File size: 24,027 Bytes
8589b76 |
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 |
2023-10-16 22:55:24,117 ----------------------------------------------------------------------------------------------------
2023-10-16 22:55:24,118 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=13, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-16 22:55:24,118 ----------------------------------------------------------------------------------------------------
2023-10-16 22:55:24,118 MultiCorpus: 6183 train + 680 dev + 2113 test sentences
- NER_HIPE_2022 Corpus: 6183 train + 680 dev + 2113 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/topres19th/en/with_doc_seperator
2023-10-16 22:55:24,118 ----------------------------------------------------------------------------------------------------
2023-10-16 22:55:24,118 Train: 6183 sentences
2023-10-16 22:55:24,118 (train_with_dev=False, train_with_test=False)
2023-10-16 22:55:24,118 ----------------------------------------------------------------------------------------------------
2023-10-16 22:55:24,118 Training Params:
2023-10-16 22:55:24,118 - learning_rate: "3e-05"
2023-10-16 22:55:24,118 - mini_batch_size: "4"
2023-10-16 22:55:24,119 - max_epochs: "10"
2023-10-16 22:55:24,119 - shuffle: "True"
2023-10-16 22:55:24,119 ----------------------------------------------------------------------------------------------------
2023-10-16 22:55:24,119 Plugins:
2023-10-16 22:55:24,119 - LinearScheduler | warmup_fraction: '0.1'
2023-10-16 22:55:24,119 ----------------------------------------------------------------------------------------------------
2023-10-16 22:55:24,119 Final evaluation on model from best epoch (best-model.pt)
2023-10-16 22:55:24,119 - metric: "('micro avg', 'f1-score')"
2023-10-16 22:55:24,119 ----------------------------------------------------------------------------------------------------
2023-10-16 22:55:24,119 Computation:
2023-10-16 22:55:24,119 - compute on device: cuda:0
2023-10-16 22:55:24,119 - embedding storage: none
2023-10-16 22:55:24,119 ----------------------------------------------------------------------------------------------------
2023-10-16 22:55:24,119 Model training base path: "hmbench-topres19th/en-dbmdz/bert-base-historic-multilingual-cased-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-4"
2023-10-16 22:55:24,119 ----------------------------------------------------------------------------------------------------
2023-10-16 22:55:24,119 ----------------------------------------------------------------------------------------------------
2023-10-16 22:55:30,952 epoch 1 - iter 154/1546 - loss 2.02538090 - time (sec): 6.83 - samples/sec: 1748.19 - lr: 0.000003 - momentum: 0.000000
2023-10-16 22:55:37,944 epoch 1 - iter 308/1546 - loss 1.10084678 - time (sec): 13.82 - samples/sec: 1753.65 - lr: 0.000006 - momentum: 0.000000
2023-10-16 22:55:45,034 epoch 1 - iter 462/1546 - loss 0.78647526 - time (sec): 20.91 - samples/sec: 1743.25 - lr: 0.000009 - momentum: 0.000000
2023-10-16 22:55:51,980 epoch 1 - iter 616/1546 - loss 0.62446099 - time (sec): 27.86 - samples/sec: 1741.14 - lr: 0.000012 - momentum: 0.000000
2023-10-16 22:55:58,942 epoch 1 - iter 770/1546 - loss 0.52442561 - time (sec): 34.82 - samples/sec: 1740.59 - lr: 0.000015 - momentum: 0.000000
2023-10-16 22:56:06,056 epoch 1 - iter 924/1546 - loss 0.45595178 - time (sec): 41.94 - samples/sec: 1747.95 - lr: 0.000018 - momentum: 0.000000
2023-10-16 22:56:12,979 epoch 1 - iter 1078/1546 - loss 0.40906220 - time (sec): 48.86 - samples/sec: 1748.23 - lr: 0.000021 - momentum: 0.000000
2023-10-16 22:56:19,931 epoch 1 - iter 1232/1546 - loss 0.36522487 - time (sec): 55.81 - samples/sec: 1782.15 - lr: 0.000024 - momentum: 0.000000
2023-10-16 22:56:26,757 epoch 1 - iter 1386/1546 - loss 0.33735166 - time (sec): 62.64 - samples/sec: 1778.01 - lr: 0.000027 - momentum: 0.000000
2023-10-16 22:56:33,780 epoch 1 - iter 1540/1546 - loss 0.31510388 - time (sec): 69.66 - samples/sec: 1778.34 - lr: 0.000030 - momentum: 0.000000
2023-10-16 22:56:34,038 ----------------------------------------------------------------------------------------------------
2023-10-16 22:56:34,038 EPOCH 1 done: loss 0.3141 - lr: 0.000030
2023-10-16 22:56:35,849 DEV : loss 0.09308421611785889 - f1-score (micro avg) 0.6875
2023-10-16 22:56:35,862 saving best model
2023-10-16 22:56:36,248 ----------------------------------------------------------------------------------------------------
2023-10-16 22:56:43,180 epoch 2 - iter 154/1546 - loss 0.08398905 - time (sec): 6.93 - samples/sec: 1878.88 - lr: 0.000030 - momentum: 0.000000
2023-10-16 22:56:50,219 epoch 2 - iter 308/1546 - loss 0.09542073 - time (sec): 13.97 - samples/sec: 1876.03 - lr: 0.000029 - momentum: 0.000000
2023-10-16 22:56:57,201 epoch 2 - iter 462/1546 - loss 0.08471087 - time (sec): 20.95 - samples/sec: 1892.41 - lr: 0.000029 - momentum: 0.000000
2023-10-16 22:57:04,205 epoch 2 - iter 616/1546 - loss 0.08071299 - time (sec): 27.95 - samples/sec: 1880.42 - lr: 0.000029 - momentum: 0.000000
2023-10-16 22:57:11,180 epoch 2 - iter 770/1546 - loss 0.08054955 - time (sec): 34.93 - samples/sec: 1852.22 - lr: 0.000028 - momentum: 0.000000
2023-10-16 22:57:18,042 epoch 2 - iter 924/1546 - loss 0.08042291 - time (sec): 41.79 - samples/sec: 1822.50 - lr: 0.000028 - momentum: 0.000000
2023-10-16 22:57:24,993 epoch 2 - iter 1078/1546 - loss 0.07999482 - time (sec): 48.74 - samples/sec: 1795.97 - lr: 0.000028 - momentum: 0.000000
2023-10-16 22:57:31,929 epoch 2 - iter 1232/1546 - loss 0.08098848 - time (sec): 55.68 - samples/sec: 1795.29 - lr: 0.000027 - momentum: 0.000000
2023-10-16 22:57:38,721 epoch 2 - iter 1386/1546 - loss 0.08104893 - time (sec): 62.47 - samples/sec: 1792.03 - lr: 0.000027 - momentum: 0.000000
2023-10-16 22:57:45,660 epoch 2 - iter 1540/1546 - loss 0.08159740 - time (sec): 69.41 - samples/sec: 1785.43 - lr: 0.000027 - momentum: 0.000000
2023-10-16 22:57:45,929 ----------------------------------------------------------------------------------------------------
2023-10-16 22:57:45,930 EPOCH 2 done: loss 0.0815 - lr: 0.000027
2023-10-16 22:57:48,443 DEV : loss 0.07893380522727966 - f1-score (micro avg) 0.7179
2023-10-16 22:57:48,456 saving best model
2023-10-16 22:57:48,916 ----------------------------------------------------------------------------------------------------
2023-10-16 22:57:55,989 epoch 3 - iter 154/1546 - loss 0.05779724 - time (sec): 7.07 - samples/sec: 1866.59 - lr: 0.000026 - momentum: 0.000000
2023-10-16 22:58:02,957 epoch 3 - iter 308/1546 - loss 0.05329053 - time (sec): 14.04 - samples/sec: 1877.95 - lr: 0.000026 - momentum: 0.000000
2023-10-16 22:58:09,919 epoch 3 - iter 462/1546 - loss 0.05290610 - time (sec): 21.00 - samples/sec: 1827.68 - lr: 0.000026 - momentum: 0.000000
2023-10-16 22:58:16,734 epoch 3 - iter 616/1546 - loss 0.05251651 - time (sec): 27.82 - samples/sec: 1852.75 - lr: 0.000025 - momentum: 0.000000
2023-10-16 22:58:23,802 epoch 3 - iter 770/1546 - loss 0.05620252 - time (sec): 34.88 - samples/sec: 1830.94 - lr: 0.000025 - momentum: 0.000000
2023-10-16 22:58:30,728 epoch 3 - iter 924/1546 - loss 0.05327156 - time (sec): 41.81 - samples/sec: 1810.74 - lr: 0.000025 - momentum: 0.000000
2023-10-16 22:58:37,749 epoch 3 - iter 1078/1546 - loss 0.05179250 - time (sec): 48.83 - samples/sec: 1801.70 - lr: 0.000024 - momentum: 0.000000
2023-10-16 22:58:44,683 epoch 3 - iter 1232/1546 - loss 0.05111602 - time (sec): 55.76 - samples/sec: 1787.00 - lr: 0.000024 - momentum: 0.000000
2023-10-16 22:58:51,659 epoch 3 - iter 1386/1546 - loss 0.05503167 - time (sec): 62.74 - samples/sec: 1784.02 - lr: 0.000024 - momentum: 0.000000
2023-10-16 22:58:58,674 epoch 3 - iter 1540/1546 - loss 0.05356182 - time (sec): 69.76 - samples/sec: 1775.63 - lr: 0.000023 - momentum: 0.000000
2023-10-16 22:58:58,969 ----------------------------------------------------------------------------------------------------
2023-10-16 22:58:58,969 EPOCH 3 done: loss 0.0534 - lr: 0.000023
2023-10-16 22:59:01,104 DEV : loss 0.07748623192310333 - f1-score (micro avg) 0.7722
2023-10-16 22:59:01,123 saving best model
2023-10-16 22:59:01,558 ----------------------------------------------------------------------------------------------------
2023-10-16 22:59:08,656 epoch 4 - iter 154/1546 - loss 0.02829162 - time (sec): 7.10 - samples/sec: 1872.64 - lr: 0.000023 - momentum: 0.000000
2023-10-16 22:59:15,863 epoch 4 - iter 308/1546 - loss 0.03022453 - time (sec): 14.30 - samples/sec: 1765.31 - lr: 0.000023 - momentum: 0.000000
2023-10-16 22:59:23,032 epoch 4 - iter 462/1546 - loss 0.03061131 - time (sec): 21.47 - samples/sec: 1784.48 - lr: 0.000022 - momentum: 0.000000
2023-10-16 22:59:29,960 epoch 4 - iter 616/1546 - loss 0.03200286 - time (sec): 28.40 - samples/sec: 1768.39 - lr: 0.000022 - momentum: 0.000000
2023-10-16 22:59:36,962 epoch 4 - iter 770/1546 - loss 0.03054981 - time (sec): 35.40 - samples/sec: 1754.04 - lr: 0.000022 - momentum: 0.000000
2023-10-16 22:59:44,121 epoch 4 - iter 924/1546 - loss 0.03188868 - time (sec): 42.56 - samples/sec: 1770.99 - lr: 0.000021 - momentum: 0.000000
2023-10-16 22:59:51,180 epoch 4 - iter 1078/1546 - loss 0.03290334 - time (sec): 49.62 - samples/sec: 1762.82 - lr: 0.000021 - momentum: 0.000000
2023-10-16 22:59:58,373 epoch 4 - iter 1232/1546 - loss 0.03474654 - time (sec): 56.81 - samples/sec: 1764.27 - lr: 0.000021 - momentum: 0.000000
2023-10-16 23:00:05,304 epoch 4 - iter 1386/1546 - loss 0.03409150 - time (sec): 63.74 - samples/sec: 1761.62 - lr: 0.000020 - momentum: 0.000000
2023-10-16 23:00:12,109 epoch 4 - iter 1540/1546 - loss 0.03479952 - time (sec): 70.55 - samples/sec: 1757.31 - lr: 0.000020 - momentum: 0.000000
2023-10-16 23:00:12,360 ----------------------------------------------------------------------------------------------------
2023-10-16 23:00:12,360 EPOCH 4 done: loss 0.0349 - lr: 0.000020
2023-10-16 23:00:14,474 DEV : loss 0.096384696662426 - f1-score (micro avg) 0.7439
2023-10-16 23:00:14,487 ----------------------------------------------------------------------------------------------------
2023-10-16 23:00:21,420 epoch 5 - iter 154/1546 - loss 0.02586135 - time (sec): 6.93 - samples/sec: 1708.87 - lr: 0.000020 - momentum: 0.000000
2023-10-16 23:00:28,521 epoch 5 - iter 308/1546 - loss 0.02697778 - time (sec): 14.03 - samples/sec: 1731.12 - lr: 0.000019 - momentum: 0.000000
2023-10-16 23:00:35,515 epoch 5 - iter 462/1546 - loss 0.02633540 - time (sec): 21.03 - samples/sec: 1744.38 - lr: 0.000019 - momentum: 0.000000
2023-10-16 23:00:42,412 epoch 5 - iter 616/1546 - loss 0.02506817 - time (sec): 27.92 - samples/sec: 1769.86 - lr: 0.000019 - momentum: 0.000000
2023-10-16 23:00:49,241 epoch 5 - iter 770/1546 - loss 0.02333240 - time (sec): 34.75 - samples/sec: 1771.95 - lr: 0.000018 - momentum: 0.000000
2023-10-16 23:00:56,162 epoch 5 - iter 924/1546 - loss 0.02299940 - time (sec): 41.67 - samples/sec: 1782.02 - lr: 0.000018 - momentum: 0.000000
2023-10-16 23:01:03,068 epoch 5 - iter 1078/1546 - loss 0.02391515 - time (sec): 48.58 - samples/sec: 1787.02 - lr: 0.000018 - momentum: 0.000000
2023-10-16 23:01:09,932 epoch 5 - iter 1232/1546 - loss 0.02530811 - time (sec): 55.44 - samples/sec: 1782.95 - lr: 0.000017 - momentum: 0.000000
2023-10-16 23:01:16,872 epoch 5 - iter 1386/1546 - loss 0.02533599 - time (sec): 62.38 - samples/sec: 1780.89 - lr: 0.000017 - momentum: 0.000000
2023-10-16 23:01:23,709 epoch 5 - iter 1540/1546 - loss 0.02481519 - time (sec): 69.22 - samples/sec: 1791.49 - lr: 0.000017 - momentum: 0.000000
2023-10-16 23:01:23,966 ----------------------------------------------------------------------------------------------------
2023-10-16 23:01:23,966 EPOCH 5 done: loss 0.0248 - lr: 0.000017
2023-10-16 23:01:26,008 DEV : loss 0.09776511788368225 - f1-score (micro avg) 0.7849
2023-10-16 23:01:26,022 saving best model
2023-10-16 23:01:26,478 ----------------------------------------------------------------------------------------------------
2023-10-16 23:01:33,699 epoch 6 - iter 154/1546 - loss 0.01891335 - time (sec): 7.22 - samples/sec: 1656.83 - lr: 0.000016 - momentum: 0.000000
2023-10-16 23:01:40,520 epoch 6 - iter 308/1546 - loss 0.01505548 - time (sec): 14.04 - samples/sec: 1735.36 - lr: 0.000016 - momentum: 0.000000
2023-10-16 23:01:47,391 epoch 6 - iter 462/1546 - loss 0.01687048 - time (sec): 20.91 - samples/sec: 1770.14 - lr: 0.000016 - momentum: 0.000000
2023-10-16 23:01:54,201 epoch 6 - iter 616/1546 - loss 0.01676100 - time (sec): 27.72 - samples/sec: 1795.31 - lr: 0.000015 - momentum: 0.000000
2023-10-16 23:02:01,043 epoch 6 - iter 770/1546 - loss 0.01743799 - time (sec): 34.56 - samples/sec: 1783.52 - lr: 0.000015 - momentum: 0.000000
2023-10-16 23:02:07,867 epoch 6 - iter 924/1546 - loss 0.01719557 - time (sec): 41.39 - samples/sec: 1786.16 - lr: 0.000015 - momentum: 0.000000
2023-10-16 23:02:14,832 epoch 6 - iter 1078/1546 - loss 0.01686034 - time (sec): 48.35 - samples/sec: 1810.74 - lr: 0.000014 - momentum: 0.000000
2023-10-16 23:02:21,723 epoch 6 - iter 1232/1546 - loss 0.01742908 - time (sec): 55.24 - samples/sec: 1811.68 - lr: 0.000014 - momentum: 0.000000
2023-10-16 23:02:28,514 epoch 6 - iter 1386/1546 - loss 0.01704174 - time (sec): 62.03 - samples/sec: 1803.99 - lr: 0.000014 - momentum: 0.000000
2023-10-16 23:02:35,311 epoch 6 - iter 1540/1546 - loss 0.01782016 - time (sec): 68.83 - samples/sec: 1797.96 - lr: 0.000013 - momentum: 0.000000
2023-10-16 23:02:35,582 ----------------------------------------------------------------------------------------------------
2023-10-16 23:02:35,582 EPOCH 6 done: loss 0.0177 - lr: 0.000013
2023-10-16 23:02:37,681 DEV : loss 0.10087885707616806 - f1-score (micro avg) 0.7967
2023-10-16 23:02:37,694 saving best model
2023-10-16 23:02:38,137 ----------------------------------------------------------------------------------------------------
2023-10-16 23:02:45,059 epoch 7 - iter 154/1546 - loss 0.00882572 - time (sec): 6.92 - samples/sec: 1801.10 - lr: 0.000013 - momentum: 0.000000
2023-10-16 23:02:51,970 epoch 7 - iter 308/1546 - loss 0.00959269 - time (sec): 13.83 - samples/sec: 1837.87 - lr: 0.000013 - momentum: 0.000000
2023-10-16 23:02:58,728 epoch 7 - iter 462/1546 - loss 0.01074153 - time (sec): 20.59 - samples/sec: 1826.45 - lr: 0.000012 - momentum: 0.000000
2023-10-16 23:03:05,614 epoch 7 - iter 616/1546 - loss 0.01103381 - time (sec): 27.48 - samples/sec: 1815.78 - lr: 0.000012 - momentum: 0.000000
2023-10-16 23:03:12,330 epoch 7 - iter 770/1546 - loss 0.01089398 - time (sec): 34.19 - samples/sec: 1795.14 - lr: 0.000012 - momentum: 0.000000
2023-10-16 23:03:19,062 epoch 7 - iter 924/1546 - loss 0.01051405 - time (sec): 40.92 - samples/sec: 1791.40 - lr: 0.000011 - momentum: 0.000000
2023-10-16 23:03:25,860 epoch 7 - iter 1078/1546 - loss 0.01078117 - time (sec): 47.72 - samples/sec: 1789.95 - lr: 0.000011 - momentum: 0.000000
2023-10-16 23:03:32,683 epoch 7 - iter 1232/1546 - loss 0.01046656 - time (sec): 54.55 - samples/sec: 1783.28 - lr: 0.000011 - momentum: 0.000000
2023-10-16 23:03:39,483 epoch 7 - iter 1386/1546 - loss 0.01062877 - time (sec): 61.34 - samples/sec: 1781.78 - lr: 0.000010 - momentum: 0.000000
2023-10-16 23:03:46,278 epoch 7 - iter 1540/1546 - loss 0.01097604 - time (sec): 68.14 - samples/sec: 1816.28 - lr: 0.000010 - momentum: 0.000000
2023-10-16 23:03:46,535 ----------------------------------------------------------------------------------------------------
2023-10-16 23:03:46,535 EPOCH 7 done: loss 0.0109 - lr: 0.000010
2023-10-16 23:03:48,934 DEV : loss 0.1023801639676094 - f1-score (micro avg) 0.7807
2023-10-16 23:03:48,947 ----------------------------------------------------------------------------------------------------
2023-10-16 23:03:55,722 epoch 8 - iter 154/1546 - loss 0.00276108 - time (sec): 6.77 - samples/sec: 1772.81 - lr: 0.000010 - momentum: 0.000000
2023-10-16 23:04:02,248 epoch 8 - iter 308/1546 - loss 0.00679671 - time (sec): 13.30 - samples/sec: 1827.75 - lr: 0.000009 - momentum: 0.000000
2023-10-16 23:04:08,788 epoch 8 - iter 462/1546 - loss 0.00528715 - time (sec): 19.84 - samples/sec: 1852.41 - lr: 0.000009 - momentum: 0.000000
2023-10-16 23:04:15,408 epoch 8 - iter 616/1546 - loss 0.00664770 - time (sec): 26.46 - samples/sec: 1867.03 - lr: 0.000009 - momentum: 0.000000
2023-10-16 23:04:22,270 epoch 8 - iter 770/1546 - loss 0.00694105 - time (sec): 33.32 - samples/sec: 1846.72 - lr: 0.000008 - momentum: 0.000000
2023-10-16 23:04:29,131 epoch 8 - iter 924/1546 - loss 0.00772621 - time (sec): 40.18 - samples/sec: 1837.91 - lr: 0.000008 - momentum: 0.000000
2023-10-16 23:04:35,982 epoch 8 - iter 1078/1546 - loss 0.00795917 - time (sec): 47.03 - samples/sec: 1842.25 - lr: 0.000008 - momentum: 0.000000
2023-10-16 23:04:42,806 epoch 8 - iter 1232/1546 - loss 0.00817950 - time (sec): 53.86 - samples/sec: 1845.29 - lr: 0.000007 - momentum: 0.000000
2023-10-16 23:04:49,669 epoch 8 - iter 1386/1546 - loss 0.00800634 - time (sec): 60.72 - samples/sec: 1838.42 - lr: 0.000007 - momentum: 0.000000
2023-10-16 23:04:56,447 epoch 8 - iter 1540/1546 - loss 0.00792214 - time (sec): 67.50 - samples/sec: 1834.35 - lr: 0.000007 - momentum: 0.000000
2023-10-16 23:04:56,713 ----------------------------------------------------------------------------------------------------
2023-10-16 23:04:56,713 EPOCH 8 done: loss 0.0079 - lr: 0.000007
2023-10-16 23:04:58,765 DEV : loss 0.1137828379869461 - f1-score (micro avg) 0.7879
2023-10-16 23:04:58,779 ----------------------------------------------------------------------------------------------------
2023-10-16 23:05:05,640 epoch 9 - iter 154/1546 - loss 0.00625578 - time (sec): 6.86 - samples/sec: 1822.09 - lr: 0.000006 - momentum: 0.000000
2023-10-16 23:05:12,362 epoch 9 - iter 308/1546 - loss 0.00569991 - time (sec): 13.58 - samples/sec: 1786.38 - lr: 0.000006 - momentum: 0.000000
2023-10-16 23:05:19,230 epoch 9 - iter 462/1546 - loss 0.00580179 - time (sec): 20.45 - samples/sec: 1833.99 - lr: 0.000006 - momentum: 0.000000
2023-10-16 23:05:25,993 epoch 9 - iter 616/1546 - loss 0.00507301 - time (sec): 27.21 - samples/sec: 1813.16 - lr: 0.000005 - momentum: 0.000000
2023-10-16 23:05:32,807 epoch 9 - iter 770/1546 - loss 0.00469977 - time (sec): 34.03 - samples/sec: 1799.76 - lr: 0.000005 - momentum: 0.000000
2023-10-16 23:05:39,691 epoch 9 - iter 924/1546 - loss 0.00605868 - time (sec): 40.91 - samples/sec: 1821.49 - lr: 0.000005 - momentum: 0.000000
2023-10-16 23:05:46,528 epoch 9 - iter 1078/1546 - loss 0.00553326 - time (sec): 47.75 - samples/sec: 1821.57 - lr: 0.000004 - momentum: 0.000000
2023-10-16 23:05:53,539 epoch 9 - iter 1232/1546 - loss 0.00506175 - time (sec): 54.76 - samples/sec: 1815.38 - lr: 0.000004 - momentum: 0.000000
2023-10-16 23:06:00,472 epoch 9 - iter 1386/1546 - loss 0.00474763 - time (sec): 61.69 - samples/sec: 1817.48 - lr: 0.000004 - momentum: 0.000000
2023-10-16 23:06:07,322 epoch 9 - iter 1540/1546 - loss 0.00470313 - time (sec): 68.54 - samples/sec: 1802.45 - lr: 0.000003 - momentum: 0.000000
2023-10-16 23:06:07,604 ----------------------------------------------------------------------------------------------------
2023-10-16 23:06:07,604 EPOCH 9 done: loss 0.0047 - lr: 0.000003
2023-10-16 23:06:09,632 DEV : loss 0.11162678897380829 - f1-score (micro avg) 0.8065
2023-10-16 23:06:09,645 saving best model
2023-10-16 23:06:10,098 ----------------------------------------------------------------------------------------------------
2023-10-16 23:06:16,992 epoch 10 - iter 154/1546 - loss 0.00029590 - time (sec): 6.89 - samples/sec: 1868.58 - lr: 0.000003 - momentum: 0.000000
2023-10-16 23:06:23,924 epoch 10 - iter 308/1546 - loss 0.00287854 - time (sec): 13.82 - samples/sec: 1847.28 - lr: 0.000003 - momentum: 0.000000
2023-10-16 23:06:30,870 epoch 10 - iter 462/1546 - loss 0.00262317 - time (sec): 20.77 - samples/sec: 1847.50 - lr: 0.000002 - momentum: 0.000000
2023-10-16 23:06:37,720 epoch 10 - iter 616/1546 - loss 0.00288842 - time (sec): 27.62 - samples/sec: 1830.00 - lr: 0.000002 - momentum: 0.000000
2023-10-16 23:06:44,747 epoch 10 - iter 770/1546 - loss 0.00280402 - time (sec): 34.65 - samples/sec: 1827.44 - lr: 0.000002 - momentum: 0.000000
2023-10-16 23:06:51,575 epoch 10 - iter 924/1546 - loss 0.00318869 - time (sec): 41.48 - samples/sec: 1806.38 - lr: 0.000001 - momentum: 0.000000
2023-10-16 23:06:58,391 epoch 10 - iter 1078/1546 - loss 0.00298211 - time (sec): 48.29 - samples/sec: 1804.58 - lr: 0.000001 - momentum: 0.000000
2023-10-16 23:07:05,298 epoch 10 - iter 1232/1546 - loss 0.00271568 - time (sec): 55.20 - samples/sec: 1808.08 - lr: 0.000001 - momentum: 0.000000
2023-10-16 23:07:12,158 epoch 10 - iter 1386/1546 - loss 0.00309708 - time (sec): 62.06 - samples/sec: 1794.54 - lr: 0.000000 - momentum: 0.000000
2023-10-16 23:07:19,086 epoch 10 - iter 1540/1546 - loss 0.00330551 - time (sec): 68.99 - samples/sec: 1793.54 - lr: 0.000000 - momentum: 0.000000
2023-10-16 23:07:19,363 ----------------------------------------------------------------------------------------------------
2023-10-16 23:07:19,363 EPOCH 10 done: loss 0.0033 - lr: 0.000000
2023-10-16 23:07:21,467 DEV : loss 0.11447464674711227 - f1-score (micro avg) 0.8
2023-10-16 23:07:21,847 ----------------------------------------------------------------------------------------------------
2023-10-16 23:07:21,848 Loading model from best epoch ...
2023-10-16 23:07:23,353 SequenceTagger predicts: Dictionary with 13 tags: O, S-LOC, B-LOC, E-LOC, I-LOC, S-BUILDING, B-BUILDING, E-BUILDING, I-BUILDING, S-STREET, B-STREET, E-STREET, I-STREET
2023-10-16 23:07:29,624
Results:
- F-score (micro) 0.8094
- F-score (macro) 0.7264
- Accuracy 0.7028
By class:
precision recall f1-score support
LOC 0.8396 0.8742 0.8566 946
BUILDING 0.5812 0.6000 0.5904 185
STREET 0.7321 0.7321 0.7321 56
micro avg 0.7946 0.8248 0.8094 1187
macro avg 0.7176 0.7355 0.7264 1187
weighted avg 0.7942 0.8248 0.8092 1187
2023-10-16 23:07:29,625 ----------------------------------------------------------------------------------------------------
|