 1.385,
|
8 |
+
"eval_steps_per_second": 0.055
|
9 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,458 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.025447649881243706,
|
3 |
+
"best_model_checkpoint": "/kaggle/working/output/checkpoint-98",
|
4 |
+
"epoch": 32.857142857142854,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 115,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.8571428571428571,
|
13 |
+
"eval_LCC": -0.13879293215538938,
|
14 |
+
"eval_SROCC": -0.15851140456182472,
|
15 |
+
"eval_loss": 0.26562368869781494,
|
16 |
+
"eval_runtime": 38.0061,
|
17 |
+
"eval_samples_per_second": 1.316,
|
18 |
+
"eval_steps_per_second": 0.053,
|
19 |
+
"step": 3
|
20 |
+
},
|
21 |
+
{
|
22 |
+
"epoch": 2.0,
|
23 |
+
"eval_LCC": -0.12769050797605289,
|
24 |
+
"eval_SROCC": -0.18098439375750297,
|
25 |
+
"eval_loss": 0.06580950319766998,
|
26 |
+
"eval_runtime": 35.9745,
|
27 |
+
"eval_samples_per_second": 1.39,
|
28 |
+
"eval_steps_per_second": 0.056,
|
29 |
+
"step": 7
|
30 |
+
},
|
31 |
+
{
|
32 |
+
"epoch": 2.857142857142857,
|
33 |
+
"grad_norm": 6.6160054206848145,
|
34 |
+
"learning_rate": 1.3333333333333333e-05,
|
35 |
+
"loss": 0.2144,
|
36 |
+
"step": 10
|
37 |
+
},
|
38 |
+
{
|
39 |
+
"epoch": 2.857142857142857,
|
40 |
+
"eval_LCC": -0.11849463789687457,
|
41 |
+
"eval_SROCC": -0.16840336134453782,
|
42 |
+
"eval_loss": 0.1341191679239273,
|
43 |
+
"eval_runtime": 35.7755,
|
44 |
+
"eval_samples_per_second": 1.398,
|
45 |
+
"eval_steps_per_second": 0.056,
|
46 |
+
"step": 10
|
47 |
+
},
|
48 |
+
{
|
49 |
+
"epoch": 4.0,
|
50 |
+
"eval_LCC": -0.13792803325263353,
|
51 |
+
"eval_SROCC": -0.23822328931572626,
|
52 |
+
"eval_loss": 0.05623332038521767,
|
53 |
+
"eval_runtime": 36.0015,
|
54 |
+
"eval_samples_per_second": 1.389,
|
55 |
+
"eval_steps_per_second": 0.056,
|
56 |
+
"step": 14
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 4.857142857142857,
|
60 |
+
"eval_LCC": -0.14276549217386758,
|
61 |
+
"eval_SROCC": -0.1657142857142857,
|
62 |
+
"eval_loss": 0.061644963920116425,
|
63 |
+
"eval_runtime": 35.7987,
|
64 |
+
"eval_samples_per_second": 1.397,
|
65 |
+
"eval_steps_per_second": 0.056,
|
66 |
+
"step": 17
|
67 |
+
},
|
68 |
+
{
|
69 |
+
"epoch": 5.714285714285714,
|
70 |
+
"grad_norm": 1.571254014968872,
|
71 |
+
"learning_rate": 1.925925925925926e-05,
|
72 |
+
"loss": 0.0575,
|
73 |
+
"step": 20
|
74 |
+
},
|
75 |
+
{
|
76 |
+
"epoch": 6.0,
|
77 |
+
"eval_LCC": -0.1185043581646345,
|
78 |
+
"eval_SROCC": -0.15006002400960383,
|
79 |
+
"eval_loss": 0.05936155468225479,
|
80 |
+
"eval_runtime": 36.1077,
|
81 |
+
"eval_samples_per_second": 1.385,
|
82 |
+
"eval_steps_per_second": 0.055,
|
83 |
+
"step": 21
|
84 |
+
},
|
85 |
+
{
|
86 |
+
"epoch": 6.857142857142857,
|
87 |
+
"eval_LCC": -0.11221146661280773,
|
88 |
+
"eval_SROCC": -0.12038415366146459,
|
89 |
+
"eval_loss": 0.05129233002662659,
|
90 |
+
"eval_runtime": 35.8427,
|
91 |
+
"eval_samples_per_second": 1.395,
|
92 |
+
"eval_steps_per_second": 0.056,
|
93 |
+
"step": 24
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 8.0,
|
97 |
+
"eval_LCC": -0.09867320013209922,
|
98 |
+
"eval_SROCC": -0.06929171668667466,
|
99 |
+
"eval_loss": 0.041676923632621765,
|
100 |
+
"eval_runtime": 36.0348,
|
101 |
+
"eval_samples_per_second": 1.388,
|
102 |
+
"eval_steps_per_second": 0.056,
|
103 |
+
"step": 28
|
104 |
+
},
|
105 |
+
{
|
106 |
+
"epoch": 8.571428571428571,
|
107 |
+
"grad_norm": 0.800278902053833,
|
108 |
+
"learning_rate": 1.7777777777777777e-05,
|
109 |
+
"loss": 0.0201,
|
110 |
+
"step": 30
|
111 |
+
},
|
112 |
+
{
|
113 |
+
"epoch": 8.857142857142858,
|
114 |
+
"eval_LCC": -0.08130033550198683,
|
115 |
+
"eval_SROCC": -0.07140456182472989,
|
116 |
+
"eval_loss": 0.04058969393372536,
|
117 |
+
"eval_runtime": 36.0799,
|
118 |
+
"eval_samples_per_second": 1.386,
|
119 |
+
"eval_steps_per_second": 0.055,
|
120 |
+
"step": 31
|
121 |
+
},
|
122 |
+
{
|
123 |
+
"epoch": 10.0,
|
124 |
+
"eval_LCC": -0.0384540491268753,
|
125 |
+
"eval_SROCC": -0.02357743097238895,
|
126 |
+
"eval_loss": 0.034295279532670975,
|
127 |
+
"eval_runtime": 36.1741,
|
128 |
+
"eval_samples_per_second": 1.382,
|
129 |
+
"eval_steps_per_second": 0.055,
|
130 |
+
"step": 35
|
131 |
+
},
|
132 |
+
{
|
133 |
+
"epoch": 10.857142857142858,
|
134 |
+
"eval_LCC": 0.0010666390633979736,
|
135 |
+
"eval_SROCC": 0.00744297719087635,
|
136 |
+
"eval_loss": 0.031487837433815,
|
137 |
+
"eval_runtime": 36.0877,
|
138 |
+
"eval_samples_per_second": 1.386,
|
139 |
+
"eval_steps_per_second": 0.055,
|
140 |
+
"step": 38
|
141 |
+
},
|
142 |
+
{
|
143 |
+
"epoch": 11.428571428571429,
|
144 |
+
"grad_norm": 0.3433558940887451,
|
145 |
+
"learning_rate": 1.6296296296296297e-05,
|
146 |
+
"loss": 0.0142,
|
147 |
+
"step": 40
|
148 |
+
},
|
149 |
+
{
|
150 |
+
"epoch": 12.0,
|
151 |
+
"eval_LCC": 0.03256796518969046,
|
152 |
+
"eval_SROCC": 0.02175270108043217,
|
153 |
+
"eval_loss": 0.03087497688829899,
|
154 |
+
"eval_runtime": 35.9382,
|
155 |
+
"eval_samples_per_second": 1.391,
|
156 |
+
"eval_steps_per_second": 0.056,
|
157 |
+
"step": 42
|
158 |
+
},
|
159 |
+
{
|
160 |
+
"epoch": 12.857142857142858,
|
161 |
+
"eval_LCC": 0.04658237209190174,
|
162 |
+
"eval_SROCC": 0.03807923169267707,
|
163 |
+
"eval_loss": 0.030963044613599777,
|
164 |
+
"eval_runtime": 36.1523,
|
165 |
+
"eval_samples_per_second": 1.383,
|
166 |
+
"eval_steps_per_second": 0.055,
|
167 |
+
"step": 45
|
168 |
+
},
|
169 |
+
{
|
170 |
+
"epoch": 14.0,
|
171 |
+
"eval_LCC": 0.06813425647249395,
|
172 |
+
"eval_SROCC": 0.05354141656662665,
|
173 |
+
"eval_loss": 0.029918596148490906,
|
174 |
+
"eval_runtime": 35.716,
|
175 |
+
"eval_samples_per_second": 1.4,
|
176 |
+
"eval_steps_per_second": 0.056,
|
177 |
+
"step": 49
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 14.285714285714286,
|
181 |
+
"grad_norm": 0.8214556574821472,
|
182 |
+
"learning_rate": 1.4814814814814815e-05,
|
183 |
+
"loss": 0.0097,
|
184 |
+
"step": 50
|
185 |
+
},
|
186 |
+
{
|
187 |
+
"epoch": 14.857142857142858,
|
188 |
+
"eval_LCC": 0.07492776273224191,
|
189 |
+
"eval_SROCC": 0.06036014405762304,
|
190 |
+
"eval_loss": 0.03141804039478302,
|
191 |
+
"eval_runtime": 36.0969,
|
192 |
+
"eval_samples_per_second": 1.385,
|
193 |
+
"eval_steps_per_second": 0.055,
|
194 |
+
"step": 52
|
195 |
+
},
|
196 |
+
{
|
197 |
+
"epoch": 16.0,
|
198 |
+
"eval_LCC": 0.10494559746287975,
|
199 |
+
"eval_SROCC": 0.07851140456182472,
|
200 |
+
"eval_loss": 0.02881774678826332,
|
201 |
+
"eval_runtime": 35.9818,
|
202 |
+
"eval_samples_per_second": 1.39,
|
203 |
+
"eval_steps_per_second": 0.056,
|
204 |
+
"step": 56
|
205 |
+
},
|
206 |
+
{
|
207 |
+
"epoch": 16.857142857142858,
|
208 |
+
"eval_LCC": 0.12693485096467225,
|
209 |
+
"eval_SROCC": 0.09435774309723889,
|
210 |
+
"eval_loss": 0.0282765943557024,
|
211 |
+
"eval_runtime": 36.0917,
|
212 |
+
"eval_samples_per_second": 1.385,
|
213 |
+
"eval_steps_per_second": 0.055,
|
214 |
+
"step": 59
|
215 |
+
},
|
216 |
+
{
|
217 |
+
"epoch": 17.142857142857142,
|
218 |
+
"grad_norm": 0.5231106877326965,
|
219 |
+
"learning_rate": 1.3333333333333333e-05,
|
220 |
+
"loss": 0.0083,
|
221 |
+
"step": 60
|
222 |
+
},
|
223 |
+
{
|
224 |
+
"epoch": 18.0,
|
225 |
+
"eval_LCC": 0.14480085260022996,
|
226 |
+
"eval_SROCC": 0.1022328931572629,
|
227 |
+
"eval_loss": 0.0297652930021286,
|
228 |
+
"eval_runtime": 35.9054,
|
229 |
+
"eval_samples_per_second": 1.393,
|
230 |
+
"eval_steps_per_second": 0.056,
|
231 |
+
"step": 63
|
232 |
+
},
|
233 |
+
{
|
234 |
+
"epoch": 18.857142857142858,
|
235 |
+
"eval_LCC": 0.16514532692258366,
|
236 |
+
"eval_SROCC": 0.11183673469387755,
|
237 |
+
"eval_loss": 0.02742738462984562,
|
238 |
+
"eval_runtime": 35.7423,
|
239 |
+
"eval_samples_per_second": 1.399,
|
240 |
+
"eval_steps_per_second": 0.056,
|
241 |
+
"step": 66
|
242 |
+
},
|
243 |
+
{
|
244 |
+
"epoch": 20.0,
|
245 |
+
"grad_norm": 0.43896809220314026,
|
246 |
+
"learning_rate": 1.1851851851851852e-05,
|
247 |
+
"loss": 0.0063,
|
248 |
+
"step": 70
|
249 |
+
},
|
250 |
+
{
|
251 |
+
"epoch": 20.0,
|
252 |
+
"eval_LCC": 0.17032522192761249,
|
253 |
+
"eval_SROCC": 0.12240096038415366,
|
254 |
+
"eval_loss": 0.028603849932551384,
|
255 |
+
"eval_runtime": 36.0386,
|
256 |
+
"eval_samples_per_second": 1.387,
|
257 |
+
"eval_steps_per_second": 0.055,
|
258 |
+
"step": 70
|
259 |
+
},
|
260 |
+
{
|
261 |
+
"epoch": 20.857142857142858,
|
262 |
+
"eval_LCC": 0.1833462608756279,
|
263 |
+
"eval_SROCC": 0.13709483793517407,
|
264 |
+
"eval_loss": 0.028336353600025177,
|
265 |
+
"eval_runtime": 35.9704,
|
266 |
+
"eval_samples_per_second": 1.39,
|
267 |
+
"eval_steps_per_second": 0.056,
|
268 |
+
"step": 73
|
269 |
+
},
|
270 |
+
{
|
271 |
+
"epoch": 22.0,
|
272 |
+
"eval_LCC": 0.19435351727111327,
|
273 |
+
"eval_SROCC": 0.13171668667466988,
|
274 |
+
"eval_loss": 0.028169861063361168,
|
275 |
+
"eval_runtime": 36.261,
|
276 |
+
"eval_samples_per_second": 1.379,
|
277 |
+
"eval_steps_per_second": 0.055,
|
278 |
+
"step": 77
|
279 |
+
},
|
280 |
+
{
|
281 |
+
"epoch": 22.857142857142858,
|
282 |
+
"grad_norm": 0.3503696024417877,
|
283 |
+
"learning_rate": 1.037037037037037e-05,
|
284 |
+
"loss": 0.0059,
|
285 |
+
"step": 80
|
286 |
+
},
|
287 |
+
{
|
288 |
+
"epoch": 22.857142857142858,
|
289 |
+
"eval_LCC": 0.20354883288709705,
|
290 |
+
"eval_SROCC": 0.13815126050420168,
|
291 |
+
"eval_loss": 0.027723778039216995,
|
292 |
+
"eval_runtime": 36.0717,
|
293 |
+
"eval_samples_per_second": 1.386,
|
294 |
+
"eval_steps_per_second": 0.055,
|
295 |
+
"step": 80
|
296 |
+
},
|
297 |
+
{
|
298 |
+
"epoch": 24.0,
|
299 |
+
"eval_LCC": 0.21464098120344227,
|
300 |
+
"eval_SROCC": 0.1479471788715486,
|
301 |
+
"eval_loss": 0.027026496827602386,
|
302 |
+
"eval_runtime": 36.0759,
|
303 |
+
"eval_samples_per_second": 1.386,
|
304 |
+
"eval_steps_per_second": 0.055,
|
305 |
+
"step": 84
|
306 |
+
},
|
307 |
+
{
|
308 |
+
"epoch": 24.857142857142858,
|
309 |
+
"eval_LCC": 0.21974470798595805,
|
310 |
+
"eval_SROCC": 0.1499639855942377,
|
311 |
+
"eval_loss": 0.026259804144501686,
|
312 |
+
"eval_runtime": 35.9282,
|
313 |
+
"eval_samples_per_second": 1.392,
|
314 |
+
"eval_steps_per_second": 0.056,
|
315 |
+
"step": 87
|
316 |
+
},
|
317 |
+
{
|
318 |
+
"epoch": 25.714285714285715,
|
319 |
+
"grad_norm": 0.24314194917678833,
|
320 |
+
"learning_rate": 8.888888888888888e-06,
|
321 |
+
"loss": 0.0046,
|
322 |
+
"step": 90
|
323 |
+
},
|
324 |
+
{
|
325 |
+
"epoch": 26.0,
|
326 |
+
"eval_LCC": 0.21990605203497599,
|
327 |
+
"eval_SROCC": 0.13642256902761105,
|
328 |
+
"eval_loss": 0.026854444295167923,
|
329 |
+
"eval_runtime": 36.0574,
|
330 |
+
"eval_samples_per_second": 1.387,
|
331 |
+
"eval_steps_per_second": 0.055,
|
332 |
+
"step": 91
|
333 |
+
},
|
334 |
+
{
|
335 |
+
"epoch": 26.857142857142858,
|
336 |
+
"eval_LCC": 0.22513542090356625,
|
337 |
+
"eval_SROCC": 0.1406482593037215,
|
338 |
+
"eval_loss": 0.025891508907079697,
|
339 |
+
"eval_runtime": 36.0239,
|
340 |
+
"eval_samples_per_second": 1.388,
|
341 |
+
"eval_steps_per_second": 0.056,
|
342 |
+
"step": 94
|
343 |
+
},
|
344 |
+
{
|
345 |
+
"epoch": 28.0,
|
346 |
+
"eval_LCC": 0.23080971037643985,
|
347 |
+
"eval_SROCC": 0.15515006002400963,
|
348 |
+
"eval_loss": 0.025447649881243706,
|
349 |
+
"eval_runtime": 36.1194,
|
350 |
+
"eval_samples_per_second": 1.384,
|
351 |
+
"eval_steps_per_second": 0.055,
|
352 |
+
"step": 98
|
353 |
+
},
|
354 |
+
{
|
355 |
+
"epoch": 28.571428571428573,
|
356 |
+
"grad_norm": 0.2996827960014343,
|
357 |
+
"learning_rate": 7.4074074074074075e-06,
|
358 |
+
"loss": 0.0039,
|
359 |
+
"step": 100
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 28.857142857142858,
|
363 |
+
"eval_LCC": 0.22614882286734456,
|
364 |
+
"eval_SROCC": 0.14804321728691477,
|
365 |
+
"eval_loss": 0.026662170886993408,
|
366 |
+
"eval_runtime": 35.7837,
|
367 |
+
"eval_samples_per_second": 1.397,
|
368 |
+
"eval_steps_per_second": 0.056,
|
369 |
+
"step": 101
|
370 |
+
},
|
371 |
+
{
|
372 |
+
"epoch": 30.0,
|
373 |
+
"eval_LCC": 0.22468420274739098,
|
374 |
+
"eval_SROCC": 0.14890756302521005,
|
375 |
+
"eval_loss": 0.02701684460043907,
|
376 |
+
"eval_runtime": 35.9924,
|
377 |
+
"eval_samples_per_second": 1.389,
|
378 |
+
"eval_steps_per_second": 0.056,
|
379 |
+
"step": 105
|
380 |
+
},
|
381 |
+
{
|
382 |
+
"epoch": 30.857142857142858,
|
383 |
+
"eval_LCC": 0.2319344817291397,
|
384 |
+
"eval_SROCC": 0.15764705882352942,
|
385 |
+
"eval_loss": 0.026096588000655174,
|
386 |
+
"eval_runtime": 35.7006,
|
387 |
+
"eval_samples_per_second": 1.401,
|
388 |
+
"eval_steps_per_second": 0.056,
|
389 |
+
"step": 108
|
390 |
+
},
|
391 |
+
{
|
392 |
+
"epoch": 31.428571428571427,
|
393 |
+
"grad_norm": 0.31328070163726807,
|
394 |
+
"learning_rate": 5.925925925925926e-06,
|
395 |
+
"loss": 0.0041,
|
396 |
+
"step": 110
|
397 |
+
},
|
398 |
+
{
|
399 |
+
"epoch": 32.0,
|
400 |
+
"eval_LCC": 0.23109019043125076,
|
401 |
+
"eval_SROCC": 0.1630252100840336,
|
402 |
+
"eval_loss": 0.026817049831151962,
|
403 |
+
"eval_runtime": 36.031,
|
404 |
+
"eval_samples_per_second": 1.388,
|
405 |
+
"eval_steps_per_second": 0.056,
|
406 |
+
"step": 112
|
407 |
+
},
|
408 |
+
{
|
409 |
+
"epoch": 32.857142857142854,
|
410 |
+
"eval_LCC": 0.22573394181226267,
|
411 |
+
"eval_SROCC": 0.1569747899159664,
|
412 |
+
"eval_loss": 0.028198465704917908,
|
413 |
+
"eval_runtime": 35.9924,
|
414 |
+
"eval_samples_per_second": 1.389,
|
415 |
+
"eval_steps_per_second": 0.056,
|
416 |
+
"step": 115
|
417 |
+
},
|
418 |
+
{
|
419 |
+
"epoch": 32.857142857142854,
|
420 |
+
"step": 115,
|
421 |
+
"total_flos": 2.895570431785304e+18,
|
422 |
+
"train_loss": 0.030544885701459388,
|
423 |
+
"train_runtime": 6073.1055,
|
424 |
+
"train_samples_per_second": 1.762,
|
425 |
+
"train_steps_per_second": 0.025
|
426 |
+
}
|
427 |
+
],
|
428 |
+
"logging_steps": 10,
|
429 |
+
"max_steps": 150,
|
430 |
+
"num_input_tokens_seen": 0,
|
431 |
+
"num_train_epochs": 50,
|
432 |
+
"save_steps": 500,
|
433 |
+
"stateful_callbacks": {
|
434 |
+
"EarlyStoppingCallback": {
|
435 |
+
"args": {
|
436 |
+
"early_stopping_patience": 5,
|
437 |
+
"early_stopping_threshold": 0.0
|
438 |
+
},
|
439 |
+
"attributes": {
|
440 |
+
"early_stopping_patience_counter": 0
|
441 |
+
}
|
442 |
+
},
|
443 |
+
"TrainerControl": {
|
444 |
+
"args": {
|
445 |
+
"should_epoch_stop": false,
|
446 |
+
"should_evaluate": false,
|
447 |
+
"should_log": false,
|
448 |
+
"should_save": true,
|
449 |
+
"should_training_stop": true
|
450 |
+
},
|
451 |
+
"attributes": {}
|
452 |
+
}
|
453 |
+
},
|
454 |
+
"total_flos": 2.895570431785304e+18,
|
455 |
+
"train_batch_size": 16,
|
456 |
+
"trial_name": null,
|
457 |
+
"trial_params": null
|
458 |
+
}
|
training_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 32.857142857142854,
|
3 |
+
"total_flos": 2.895570431785304e+18,
|
4 |
+
"train_loss": 0.030544885701459388,
|
5 |
+
"train_runtime": 6073.1055,
|
6 |
+
"train_samples_per_second": 1.762,
|
7 |
+
"train_steps_per_second": 0.025
|
8 |
+
}
|