File size: 25,410 Bytes
3443056 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 |
2023-10-11 03:31:48,290 ----------------------------------------------------------------------------------------------------
2023-10-11 03:31:48,292 Model: "SequenceTagger(
(embeddings): ByT5Embeddings(
(model): T5EncoderModel(
(shared): Embedding(384, 1472)
(encoder): T5Stack(
(embed_tokens): Embedding(384, 1472)
(block): ModuleList(
(0): T5Block(
(layer): ModuleList(
(0): T5LayerSelfAttention(
(SelfAttention): T5Attention(
(q): Linear(in_features=1472, out_features=384, bias=False)
(k): Linear(in_features=1472, out_features=384, bias=False)
(v): Linear(in_features=1472, out_features=384, bias=False)
(o): Linear(in_features=384, out_features=1472, bias=False)
(relative_attention_bias): Embedding(32, 6)
)
(layer_norm): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(1): T5LayerFF(
(DenseReluDense): T5DenseGatedActDense(
(wi_0): Linear(in_features=1472, out_features=3584, bias=False)
(wi_1): Linear(in_features=1472, out_features=3584, bias=False)
(wo): Linear(in_features=3584, out_features=1472, bias=False)
(dropout): Dropout(p=0.1, inplace=False)
(act): NewGELUActivation()
)
(layer_norm): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
(1-11): 11 x T5Block(
(layer): ModuleList(
(0): T5LayerSelfAttention(
(SelfAttention): T5Attention(
(q): Linear(in_features=1472, out_features=384, bias=False)
(k): Linear(in_features=1472, out_features=384, bias=False)
(v): Linear(in_features=1472, out_features=384, bias=False)
(o): Linear(in_features=384, out_features=1472, bias=False)
)
(layer_norm): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(1): T5LayerFF(
(DenseReluDense): T5DenseGatedActDense(
(wi_0): Linear(in_features=1472, out_features=3584, bias=False)
(wi_1): Linear(in_features=1472, out_features=3584, bias=False)
(wo): Linear(in_features=3584, out_features=1472, bias=False)
(dropout): Dropout(p=0.1, inplace=False)
(act): NewGELUActivation()
)
(layer_norm): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
)
(final_layer_norm): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
(locked_dropout): LockedDropout(p=0.5)
(linear): Linear(in_features=1472, out_features=17, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-11 03:31:48,292 ----------------------------------------------------------------------------------------------------
2023-10-11 03:31:48,292 MultiCorpus: 7142 train + 698 dev + 2570 test sentences
- NER_HIPE_2022 Corpus: 7142 train + 698 dev + 2570 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/newseye/fr/with_doc_seperator
2023-10-11 03:31:48,293 ----------------------------------------------------------------------------------------------------
2023-10-11 03:31:48,293 Train: 7142 sentences
2023-10-11 03:31:48,293 (train_with_dev=False, train_with_test=False)
2023-10-11 03:31:48,293 ----------------------------------------------------------------------------------------------------
2023-10-11 03:31:48,293 Training Params:
2023-10-11 03:31:48,293 - learning_rate: "0.00016"
2023-10-11 03:31:48,293 - mini_batch_size: "8"
2023-10-11 03:31:48,293 - max_epochs: "10"
2023-10-11 03:31:48,293 - shuffle: "True"
2023-10-11 03:31:48,293 ----------------------------------------------------------------------------------------------------
2023-10-11 03:31:48,293 Plugins:
2023-10-11 03:31:48,293 - TensorboardLogger
2023-10-11 03:31:48,293 - LinearScheduler | warmup_fraction: '0.1'
2023-10-11 03:31:48,293 ----------------------------------------------------------------------------------------------------
2023-10-11 03:31:48,293 Final evaluation on model from best epoch (best-model.pt)
2023-10-11 03:31:48,294 - metric: "('micro avg', 'f1-score')"
2023-10-11 03:31:48,294 ----------------------------------------------------------------------------------------------------
2023-10-11 03:31:48,294 Computation:
2023-10-11 03:31:48,294 - compute on device: cuda:0
2023-10-11 03:31:48,294 - embedding storage: none
2023-10-11 03:31:48,294 ----------------------------------------------------------------------------------------------------
2023-10-11 03:31:48,294 Model training base path: "hmbench-newseye/fr-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs8-wsFalse-e10-lr0.00016-poolingfirst-layers-1-crfFalse-2"
2023-10-11 03:31:48,294 ----------------------------------------------------------------------------------------------------
2023-10-11 03:31:48,294 ----------------------------------------------------------------------------------------------------
2023-10-11 03:31:48,294 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-11 03:32:38,651 epoch 1 - iter 89/893 - loss 2.83712457 - time (sec): 50.36 - samples/sec: 527.35 - lr: 0.000016 - momentum: 0.000000
2023-10-11 03:33:27,129 epoch 1 - iter 178/893 - loss 2.76856474 - time (sec): 98.83 - samples/sec: 523.88 - lr: 0.000032 - momentum: 0.000000
2023-10-11 03:34:17,244 epoch 1 - iter 267/893 - loss 2.56645727 - time (sec): 148.95 - samples/sec: 521.49 - lr: 0.000048 - momentum: 0.000000
2023-10-11 03:35:06,565 epoch 1 - iter 356/893 - loss 2.33448011 - time (sec): 198.27 - samples/sec: 519.23 - lr: 0.000064 - momentum: 0.000000
2023-10-11 03:36:00,091 epoch 1 - iter 445/893 - loss 2.10432695 - time (sec): 251.80 - samples/sec: 505.41 - lr: 0.000080 - momentum: 0.000000
2023-10-11 03:36:51,822 epoch 1 - iter 534/893 - loss 1.88059820 - time (sec): 303.53 - samples/sec: 504.20 - lr: 0.000095 - momentum: 0.000000
2023-10-11 03:37:42,586 epoch 1 - iter 623/893 - loss 1.70794075 - time (sec): 354.29 - samples/sec: 502.17 - lr: 0.000111 - momentum: 0.000000
2023-10-11 03:38:32,514 epoch 1 - iter 712/893 - loss 1.57064750 - time (sec): 404.22 - samples/sec: 497.50 - lr: 0.000127 - momentum: 0.000000
2023-10-11 03:39:22,687 epoch 1 - iter 801/893 - loss 1.45016177 - time (sec): 454.39 - samples/sec: 494.09 - lr: 0.000143 - momentum: 0.000000
2023-10-11 03:40:13,992 epoch 1 - iter 890/893 - loss 1.34843504 - time (sec): 505.70 - samples/sec: 490.92 - lr: 0.000159 - momentum: 0.000000
2023-10-11 03:40:15,430 ----------------------------------------------------------------------------------------------------
2023-10-11 03:40:15,430 EPOCH 1 done: loss 1.3466 - lr: 0.000159
2023-10-11 03:40:35,102 DEV : loss 0.2833699882030487 - f1-score (micro avg) 0.1705
2023-10-11 03:40:35,135 saving best model
2023-10-11 03:40:36,033 ----------------------------------------------------------------------------------------------------
2023-10-11 03:41:27,920 epoch 2 - iter 89/893 - loss 0.31854275 - time (sec): 51.89 - samples/sec: 484.38 - lr: 0.000158 - momentum: 0.000000
2023-10-11 03:42:20,133 epoch 2 - iter 178/893 - loss 0.30105422 - time (sec): 104.10 - samples/sec: 494.94 - lr: 0.000156 - momentum: 0.000000
2023-10-11 03:43:10,984 epoch 2 - iter 267/893 - loss 0.28521306 - time (sec): 154.95 - samples/sec: 492.72 - lr: 0.000155 - momentum: 0.000000
2023-10-11 03:44:01,012 epoch 2 - iter 356/893 - loss 0.26928774 - time (sec): 204.98 - samples/sec: 490.36 - lr: 0.000153 - momentum: 0.000000
2023-10-11 03:44:50,907 epoch 2 - iter 445/893 - loss 0.25483997 - time (sec): 254.87 - samples/sec: 488.75 - lr: 0.000151 - momentum: 0.000000
2023-10-11 03:45:42,828 epoch 2 - iter 534/893 - loss 0.24373078 - time (sec): 306.79 - samples/sec: 483.89 - lr: 0.000149 - momentum: 0.000000
2023-10-11 03:46:32,863 epoch 2 - iter 623/893 - loss 0.23211977 - time (sec): 356.83 - samples/sec: 482.34 - lr: 0.000148 - momentum: 0.000000
2023-10-11 03:47:24,191 epoch 2 - iter 712/893 - loss 0.22071424 - time (sec): 408.16 - samples/sec: 483.45 - lr: 0.000146 - momentum: 0.000000
2023-10-11 03:48:15,045 epoch 2 - iter 801/893 - loss 0.21094737 - time (sec): 459.01 - samples/sec: 486.71 - lr: 0.000144 - momentum: 0.000000
2023-10-11 03:49:04,867 epoch 2 - iter 890/893 - loss 0.20114803 - time (sec): 508.83 - samples/sec: 487.40 - lr: 0.000142 - momentum: 0.000000
2023-10-11 03:49:06,375 ----------------------------------------------------------------------------------------------------
2023-10-11 03:49:06,376 EPOCH 2 done: loss 0.2009 - lr: 0.000142
2023-10-11 03:49:28,304 DEV : loss 0.11237920075654984 - f1-score (micro avg) 0.7166
2023-10-11 03:49:28,336 saving best model
2023-10-11 03:49:30,905 ----------------------------------------------------------------------------------------------------
2023-10-11 03:50:22,507 epoch 3 - iter 89/893 - loss 0.09483268 - time (sec): 51.60 - samples/sec: 500.89 - lr: 0.000140 - momentum: 0.000000
2023-10-11 03:51:12,638 epoch 3 - iter 178/893 - loss 0.09263939 - time (sec): 101.73 - samples/sec: 478.43 - lr: 0.000139 - momentum: 0.000000
2023-10-11 03:52:02,580 epoch 3 - iter 267/893 - loss 0.09132766 - time (sec): 151.67 - samples/sec: 487.98 - lr: 0.000137 - momentum: 0.000000
2023-10-11 03:52:51,470 epoch 3 - iter 356/893 - loss 0.09076148 - time (sec): 200.56 - samples/sec: 489.76 - lr: 0.000135 - momentum: 0.000000
2023-10-11 03:53:42,098 epoch 3 - iter 445/893 - loss 0.08608269 - time (sec): 251.19 - samples/sec: 489.35 - lr: 0.000133 - momentum: 0.000000
2023-10-11 03:54:31,559 epoch 3 - iter 534/893 - loss 0.08616002 - time (sec): 300.65 - samples/sec: 492.76 - lr: 0.000132 - momentum: 0.000000
2023-10-11 03:55:21,855 epoch 3 - iter 623/893 - loss 0.08359586 - time (sec): 350.95 - samples/sec: 498.06 - lr: 0.000130 - momentum: 0.000000
2023-10-11 03:56:11,917 epoch 3 - iter 712/893 - loss 0.08345896 - time (sec): 401.01 - samples/sec: 497.89 - lr: 0.000128 - momentum: 0.000000
2023-10-11 03:57:01,663 epoch 3 - iter 801/893 - loss 0.08264139 - time (sec): 450.75 - samples/sec: 496.55 - lr: 0.000126 - momentum: 0.000000
2023-10-11 03:57:53,429 epoch 3 - iter 890/893 - loss 0.08208483 - time (sec): 502.52 - samples/sec: 493.86 - lr: 0.000125 - momentum: 0.000000
2023-10-11 03:57:54,954 ----------------------------------------------------------------------------------------------------
2023-10-11 03:57:54,954 EPOCH 3 done: loss 0.0822 - lr: 0.000125
2023-10-11 03:58:16,128 DEV : loss 0.09909958392381668 - f1-score (micro avg) 0.7795
2023-10-11 03:58:16,162 saving best model
2023-10-11 03:58:18,810 ----------------------------------------------------------------------------------------------------
2023-10-11 03:59:10,515 epoch 4 - iter 89/893 - loss 0.05729177 - time (sec): 51.70 - samples/sec: 478.76 - lr: 0.000123 - momentum: 0.000000
2023-10-11 04:00:01,270 epoch 4 - iter 178/893 - loss 0.05132805 - time (sec): 102.46 - samples/sec: 466.91 - lr: 0.000121 - momentum: 0.000000
2023-10-11 04:00:53,652 epoch 4 - iter 267/893 - loss 0.05168862 - time (sec): 154.84 - samples/sec: 477.37 - lr: 0.000119 - momentum: 0.000000
2023-10-11 04:01:46,002 epoch 4 - iter 356/893 - loss 0.05215912 - time (sec): 207.19 - samples/sec: 485.67 - lr: 0.000117 - momentum: 0.000000
2023-10-11 04:02:34,693 epoch 4 - iter 445/893 - loss 0.05079508 - time (sec): 255.88 - samples/sec: 483.73 - lr: 0.000116 - momentum: 0.000000
2023-10-11 04:03:24,842 epoch 4 - iter 534/893 - loss 0.05130903 - time (sec): 306.03 - samples/sec: 485.64 - lr: 0.000114 - momentum: 0.000000
2023-10-11 04:04:15,425 epoch 4 - iter 623/893 - loss 0.05198355 - time (sec): 356.61 - samples/sec: 489.38 - lr: 0.000112 - momentum: 0.000000
2023-10-11 04:05:05,416 epoch 4 - iter 712/893 - loss 0.05235379 - time (sec): 406.60 - samples/sec: 488.86 - lr: 0.000110 - momentum: 0.000000
2023-10-11 04:05:55,152 epoch 4 - iter 801/893 - loss 0.05215177 - time (sec): 456.34 - samples/sec: 488.91 - lr: 0.000109 - momentum: 0.000000
2023-10-11 04:06:45,629 epoch 4 - iter 890/893 - loss 0.05242878 - time (sec): 506.82 - samples/sec: 489.86 - lr: 0.000107 - momentum: 0.000000
2023-10-11 04:06:47,035 ----------------------------------------------------------------------------------------------------
2023-10-11 04:06:47,035 EPOCH 4 done: loss 0.0523 - lr: 0.000107
2023-10-11 04:07:09,433 DEV : loss 0.11436811834573746 - f1-score (micro avg) 0.7913
2023-10-11 04:07:09,464 saving best model
2023-10-11 04:07:12,074 ----------------------------------------------------------------------------------------------------
2023-10-11 04:08:03,836 epoch 5 - iter 89/893 - loss 0.04449798 - time (sec): 51.76 - samples/sec: 480.80 - lr: 0.000105 - momentum: 0.000000
2023-10-11 04:08:54,019 epoch 5 - iter 178/893 - loss 0.04209674 - time (sec): 101.94 - samples/sec: 466.28 - lr: 0.000103 - momentum: 0.000000
2023-10-11 04:09:45,794 epoch 5 - iter 267/893 - loss 0.04174679 - time (sec): 153.72 - samples/sec: 470.86 - lr: 0.000101 - momentum: 0.000000
2023-10-11 04:10:36,630 epoch 5 - iter 356/893 - loss 0.04092342 - time (sec): 204.55 - samples/sec: 475.82 - lr: 0.000100 - momentum: 0.000000
2023-10-11 04:11:25,166 epoch 5 - iter 445/893 - loss 0.04167428 - time (sec): 253.09 - samples/sec: 479.16 - lr: 0.000098 - momentum: 0.000000
2023-10-11 04:12:14,988 epoch 5 - iter 534/893 - loss 0.04028609 - time (sec): 302.91 - samples/sec: 482.87 - lr: 0.000096 - momentum: 0.000000
2023-10-11 04:13:04,569 epoch 5 - iter 623/893 - loss 0.04068915 - time (sec): 352.49 - samples/sec: 490.07 - lr: 0.000094 - momentum: 0.000000
2023-10-11 04:13:54,291 epoch 5 - iter 712/893 - loss 0.04043929 - time (sec): 402.21 - samples/sec: 491.30 - lr: 0.000093 - momentum: 0.000000
2023-10-11 04:14:47,153 epoch 5 - iter 801/893 - loss 0.03903387 - time (sec): 455.07 - samples/sec: 488.39 - lr: 0.000091 - momentum: 0.000000
2023-10-11 04:15:37,354 epoch 5 - iter 890/893 - loss 0.03924486 - time (sec): 505.28 - samples/sec: 490.96 - lr: 0.000089 - momentum: 0.000000
2023-10-11 04:15:38,821 ----------------------------------------------------------------------------------------------------
2023-10-11 04:15:38,821 EPOCH 5 done: loss 0.0392 - lr: 0.000089
2023-10-11 04:16:00,816 DEV : loss 0.1378975659608841 - f1-score (micro avg) 0.7957
2023-10-11 04:16:00,847 saving best model
2023-10-11 04:16:03,500 ----------------------------------------------------------------------------------------------------
2023-10-11 04:16:53,271 epoch 6 - iter 89/893 - loss 0.02839597 - time (sec): 49.77 - samples/sec: 497.48 - lr: 0.000087 - momentum: 0.000000
2023-10-11 04:17:43,066 epoch 6 - iter 178/893 - loss 0.02969544 - time (sec): 99.56 - samples/sec: 500.48 - lr: 0.000085 - momentum: 0.000000
2023-10-11 04:18:32,285 epoch 6 - iter 267/893 - loss 0.02762991 - time (sec): 148.78 - samples/sec: 500.47 - lr: 0.000084 - momentum: 0.000000
2023-10-11 04:19:21,556 epoch 6 - iter 356/893 - loss 0.02793419 - time (sec): 198.05 - samples/sec: 501.04 - lr: 0.000082 - momentum: 0.000000
2023-10-11 04:20:11,648 epoch 6 - iter 445/893 - loss 0.02784170 - time (sec): 248.14 - samples/sec: 500.36 - lr: 0.000080 - momentum: 0.000000
2023-10-11 04:21:02,540 epoch 6 - iter 534/893 - loss 0.02833000 - time (sec): 299.04 - samples/sec: 502.29 - lr: 0.000078 - momentum: 0.000000
2023-10-11 04:21:51,592 epoch 6 - iter 623/893 - loss 0.02781878 - time (sec): 348.09 - samples/sec: 501.02 - lr: 0.000077 - momentum: 0.000000
2023-10-11 04:22:41,493 epoch 6 - iter 712/893 - loss 0.02807748 - time (sec): 397.99 - samples/sec: 501.31 - lr: 0.000075 - momentum: 0.000000
2023-10-11 04:23:32,397 epoch 6 - iter 801/893 - loss 0.02889368 - time (sec): 448.89 - samples/sec: 501.88 - lr: 0.000073 - momentum: 0.000000
2023-10-11 04:24:20,319 epoch 6 - iter 890/893 - loss 0.02940820 - time (sec): 496.81 - samples/sec: 499.77 - lr: 0.000071 - momentum: 0.000000
2023-10-11 04:24:21,612 ----------------------------------------------------------------------------------------------------
2023-10-11 04:24:21,612 EPOCH 6 done: loss 0.0294 - lr: 0.000071
2023-10-11 04:24:43,256 DEV : loss 0.1482623666524887 - f1-score (micro avg) 0.8011
2023-10-11 04:24:43,286 saving best model
2023-10-11 04:24:45,866 ----------------------------------------------------------------------------------------------------
2023-10-11 04:25:37,622 epoch 7 - iter 89/893 - loss 0.02216521 - time (sec): 51.75 - samples/sec: 528.67 - lr: 0.000069 - momentum: 0.000000
2023-10-11 04:26:27,082 epoch 7 - iter 178/893 - loss 0.02497855 - time (sec): 101.21 - samples/sec: 513.77 - lr: 0.000068 - momentum: 0.000000
2023-10-11 04:27:15,549 epoch 7 - iter 267/893 - loss 0.02422271 - time (sec): 149.68 - samples/sec: 505.79 - lr: 0.000066 - momentum: 0.000000
2023-10-11 04:28:04,841 epoch 7 - iter 356/893 - loss 0.02369475 - time (sec): 198.97 - samples/sec: 502.47 - lr: 0.000064 - momentum: 0.000000
2023-10-11 04:28:54,382 epoch 7 - iter 445/893 - loss 0.02341763 - time (sec): 248.51 - samples/sec: 500.38 - lr: 0.000062 - momentum: 0.000000
2023-10-11 04:29:44,889 epoch 7 - iter 534/893 - loss 0.02214307 - time (sec): 299.02 - samples/sec: 501.17 - lr: 0.000061 - momentum: 0.000000
2023-10-11 04:30:33,572 epoch 7 - iter 623/893 - loss 0.02180161 - time (sec): 347.70 - samples/sec: 499.52 - lr: 0.000059 - momentum: 0.000000
2023-10-11 04:31:25,741 epoch 7 - iter 712/893 - loss 0.02147087 - time (sec): 399.87 - samples/sec: 500.34 - lr: 0.000057 - momentum: 0.000000
2023-10-11 04:32:17,029 epoch 7 - iter 801/893 - loss 0.02193032 - time (sec): 451.16 - samples/sec: 497.92 - lr: 0.000055 - momentum: 0.000000
2023-10-11 04:33:08,676 epoch 7 - iter 890/893 - loss 0.02228242 - time (sec): 502.81 - samples/sec: 493.32 - lr: 0.000053 - momentum: 0.000000
2023-10-11 04:33:10,184 ----------------------------------------------------------------------------------------------------
2023-10-11 04:33:10,184 EPOCH 7 done: loss 0.0222 - lr: 0.000053
2023-10-11 04:33:31,045 DEV : loss 0.16061536967754364 - f1-score (micro avg) 0.7955
2023-10-11 04:33:31,076 ----------------------------------------------------------------------------------------------------
2023-10-11 04:34:18,933 epoch 8 - iter 89/893 - loss 0.01609981 - time (sec): 47.86 - samples/sec: 503.12 - lr: 0.000052 - momentum: 0.000000
2023-10-11 04:35:08,167 epoch 8 - iter 178/893 - loss 0.01581475 - time (sec): 97.09 - samples/sec: 500.45 - lr: 0.000050 - momentum: 0.000000
2023-10-11 04:35:58,306 epoch 8 - iter 267/893 - loss 0.01509213 - time (sec): 147.23 - samples/sec: 499.65 - lr: 0.000048 - momentum: 0.000000
2023-10-11 04:36:49,077 epoch 8 - iter 356/893 - loss 0.01826549 - time (sec): 198.00 - samples/sec: 501.64 - lr: 0.000046 - momentum: 0.000000
2023-10-11 04:37:38,846 epoch 8 - iter 445/893 - loss 0.01880460 - time (sec): 247.77 - samples/sec: 498.52 - lr: 0.000045 - momentum: 0.000000
2023-10-11 04:38:29,491 epoch 8 - iter 534/893 - loss 0.01951515 - time (sec): 298.41 - samples/sec: 497.04 - lr: 0.000043 - momentum: 0.000000
2023-10-11 04:39:19,883 epoch 8 - iter 623/893 - loss 0.01961892 - time (sec): 348.81 - samples/sec: 495.24 - lr: 0.000041 - momentum: 0.000000
2023-10-11 04:40:11,819 epoch 8 - iter 712/893 - loss 0.01925037 - time (sec): 400.74 - samples/sec: 495.14 - lr: 0.000039 - momentum: 0.000000
2023-10-11 04:41:01,099 epoch 8 - iter 801/893 - loss 0.01911293 - time (sec): 450.02 - samples/sec: 494.92 - lr: 0.000037 - momentum: 0.000000
2023-10-11 04:41:50,784 epoch 8 - iter 890/893 - loss 0.01886592 - time (sec): 499.71 - samples/sec: 495.88 - lr: 0.000036 - momentum: 0.000000
2023-10-11 04:41:52,459 ----------------------------------------------------------------------------------------------------
2023-10-11 04:41:52,459 EPOCH 8 done: loss 0.0189 - lr: 0.000036
2023-10-11 04:42:13,821 DEV : loss 0.17750906944274902 - f1-score (micro avg) 0.8094
2023-10-11 04:42:13,851 saving best model
2023-10-11 04:42:16,513 ----------------------------------------------------------------------------------------------------
2023-10-11 04:43:08,166 epoch 9 - iter 89/893 - loss 0.01085198 - time (sec): 51.65 - samples/sec: 499.86 - lr: 0.000034 - momentum: 0.000000
2023-10-11 04:43:58,320 epoch 9 - iter 178/893 - loss 0.01500659 - time (sec): 101.80 - samples/sec: 493.47 - lr: 0.000032 - momentum: 0.000000
2023-10-11 04:44:49,824 epoch 9 - iter 267/893 - loss 0.01365750 - time (sec): 153.31 - samples/sec: 488.25 - lr: 0.000030 - momentum: 0.000000
2023-10-11 04:45:39,299 epoch 9 - iter 356/893 - loss 0.01461229 - time (sec): 202.78 - samples/sec: 489.26 - lr: 0.000029 - momentum: 0.000000
2023-10-11 04:46:28,453 epoch 9 - iter 445/893 - loss 0.01350951 - time (sec): 251.94 - samples/sec: 491.86 - lr: 0.000027 - momentum: 0.000000
2023-10-11 04:47:17,606 epoch 9 - iter 534/893 - loss 0.01395459 - time (sec): 301.09 - samples/sec: 493.39 - lr: 0.000025 - momentum: 0.000000
2023-10-11 04:48:05,086 epoch 9 - iter 623/893 - loss 0.01396505 - time (sec): 348.57 - samples/sec: 492.02 - lr: 0.000023 - momentum: 0.000000
2023-10-11 04:48:54,590 epoch 9 - iter 712/893 - loss 0.01473988 - time (sec): 398.07 - samples/sec: 494.86 - lr: 0.000022 - momentum: 0.000000
2023-10-11 04:49:44,244 epoch 9 - iter 801/893 - loss 0.01491773 - time (sec): 447.73 - samples/sec: 498.75 - lr: 0.000020 - momentum: 0.000000
2023-10-11 04:50:32,777 epoch 9 - iter 890/893 - loss 0.01457607 - time (sec): 496.26 - samples/sec: 499.51 - lr: 0.000018 - momentum: 0.000000
2023-10-11 04:50:34,391 ----------------------------------------------------------------------------------------------------
2023-10-11 04:50:34,391 EPOCH 9 done: loss 0.0146 - lr: 0.000018
2023-10-11 04:50:55,729 DEV : loss 0.18004417419433594 - f1-score (micro avg) 0.8005
2023-10-11 04:50:55,761 ----------------------------------------------------------------------------------------------------
2023-10-11 04:51:43,006 epoch 10 - iter 89/893 - loss 0.01256475 - time (sec): 47.24 - samples/sec: 497.36 - lr: 0.000016 - momentum: 0.000000
2023-10-11 04:52:33,364 epoch 10 - iter 178/893 - loss 0.01131584 - time (sec): 97.60 - samples/sec: 495.44 - lr: 0.000014 - momentum: 0.000000
2023-10-11 04:53:24,321 epoch 10 - iter 267/893 - loss 0.01086775 - time (sec): 148.56 - samples/sec: 497.67 - lr: 0.000013 - momentum: 0.000000
2023-10-11 04:54:13,799 epoch 10 - iter 356/893 - loss 0.01031103 - time (sec): 198.04 - samples/sec: 493.94 - lr: 0.000011 - momentum: 0.000000
2023-10-11 04:55:04,876 epoch 10 - iter 445/893 - loss 0.01121298 - time (sec): 249.11 - samples/sec: 497.61 - lr: 0.000009 - momentum: 0.000000
2023-10-11 04:55:56,369 epoch 10 - iter 534/893 - loss 0.01191879 - time (sec): 300.61 - samples/sec: 501.27 - lr: 0.000007 - momentum: 0.000000
2023-10-11 04:56:45,593 epoch 10 - iter 623/893 - loss 0.01245382 - time (sec): 349.83 - samples/sec: 496.96 - lr: 0.000006 - momentum: 0.000000
2023-10-11 04:57:34,479 epoch 10 - iter 712/893 - loss 0.01244378 - time (sec): 398.72 - samples/sec: 499.96 - lr: 0.000004 - momentum: 0.000000
2023-10-11 04:58:23,033 epoch 10 - iter 801/893 - loss 0.01246859 - time (sec): 447.27 - samples/sec: 500.35 - lr: 0.000002 - momentum: 0.000000
2023-10-11 04:59:10,669 epoch 10 - iter 890/893 - loss 0.01218169 - time (sec): 494.91 - samples/sec: 501.30 - lr: 0.000000 - momentum: 0.000000
2023-10-11 04:59:12,200 ----------------------------------------------------------------------------------------------------
2023-10-11 04:59:12,200 EPOCH 10 done: loss 0.0122 - lr: 0.000000
2023-10-11 04:59:34,117 DEV : loss 0.185577392578125 - f1-score (micro avg) 0.8069
2023-10-11 04:59:35,005 ----------------------------------------------------------------------------------------------------
2023-10-11 04:59:35,007 Loading model from best epoch ...
2023-10-11 04:59:39,747 SequenceTagger predicts: Dictionary with 17 tags: O, S-PER, B-PER, E-PER, I-PER, S-LOC, B-LOC, E-LOC, I-LOC, S-ORG, B-ORG, E-ORG, I-ORG, S-HumanProd, B-HumanProd, E-HumanProd, I-HumanProd
2023-10-11 05:00:48,937
Results:
- F-score (micro) 0.69
- F-score (macro) 0.6101
- Accuracy 0.5402
By class:
precision recall f1-score support
LOC 0.6901 0.6913 0.6907 1095
PER 0.7741 0.7856 0.7798 1012
ORG 0.4310 0.5602 0.4872 357
HumanProd 0.3889 0.6364 0.4828 33
micro avg 0.6711 0.7101 0.6900 2497
macro avg 0.5710 0.6684 0.6101 2497
weighted avg 0.6831 0.7101 0.6950 2497
2023-10-11 05:00:48,937 ----------------------------------------------------------------------------------------------------
|