File size: 24,142 Bytes
75710eb |
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 |
2023-10-13 15:16:58,271 ----------------------------------------------------------------------------------------------------
2023-10-13 15:16:58,272 Model: "SequenceTagger(
(embeddings): TransformerWordEmbeddings(
(model): BertModel(
(embeddings): BertEmbeddings(
(word_embeddings): Embedding(32001, 768)
(position_embeddings): Embedding(512, 768)
(token_type_embeddings): Embedding(2, 768)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(encoder): BertEncoder(
(layer): ModuleList(
(0-11): 12 x BertLayer(
(attention): BertAttention(
(self): BertSelfAttention(
(query): Linear(in_features=768, out_features=768, bias=True)
(key): Linear(in_features=768, out_features=768, bias=True)
(value): Linear(in_features=768, out_features=768, bias=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(output): BertSelfOutput(
(dense): Linear(in_features=768, out_features=768, bias=True)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
(intermediate): BertIntermediate(
(dense): Linear(in_features=768, out_features=3072, bias=True)
(intermediate_act_fn): GELUActivation()
)
(output): BertOutput(
(dense): Linear(in_features=3072, out_features=768, bias=True)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
)
(pooler): BertPooler(
(dense): Linear(in_features=768, out_features=768, bias=True)
(activation): Tanh()
)
)
)
(locked_dropout): LockedDropout(p=0.5)
(linear): Linear(in_features=768, out_features=21, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-13 15:16:58,272 ----------------------------------------------------------------------------------------------------
2023-10-13 15:16:58,272 MultiCorpus: 5901 train + 1287 dev + 1505 test sentences
- NER_HIPE_2022 Corpus: 5901 train + 1287 dev + 1505 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/hipe2020/fr/with_doc_seperator
2023-10-13 15:16:58,272 ----------------------------------------------------------------------------------------------------
2023-10-13 15:16:58,272 Train: 5901 sentences
2023-10-13 15:16:58,272 (train_with_dev=False, train_with_test=False)
2023-10-13 15:16:58,272 ----------------------------------------------------------------------------------------------------
2023-10-13 15:16:58,272 Training Params:
2023-10-13 15:16:58,272 - learning_rate: "5e-05"
2023-10-13 15:16:58,272 - mini_batch_size: "8"
2023-10-13 15:16:58,272 - max_epochs: "10"
2023-10-13 15:16:58,272 - shuffle: "True"
2023-10-13 15:16:58,272 ----------------------------------------------------------------------------------------------------
2023-10-13 15:16:58,272 Plugins:
2023-10-13 15:16:58,272 - LinearScheduler | warmup_fraction: '0.1'
2023-10-13 15:16:58,272 ----------------------------------------------------------------------------------------------------
2023-10-13 15:16:58,273 Final evaluation on model from best epoch (best-model.pt)
2023-10-13 15:16:58,273 - metric: "('micro avg', 'f1-score')"
2023-10-13 15:16:58,273 ----------------------------------------------------------------------------------------------------
2023-10-13 15:16:58,273 Computation:
2023-10-13 15:16:58,273 - compute on device: cuda:0
2023-10-13 15:16:58,273 - embedding storage: none
2023-10-13 15:16:58,273 ----------------------------------------------------------------------------------------------------
2023-10-13 15:16:58,273 Model training base path: "hmbench-hipe2020/fr-dbmdz/bert-base-historic-multilingual-cased-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-1"
2023-10-13 15:16:58,273 ----------------------------------------------------------------------------------------------------
2023-10-13 15:16:58,273 ----------------------------------------------------------------------------------------------------
2023-10-13 15:17:03,084 epoch 1 - iter 73/738 - loss 2.74602808 - time (sec): 4.81 - samples/sec: 3466.71 - lr: 0.000005 - momentum: 0.000000
2023-10-13 15:17:07,826 epoch 1 - iter 146/738 - loss 1.71361771 - time (sec): 9.55 - samples/sec: 3449.67 - lr: 0.000010 - momentum: 0.000000
2023-10-13 15:17:13,391 epoch 1 - iter 219/738 - loss 1.24565664 - time (sec): 15.12 - samples/sec: 3430.28 - lr: 0.000015 - momentum: 0.000000
2023-10-13 15:17:17,821 epoch 1 - iter 292/738 - loss 1.03626899 - time (sec): 19.55 - samples/sec: 3433.30 - lr: 0.000020 - momentum: 0.000000
2023-10-13 15:17:22,517 epoch 1 - iter 365/738 - loss 0.89538384 - time (sec): 24.24 - samples/sec: 3426.05 - lr: 0.000025 - momentum: 0.000000
2023-10-13 15:17:27,166 epoch 1 - iter 438/738 - loss 0.79251444 - time (sec): 28.89 - samples/sec: 3409.99 - lr: 0.000030 - momentum: 0.000000
2023-10-13 15:17:31,550 epoch 1 - iter 511/738 - loss 0.71725793 - time (sec): 33.28 - samples/sec: 3417.41 - lr: 0.000035 - momentum: 0.000000
2023-10-13 15:17:36,209 epoch 1 - iter 584/738 - loss 0.65519603 - time (sec): 37.94 - samples/sec: 3398.73 - lr: 0.000039 - momentum: 0.000000
2023-10-13 15:17:41,869 epoch 1 - iter 657/738 - loss 0.59427688 - time (sec): 43.60 - samples/sec: 3399.96 - lr: 0.000044 - momentum: 0.000000
2023-10-13 15:17:46,971 epoch 1 - iter 730/738 - loss 0.55319814 - time (sec): 48.70 - samples/sec: 3385.84 - lr: 0.000049 - momentum: 0.000000
2023-10-13 15:17:47,413 ----------------------------------------------------------------------------------------------------
2023-10-13 15:17:47,413 EPOCH 1 done: loss 0.5489 - lr: 0.000049
2023-10-13 15:17:53,443 DEV : loss 0.14253994822502136 - f1-score (micro avg) 0.7035
2023-10-13 15:17:53,476 saving best model
2023-10-13 15:17:53,841 ----------------------------------------------------------------------------------------------------
2023-10-13 15:17:58,952 epoch 2 - iter 73/738 - loss 0.15502447 - time (sec): 5.11 - samples/sec: 3311.80 - lr: 0.000049 - momentum: 0.000000
2023-10-13 15:18:03,246 epoch 2 - iter 146/738 - loss 0.14085048 - time (sec): 9.40 - samples/sec: 3306.19 - lr: 0.000049 - momentum: 0.000000
2023-10-13 15:18:08,012 epoch 2 - iter 219/738 - loss 0.13869781 - time (sec): 14.17 - samples/sec: 3275.65 - lr: 0.000048 - momentum: 0.000000
2023-10-13 15:18:12,900 epoch 2 - iter 292/738 - loss 0.13704243 - time (sec): 19.06 - samples/sec: 3288.19 - lr: 0.000048 - momentum: 0.000000
2023-10-13 15:18:17,212 epoch 2 - iter 365/738 - loss 0.13342204 - time (sec): 23.37 - samples/sec: 3336.73 - lr: 0.000047 - momentum: 0.000000
2023-10-13 15:18:23,427 epoch 2 - iter 438/738 - loss 0.13259264 - time (sec): 29.58 - samples/sec: 3372.56 - lr: 0.000047 - momentum: 0.000000
2023-10-13 15:18:28,287 epoch 2 - iter 511/738 - loss 0.13044756 - time (sec): 34.44 - samples/sec: 3369.54 - lr: 0.000046 - momentum: 0.000000
2023-10-13 15:18:33,290 epoch 2 - iter 584/738 - loss 0.12937219 - time (sec): 39.45 - samples/sec: 3358.12 - lr: 0.000046 - momentum: 0.000000
2023-10-13 15:18:38,093 epoch 2 - iter 657/738 - loss 0.12659105 - time (sec): 44.25 - samples/sec: 3367.71 - lr: 0.000045 - momentum: 0.000000
2023-10-13 15:18:42,598 epoch 2 - iter 730/738 - loss 0.12317234 - time (sec): 48.76 - samples/sec: 3378.72 - lr: 0.000045 - momentum: 0.000000
2023-10-13 15:18:43,094 ----------------------------------------------------------------------------------------------------
2023-10-13 15:18:43,095 EPOCH 2 done: loss 0.1229 - lr: 0.000045
2023-10-13 15:18:54,371 DEV : loss 0.11215972900390625 - f1-score (micro avg) 0.7691
2023-10-13 15:18:54,406 saving best model
2023-10-13 15:18:54,884 ----------------------------------------------------------------------------------------------------
2023-10-13 15:18:59,693 epoch 3 - iter 73/738 - loss 0.07421766 - time (sec): 4.81 - samples/sec: 3162.11 - lr: 0.000044 - momentum: 0.000000
2023-10-13 15:19:04,307 epoch 3 - iter 146/738 - loss 0.06446355 - time (sec): 9.42 - samples/sec: 3312.82 - lr: 0.000043 - momentum: 0.000000
2023-10-13 15:19:09,071 epoch 3 - iter 219/738 - loss 0.07063195 - time (sec): 14.18 - samples/sec: 3406.91 - lr: 0.000043 - momentum: 0.000000
2023-10-13 15:19:14,191 epoch 3 - iter 292/738 - loss 0.07392124 - time (sec): 19.30 - samples/sec: 3389.06 - lr: 0.000042 - momentum: 0.000000
2023-10-13 15:19:19,271 epoch 3 - iter 365/738 - loss 0.07638465 - time (sec): 24.38 - samples/sec: 3400.77 - lr: 0.000042 - momentum: 0.000000
2023-10-13 15:19:23,948 epoch 3 - iter 438/738 - loss 0.07585899 - time (sec): 29.06 - samples/sec: 3367.56 - lr: 0.000041 - momentum: 0.000000
2023-10-13 15:19:29,191 epoch 3 - iter 511/738 - loss 0.07494694 - time (sec): 34.30 - samples/sec: 3332.95 - lr: 0.000041 - momentum: 0.000000
2023-10-13 15:19:34,643 epoch 3 - iter 584/738 - loss 0.07368203 - time (sec): 39.76 - samples/sec: 3303.06 - lr: 0.000040 - momentum: 0.000000
2023-10-13 15:19:39,452 epoch 3 - iter 657/738 - loss 0.07370863 - time (sec): 44.57 - samples/sec: 3315.19 - lr: 0.000040 - momentum: 0.000000
2023-10-13 15:19:44,937 epoch 3 - iter 730/738 - loss 0.07293039 - time (sec): 50.05 - samples/sec: 3293.55 - lr: 0.000039 - momentum: 0.000000
2023-10-13 15:19:45,384 ----------------------------------------------------------------------------------------------------
2023-10-13 15:19:45,385 EPOCH 3 done: loss 0.0725 - lr: 0.000039
2023-10-13 15:19:56,548 DEV : loss 0.12536022067070007 - f1-score (micro avg) 0.7764
2023-10-13 15:19:56,576 saving best model
2023-10-13 15:19:57,053 ----------------------------------------------------------------------------------------------------
2023-10-13 15:20:01,765 epoch 4 - iter 73/738 - loss 0.04276265 - time (sec): 4.70 - samples/sec: 3350.13 - lr: 0.000038 - momentum: 0.000000
2023-10-13 15:20:06,911 epoch 4 - iter 146/738 - loss 0.05179007 - time (sec): 9.85 - samples/sec: 3393.82 - lr: 0.000038 - momentum: 0.000000
2023-10-13 15:20:12,365 epoch 4 - iter 219/738 - loss 0.04965902 - time (sec): 15.30 - samples/sec: 3402.43 - lr: 0.000037 - momentum: 0.000000
2023-10-13 15:20:17,046 epoch 4 - iter 292/738 - loss 0.04941122 - time (sec): 19.98 - samples/sec: 3373.66 - lr: 0.000037 - momentum: 0.000000
2023-10-13 15:20:21,696 epoch 4 - iter 365/738 - loss 0.04819576 - time (sec): 24.63 - samples/sec: 3367.59 - lr: 0.000036 - momentum: 0.000000
2023-10-13 15:20:26,064 epoch 4 - iter 438/738 - loss 0.04826942 - time (sec): 29.00 - samples/sec: 3362.01 - lr: 0.000036 - momentum: 0.000000
2023-10-13 15:20:31,211 epoch 4 - iter 511/738 - loss 0.04765206 - time (sec): 34.15 - samples/sec: 3370.67 - lr: 0.000035 - momentum: 0.000000
2023-10-13 15:20:35,872 epoch 4 - iter 584/738 - loss 0.04844213 - time (sec): 38.81 - samples/sec: 3360.11 - lr: 0.000035 - momentum: 0.000000
2023-10-13 15:20:41,280 epoch 4 - iter 657/738 - loss 0.04959664 - time (sec): 44.22 - samples/sec: 3353.98 - lr: 0.000034 - momentum: 0.000000
2023-10-13 15:20:46,071 epoch 4 - iter 730/738 - loss 0.04971863 - time (sec): 49.01 - samples/sec: 3365.56 - lr: 0.000033 - momentum: 0.000000
2023-10-13 15:20:46,528 ----------------------------------------------------------------------------------------------------
2023-10-13 15:20:46,528 EPOCH 4 done: loss 0.0497 - lr: 0.000033
2023-10-13 15:20:57,722 DEV : loss 0.1647169589996338 - f1-score (micro avg) 0.8109
2023-10-13 15:20:57,752 saving best model
2023-10-13 15:20:58,245 ----------------------------------------------------------------------------------------------------
2023-10-13 15:21:02,849 epoch 5 - iter 73/738 - loss 0.03178351 - time (sec): 4.59 - samples/sec: 3344.00 - lr: 0.000033 - momentum: 0.000000
2023-10-13 15:21:07,618 epoch 5 - iter 146/738 - loss 0.03011420 - time (sec): 9.36 - samples/sec: 3338.50 - lr: 0.000032 - momentum: 0.000000
2023-10-13 15:21:12,677 epoch 5 - iter 219/738 - loss 0.03759577 - time (sec): 14.42 - samples/sec: 3368.79 - lr: 0.000032 - momentum: 0.000000
2023-10-13 15:21:17,375 epoch 5 - iter 292/738 - loss 0.03522125 - time (sec): 19.12 - samples/sec: 3345.28 - lr: 0.000031 - momentum: 0.000000
2023-10-13 15:21:22,615 epoch 5 - iter 365/738 - loss 0.03386361 - time (sec): 24.36 - samples/sec: 3343.37 - lr: 0.000031 - momentum: 0.000000
2023-10-13 15:21:27,708 epoch 5 - iter 438/738 - loss 0.03481830 - time (sec): 29.45 - samples/sec: 3340.69 - lr: 0.000030 - momentum: 0.000000
2023-10-13 15:21:32,309 epoch 5 - iter 511/738 - loss 0.03493902 - time (sec): 34.05 - samples/sec: 3343.79 - lr: 0.000030 - momentum: 0.000000
2023-10-13 15:21:37,167 epoch 5 - iter 584/738 - loss 0.03490208 - time (sec): 38.91 - samples/sec: 3339.05 - lr: 0.000029 - momentum: 0.000000
2023-10-13 15:21:42,940 epoch 5 - iter 657/738 - loss 0.03505144 - time (sec): 44.68 - samples/sec: 3320.83 - lr: 0.000028 - momentum: 0.000000
2023-10-13 15:21:47,947 epoch 5 - iter 730/738 - loss 0.03538729 - time (sec): 49.69 - samples/sec: 3320.50 - lr: 0.000028 - momentum: 0.000000
2023-10-13 15:21:48,393 ----------------------------------------------------------------------------------------------------
2023-10-13 15:21:48,393 EPOCH 5 done: loss 0.0353 - lr: 0.000028
2023-10-13 15:21:59,731 DEV : loss 0.16715505719184875 - f1-score (micro avg) 0.8129
2023-10-13 15:21:59,762 saving best model
2023-10-13 15:22:00,272 ----------------------------------------------------------------------------------------------------
2023-10-13 15:22:05,043 epoch 6 - iter 73/738 - loss 0.03487724 - time (sec): 4.76 - samples/sec: 3137.17 - lr: 0.000027 - momentum: 0.000000
2023-10-13 15:22:09,824 epoch 6 - iter 146/738 - loss 0.02962974 - time (sec): 9.54 - samples/sec: 3178.81 - lr: 0.000027 - momentum: 0.000000
2023-10-13 15:22:15,490 epoch 6 - iter 219/738 - loss 0.02757779 - time (sec): 15.21 - samples/sec: 3232.17 - lr: 0.000026 - momentum: 0.000000
2023-10-13 15:22:20,455 epoch 6 - iter 292/738 - loss 0.02874868 - time (sec): 20.17 - samples/sec: 3225.44 - lr: 0.000026 - momentum: 0.000000
2023-10-13 15:22:25,165 epoch 6 - iter 365/738 - loss 0.02763650 - time (sec): 24.88 - samples/sec: 3254.83 - lr: 0.000025 - momentum: 0.000000
2023-10-13 15:22:30,551 epoch 6 - iter 438/738 - loss 0.02686535 - time (sec): 30.27 - samples/sec: 3271.81 - lr: 0.000025 - momentum: 0.000000
2023-10-13 15:22:35,096 epoch 6 - iter 511/738 - loss 0.02587544 - time (sec): 34.82 - samples/sec: 3279.27 - lr: 0.000024 - momentum: 0.000000
2023-10-13 15:22:39,857 epoch 6 - iter 584/738 - loss 0.02553302 - time (sec): 39.58 - samples/sec: 3288.34 - lr: 0.000023 - momentum: 0.000000
2023-10-13 15:22:45,402 epoch 6 - iter 657/738 - loss 0.02568474 - time (sec): 45.12 - samples/sec: 3304.11 - lr: 0.000023 - momentum: 0.000000
2023-10-13 15:22:50,134 epoch 6 - iter 730/738 - loss 0.02547133 - time (sec): 49.85 - samples/sec: 3306.39 - lr: 0.000022 - momentum: 0.000000
2023-10-13 15:22:50,602 ----------------------------------------------------------------------------------------------------
2023-10-13 15:22:50,602 EPOCH 6 done: loss 0.0254 - lr: 0.000022
2023-10-13 15:23:03,139 DEV : loss 0.19820576906204224 - f1-score (micro avg) 0.8153
2023-10-13 15:23:03,177 saving best model
2023-10-13 15:23:03,721 ----------------------------------------------------------------------------------------------------
2023-10-13 15:23:08,470 epoch 7 - iter 73/738 - loss 0.01207347 - time (sec): 4.75 - samples/sec: 3203.00 - lr: 0.000022 - momentum: 0.000000
2023-10-13 15:23:14,759 epoch 7 - iter 146/738 - loss 0.01791847 - time (sec): 11.04 - samples/sec: 3047.72 - lr: 0.000021 - momentum: 0.000000
2023-10-13 15:23:19,307 epoch 7 - iter 219/738 - loss 0.01769644 - time (sec): 15.58 - samples/sec: 3125.38 - lr: 0.000021 - momentum: 0.000000
2023-10-13 15:23:24,645 epoch 7 - iter 292/738 - loss 0.01842221 - time (sec): 20.92 - samples/sec: 3096.35 - lr: 0.000020 - momentum: 0.000000
2023-10-13 15:23:29,789 epoch 7 - iter 365/738 - loss 0.01783206 - time (sec): 26.07 - samples/sec: 3137.84 - lr: 0.000020 - momentum: 0.000000
2023-10-13 15:23:34,922 epoch 7 - iter 438/738 - loss 0.01767386 - time (sec): 31.20 - samples/sec: 3200.09 - lr: 0.000019 - momentum: 0.000000
2023-10-13 15:23:40,128 epoch 7 - iter 511/738 - loss 0.01695655 - time (sec): 36.40 - samples/sec: 3224.08 - lr: 0.000018 - momentum: 0.000000
2023-10-13 15:23:45,371 epoch 7 - iter 584/738 - loss 0.01651792 - time (sec): 41.65 - samples/sec: 3222.34 - lr: 0.000018 - momentum: 0.000000
2023-10-13 15:23:49,934 epoch 7 - iter 657/738 - loss 0.01655195 - time (sec): 46.21 - samples/sec: 3223.49 - lr: 0.000017 - momentum: 0.000000
2023-10-13 15:23:54,626 epoch 7 - iter 730/738 - loss 0.01607427 - time (sec): 50.90 - samples/sec: 3232.49 - lr: 0.000017 - momentum: 0.000000
2023-10-13 15:23:55,106 ----------------------------------------------------------------------------------------------------
2023-10-13 15:23:55,106 EPOCH 7 done: loss 0.0162 - lr: 0.000017
2023-10-13 15:24:06,529 DEV : loss 0.20719152688980103 - f1-score (micro avg) 0.8166
2023-10-13 15:24:06,573 saving best model
2023-10-13 15:24:07,354 ----------------------------------------------------------------------------------------------------
2023-10-13 15:24:12,377 epoch 8 - iter 73/738 - loss 0.00864185 - time (sec): 5.02 - samples/sec: 3218.84 - lr: 0.000016 - momentum: 0.000000
2023-10-13 15:24:17,396 epoch 8 - iter 146/738 - loss 0.01007836 - time (sec): 10.04 - samples/sec: 3180.73 - lr: 0.000016 - momentum: 0.000000
2023-10-13 15:24:22,749 epoch 8 - iter 219/738 - loss 0.01079097 - time (sec): 15.39 - samples/sec: 3223.12 - lr: 0.000015 - momentum: 0.000000
2023-10-13 15:24:27,812 epoch 8 - iter 292/738 - loss 0.01209528 - time (sec): 20.46 - samples/sec: 3195.75 - lr: 0.000015 - momentum: 0.000000
2023-10-13 15:24:32,510 epoch 8 - iter 365/738 - loss 0.01368334 - time (sec): 25.15 - samples/sec: 3219.63 - lr: 0.000014 - momentum: 0.000000
2023-10-13 15:24:37,678 epoch 8 - iter 438/738 - loss 0.01409645 - time (sec): 30.32 - samples/sec: 3206.98 - lr: 0.000013 - momentum: 0.000000
2023-10-13 15:24:42,309 epoch 8 - iter 511/738 - loss 0.01413601 - time (sec): 34.95 - samples/sec: 3220.08 - lr: 0.000013 - momentum: 0.000000
2023-10-13 15:24:47,889 epoch 8 - iter 584/738 - loss 0.01464820 - time (sec): 40.53 - samples/sec: 3230.23 - lr: 0.000012 - momentum: 0.000000
2023-10-13 15:24:52,581 epoch 8 - iter 657/738 - loss 0.01388140 - time (sec): 45.22 - samples/sec: 3247.02 - lr: 0.000012 - momentum: 0.000000
2023-10-13 15:24:57,881 epoch 8 - iter 730/738 - loss 0.01314649 - time (sec): 50.52 - samples/sec: 3263.37 - lr: 0.000011 - momentum: 0.000000
2023-10-13 15:24:58,335 ----------------------------------------------------------------------------------------------------
2023-10-13 15:24:58,335 EPOCH 8 done: loss 0.0131 - lr: 0.000011
2023-10-13 15:25:09,552 DEV : loss 0.22207467257976532 - f1-score (micro avg) 0.8239
2023-10-13 15:25:09,587 saving best model
2023-10-13 15:25:10,122 ----------------------------------------------------------------------------------------------------
2023-10-13 15:25:15,072 epoch 9 - iter 73/738 - loss 0.00622634 - time (sec): 4.95 - samples/sec: 3500.95 - lr: 0.000011 - momentum: 0.000000
2023-10-13 15:25:19,870 epoch 9 - iter 146/738 - loss 0.00678073 - time (sec): 9.75 - samples/sec: 3451.45 - lr: 0.000010 - momentum: 0.000000
2023-10-13 15:25:24,740 epoch 9 - iter 219/738 - loss 0.00883955 - time (sec): 14.62 - samples/sec: 3378.38 - lr: 0.000010 - momentum: 0.000000
2023-10-13 15:25:29,740 epoch 9 - iter 292/738 - loss 0.00808768 - time (sec): 19.62 - samples/sec: 3336.72 - lr: 0.000009 - momentum: 0.000000
2023-10-13 15:25:34,678 epoch 9 - iter 365/738 - loss 0.00752856 - time (sec): 24.55 - samples/sec: 3321.98 - lr: 0.000008 - momentum: 0.000000
2023-10-13 15:25:39,270 epoch 9 - iter 438/738 - loss 0.00836252 - time (sec): 29.15 - samples/sec: 3326.86 - lr: 0.000008 - momentum: 0.000000
2023-10-13 15:25:44,160 epoch 9 - iter 511/738 - loss 0.00810242 - time (sec): 34.04 - samples/sec: 3354.34 - lr: 0.000007 - momentum: 0.000000
2023-10-13 15:25:49,564 epoch 9 - iter 584/738 - loss 0.00765179 - time (sec): 39.44 - samples/sec: 3330.39 - lr: 0.000007 - momentum: 0.000000
2023-10-13 15:25:54,489 epoch 9 - iter 657/738 - loss 0.00751168 - time (sec): 44.36 - samples/sec: 3334.19 - lr: 0.000006 - momentum: 0.000000
2023-10-13 15:25:59,344 epoch 9 - iter 730/738 - loss 0.00780373 - time (sec): 49.22 - samples/sec: 3338.80 - lr: 0.000006 - momentum: 0.000000
2023-10-13 15:25:59,980 ----------------------------------------------------------------------------------------------------
2023-10-13 15:25:59,980 EPOCH 9 done: loss 0.0077 - lr: 0.000006
2023-10-13 15:26:11,819 DEV : loss 0.215680792927742 - f1-score (micro avg) 0.8301
2023-10-13 15:26:11,859 saving best model
2023-10-13 15:26:12,442 ----------------------------------------------------------------------------------------------------
2023-10-13 15:26:17,570 epoch 10 - iter 73/738 - loss 0.00455941 - time (sec): 5.12 - samples/sec: 3146.40 - lr: 0.000005 - momentum: 0.000000
2023-10-13 15:26:23,698 epoch 10 - iter 146/738 - loss 0.00464036 - time (sec): 11.25 - samples/sec: 3147.77 - lr: 0.000004 - momentum: 0.000000
2023-10-13 15:26:28,944 epoch 10 - iter 219/738 - loss 0.00476684 - time (sec): 16.50 - samples/sec: 3115.40 - lr: 0.000004 - momentum: 0.000000
2023-10-13 15:26:33,584 epoch 10 - iter 292/738 - loss 0.00428671 - time (sec): 21.14 - samples/sec: 3160.08 - lr: 0.000003 - momentum: 0.000000
2023-10-13 15:26:38,315 epoch 10 - iter 365/738 - loss 0.00478654 - time (sec): 25.87 - samples/sec: 3178.75 - lr: 0.000003 - momentum: 0.000000
2023-10-13 15:26:43,078 epoch 10 - iter 438/738 - loss 0.00517556 - time (sec): 30.63 - samples/sec: 3179.66 - lr: 0.000002 - momentum: 0.000000
2023-10-13 15:26:48,428 epoch 10 - iter 511/738 - loss 0.00501566 - time (sec): 35.98 - samples/sec: 3193.01 - lr: 0.000002 - momentum: 0.000000
2023-10-13 15:26:54,233 epoch 10 - iter 584/738 - loss 0.00534061 - time (sec): 41.79 - samples/sec: 3134.82 - lr: 0.000001 - momentum: 0.000000
2023-10-13 15:26:59,099 epoch 10 - iter 657/738 - loss 0.00508094 - time (sec): 46.65 - samples/sec: 3150.16 - lr: 0.000001 - momentum: 0.000000
2023-10-13 15:27:04,541 epoch 10 - iter 730/738 - loss 0.00505925 - time (sec): 52.10 - samples/sec: 3167.31 - lr: 0.000000 - momentum: 0.000000
2023-10-13 15:27:04,973 ----------------------------------------------------------------------------------------------------
2023-10-13 15:27:04,974 EPOCH 10 done: loss 0.0050 - lr: 0.000000
2023-10-13 15:27:16,194 DEV : loss 0.22398900985717773 - f1-score (micro avg) 0.8267
2023-10-13 15:27:16,617 ----------------------------------------------------------------------------------------------------
2023-10-13 15:27:16,619 Loading model from best epoch ...
2023-10-13 15:27:18,212 SequenceTagger predicts: Dictionary with 21 tags: O, S-loc, B-loc, E-loc, I-loc, S-pers, B-pers, E-pers, I-pers, S-org, B-org, E-org, I-org, S-time, B-time, E-time, I-time, S-prod, B-prod, E-prod, I-prod
2023-10-13 15:27:24,197
Results:
- F-score (micro) 0.7931
- F-score (macro) 0.6926
- Accuracy 0.6833
By class:
precision recall f1-score support
loc 0.8555 0.8765 0.8659 858
pers 0.7491 0.8007 0.7741 537
org 0.5294 0.6136 0.5684 132
time 0.4789 0.6296 0.5440 54
prod 0.7167 0.7049 0.7107 61
micro avg 0.7714 0.8161 0.7931 1642
macro avg 0.6659 0.7251 0.6926 1642
weighted avg 0.7770 0.8161 0.7956 1642
2023-10-13 15:27:24,197 ----------------------------------------------------------------------------------------------------
|