File size: 24,020 Bytes
ec16760 |
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 |
2023-10-13 08:30:02,855 ----------------------------------------------------------------------------------------------------
2023-10-13 08:30:02,856 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=25, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-13 08:30:02,856 ----------------------------------------------------------------------------------------------------
2023-10-13 08:30:02,856 MultiCorpus: 1100 train + 206 dev + 240 test sentences
- NER_HIPE_2022 Corpus: 1100 train + 206 dev + 240 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/ajmc/de/with_doc_seperator
2023-10-13 08:30:02,856 ----------------------------------------------------------------------------------------------------
2023-10-13 08:30:02,856 Train: 1100 sentences
2023-10-13 08:30:02,856 (train_with_dev=False, train_with_test=False)
2023-10-13 08:30:02,856 ----------------------------------------------------------------------------------------------------
2023-10-13 08:30:02,856 Training Params:
2023-10-13 08:30:02,856 - learning_rate: "3e-05"
2023-10-13 08:30:02,856 - mini_batch_size: "4"
2023-10-13 08:30:02,856 - max_epochs: "10"
2023-10-13 08:30:02,856 - shuffle: "True"
2023-10-13 08:30:02,856 ----------------------------------------------------------------------------------------------------
2023-10-13 08:30:02,856 Plugins:
2023-10-13 08:30:02,856 - LinearScheduler | warmup_fraction: '0.1'
2023-10-13 08:30:02,856 ----------------------------------------------------------------------------------------------------
2023-10-13 08:30:02,856 Final evaluation on model from best epoch (best-model.pt)
2023-10-13 08:30:02,856 - metric: "('micro avg', 'f1-score')"
2023-10-13 08:30:02,856 ----------------------------------------------------------------------------------------------------
2023-10-13 08:30:02,856 Computation:
2023-10-13 08:30:02,856 - compute on device: cuda:0
2023-10-13 08:30:02,856 - embedding storage: none
2023-10-13 08:30:02,856 ----------------------------------------------------------------------------------------------------
2023-10-13 08:30:02,856 Model training base path: "hmbench-ajmc/de-dbmdz/bert-base-historic-multilingual-cased-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-3"
2023-10-13 08:30:02,856 ----------------------------------------------------------------------------------------------------
2023-10-13 08:30:02,856 ----------------------------------------------------------------------------------------------------
2023-10-13 08:30:04,144 epoch 1 - iter 27/275 - loss 3.12256091 - time (sec): 1.29 - samples/sec: 1881.91 - lr: 0.000003 - momentum: 0.000000
2023-10-13 08:30:05,494 epoch 1 - iter 54/275 - loss 2.73588781 - time (sec): 2.64 - samples/sec: 1721.27 - lr: 0.000006 - momentum: 0.000000
2023-10-13 08:30:06,747 epoch 1 - iter 81/275 - loss 2.11741342 - time (sec): 3.89 - samples/sec: 1759.29 - lr: 0.000009 - momentum: 0.000000
2023-10-13 08:30:07,941 epoch 1 - iter 108/275 - loss 1.76793211 - time (sec): 5.08 - samples/sec: 1777.11 - lr: 0.000012 - momentum: 0.000000
2023-10-13 08:30:09,143 epoch 1 - iter 135/275 - loss 1.56633956 - time (sec): 6.29 - samples/sec: 1795.64 - lr: 0.000015 - momentum: 0.000000
2023-10-13 08:30:10,376 epoch 1 - iter 162/275 - loss 1.42899417 - time (sec): 7.52 - samples/sec: 1764.94 - lr: 0.000018 - momentum: 0.000000
2023-10-13 08:30:11,623 epoch 1 - iter 189/275 - loss 1.28375594 - time (sec): 8.77 - samples/sec: 1790.74 - lr: 0.000021 - momentum: 0.000000
2023-10-13 08:30:12,860 epoch 1 - iter 216/275 - loss 1.16818959 - time (sec): 10.00 - samples/sec: 1784.82 - lr: 0.000023 - momentum: 0.000000
2023-10-13 08:30:14,096 epoch 1 - iter 243/275 - loss 1.08409111 - time (sec): 11.24 - samples/sec: 1787.90 - lr: 0.000026 - momentum: 0.000000
2023-10-13 08:30:15,283 epoch 1 - iter 270/275 - loss 1.00861787 - time (sec): 12.43 - samples/sec: 1792.64 - lr: 0.000029 - momentum: 0.000000
2023-10-13 08:30:15,505 ----------------------------------------------------------------------------------------------------
2023-10-13 08:30:15,505 EPOCH 1 done: loss 0.9973 - lr: 0.000029
2023-10-13 08:30:16,318 DEV : loss 0.23283608257770538 - f1-score (micro avg) 0.6629
2023-10-13 08:30:16,323 saving best model
2023-10-13 08:30:16,660 ----------------------------------------------------------------------------------------------------
2023-10-13 08:30:17,809 epoch 2 - iter 27/275 - loss 0.20631712 - time (sec): 1.15 - samples/sec: 1806.74 - lr: 0.000030 - momentum: 0.000000
2023-10-13 08:30:18,998 epoch 2 - iter 54/275 - loss 0.22953289 - time (sec): 2.34 - samples/sec: 1904.20 - lr: 0.000029 - momentum: 0.000000
2023-10-13 08:30:20,152 epoch 2 - iter 81/275 - loss 0.22170780 - time (sec): 3.49 - samples/sec: 1937.42 - lr: 0.000029 - momentum: 0.000000
2023-10-13 08:30:21,374 epoch 2 - iter 108/275 - loss 0.21397525 - time (sec): 4.71 - samples/sec: 1972.24 - lr: 0.000029 - momentum: 0.000000
2023-10-13 08:30:22,571 epoch 2 - iter 135/275 - loss 0.21197127 - time (sec): 5.91 - samples/sec: 1977.94 - lr: 0.000028 - momentum: 0.000000
2023-10-13 08:30:23,747 epoch 2 - iter 162/275 - loss 0.20464864 - time (sec): 7.09 - samples/sec: 1956.91 - lr: 0.000028 - momentum: 0.000000
2023-10-13 08:30:24,930 epoch 2 - iter 189/275 - loss 0.20054963 - time (sec): 8.27 - samples/sec: 1933.60 - lr: 0.000028 - momentum: 0.000000
2023-10-13 08:30:26,120 epoch 2 - iter 216/275 - loss 0.19706410 - time (sec): 9.46 - samples/sec: 1930.90 - lr: 0.000027 - momentum: 0.000000
2023-10-13 08:30:27,256 epoch 2 - iter 243/275 - loss 0.18940321 - time (sec): 10.59 - samples/sec: 1928.02 - lr: 0.000027 - momentum: 0.000000
2023-10-13 08:30:28,416 epoch 2 - iter 270/275 - loss 0.18411801 - time (sec): 11.75 - samples/sec: 1902.18 - lr: 0.000027 - momentum: 0.000000
2023-10-13 08:30:28,646 ----------------------------------------------------------------------------------------------------
2023-10-13 08:30:28,647 EPOCH 2 done: loss 0.1857 - lr: 0.000027
2023-10-13 08:30:29,336 DEV : loss 0.13876736164093018 - f1-score (micro avg) 0.8377
2023-10-13 08:30:29,342 saving best model
2023-10-13 08:30:29,797 ----------------------------------------------------------------------------------------------------
2023-10-13 08:30:31,028 epoch 3 - iter 27/275 - loss 0.08676695 - time (sec): 1.23 - samples/sec: 1865.67 - lr: 0.000026 - momentum: 0.000000
2023-10-13 08:30:32,243 epoch 3 - iter 54/275 - loss 0.11105096 - time (sec): 2.44 - samples/sec: 1900.57 - lr: 0.000026 - momentum: 0.000000
2023-10-13 08:30:33,443 epoch 3 - iter 81/275 - loss 0.11026128 - time (sec): 3.64 - samples/sec: 1894.97 - lr: 0.000026 - momentum: 0.000000
2023-10-13 08:30:34,635 epoch 3 - iter 108/275 - loss 0.09728597 - time (sec): 4.84 - samples/sec: 1877.65 - lr: 0.000025 - momentum: 0.000000
2023-10-13 08:30:35,834 epoch 3 - iter 135/275 - loss 0.11057918 - time (sec): 6.03 - samples/sec: 1905.70 - lr: 0.000025 - momentum: 0.000000
2023-10-13 08:30:37,014 epoch 3 - iter 162/275 - loss 0.11211846 - time (sec): 7.21 - samples/sec: 1878.86 - lr: 0.000025 - momentum: 0.000000
2023-10-13 08:30:38,268 epoch 3 - iter 189/275 - loss 0.10686033 - time (sec): 8.47 - samples/sec: 1866.84 - lr: 0.000024 - momentum: 0.000000
2023-10-13 08:30:39,465 epoch 3 - iter 216/275 - loss 0.10678281 - time (sec): 9.66 - samples/sec: 1858.69 - lr: 0.000024 - momentum: 0.000000
2023-10-13 08:30:40,664 epoch 3 - iter 243/275 - loss 0.10493191 - time (sec): 10.86 - samples/sec: 1867.30 - lr: 0.000024 - momentum: 0.000000
2023-10-13 08:30:41,872 epoch 3 - iter 270/275 - loss 0.10312391 - time (sec): 12.07 - samples/sec: 1855.13 - lr: 0.000023 - momentum: 0.000000
2023-10-13 08:30:42,095 ----------------------------------------------------------------------------------------------------
2023-10-13 08:30:42,096 EPOCH 3 done: loss 0.1019 - lr: 0.000023
2023-10-13 08:30:42,829 DEV : loss 0.1486443281173706 - f1-score (micro avg) 0.8523
2023-10-13 08:30:42,836 saving best model
2023-10-13 08:30:43,371 ----------------------------------------------------------------------------------------------------
2023-10-13 08:30:44,793 epoch 4 - iter 27/275 - loss 0.07441677 - time (sec): 1.42 - samples/sec: 1577.23 - lr: 0.000023 - momentum: 0.000000
2023-10-13 08:30:46,179 epoch 4 - iter 54/275 - loss 0.08226076 - time (sec): 2.80 - samples/sec: 1570.96 - lr: 0.000023 - momentum: 0.000000
2023-10-13 08:30:47,582 epoch 4 - iter 81/275 - loss 0.06455288 - time (sec): 4.20 - samples/sec: 1614.32 - lr: 0.000022 - momentum: 0.000000
2023-10-13 08:30:48,954 epoch 4 - iter 108/275 - loss 0.07124220 - time (sec): 5.58 - samples/sec: 1586.17 - lr: 0.000022 - momentum: 0.000000
2023-10-13 08:30:50,353 epoch 4 - iter 135/275 - loss 0.07867153 - time (sec): 6.98 - samples/sec: 1598.33 - lr: 0.000022 - momentum: 0.000000
2023-10-13 08:30:51,688 epoch 4 - iter 162/275 - loss 0.07905647 - time (sec): 8.31 - samples/sec: 1617.34 - lr: 0.000021 - momentum: 0.000000
2023-10-13 08:30:53,061 epoch 4 - iter 189/275 - loss 0.07773678 - time (sec): 9.68 - samples/sec: 1621.71 - lr: 0.000021 - momentum: 0.000000
2023-10-13 08:30:54,274 epoch 4 - iter 216/275 - loss 0.07725977 - time (sec): 10.90 - samples/sec: 1632.74 - lr: 0.000021 - momentum: 0.000000
2023-10-13 08:30:55,444 epoch 4 - iter 243/275 - loss 0.07933465 - time (sec): 12.07 - samples/sec: 1642.09 - lr: 0.000020 - momentum: 0.000000
2023-10-13 08:30:56,612 epoch 4 - iter 270/275 - loss 0.07839244 - time (sec): 13.23 - samples/sec: 1689.29 - lr: 0.000020 - momentum: 0.000000
2023-10-13 08:30:56,828 ----------------------------------------------------------------------------------------------------
2023-10-13 08:30:56,829 EPOCH 4 done: loss 0.0774 - lr: 0.000020
2023-10-13 08:30:57,572 DEV : loss 0.15350772440433502 - f1-score (micro avg) 0.8758
2023-10-13 08:30:57,578 saving best model
2023-10-13 08:30:58,200 ----------------------------------------------------------------------------------------------------
2023-10-13 08:30:59,595 epoch 5 - iter 27/275 - loss 0.07548554 - time (sec): 1.39 - samples/sec: 1754.44 - lr: 0.000020 - momentum: 0.000000
2023-10-13 08:31:01,038 epoch 5 - iter 54/275 - loss 0.06764304 - time (sec): 2.84 - samples/sec: 1684.43 - lr: 0.000019 - momentum: 0.000000
2023-10-13 08:31:02,436 epoch 5 - iter 81/275 - loss 0.06984076 - time (sec): 4.23 - samples/sec: 1630.67 - lr: 0.000019 - momentum: 0.000000
2023-10-13 08:31:03,882 epoch 5 - iter 108/275 - loss 0.06004683 - time (sec): 5.68 - samples/sec: 1601.75 - lr: 0.000019 - momentum: 0.000000
2023-10-13 08:31:05,333 epoch 5 - iter 135/275 - loss 0.06436743 - time (sec): 7.13 - samples/sec: 1583.01 - lr: 0.000018 - momentum: 0.000000
2023-10-13 08:31:06,783 epoch 5 - iter 162/275 - loss 0.05845132 - time (sec): 8.58 - samples/sec: 1550.48 - lr: 0.000018 - momentum: 0.000000
2023-10-13 08:31:08,243 epoch 5 - iter 189/275 - loss 0.05821264 - time (sec): 10.04 - samples/sec: 1552.31 - lr: 0.000018 - momentum: 0.000000
2023-10-13 08:31:09,661 epoch 5 - iter 216/275 - loss 0.06126030 - time (sec): 11.46 - samples/sec: 1545.09 - lr: 0.000017 - momentum: 0.000000
2023-10-13 08:31:11,113 epoch 5 - iter 243/275 - loss 0.05900075 - time (sec): 12.91 - samples/sec: 1561.69 - lr: 0.000017 - momentum: 0.000000
2023-10-13 08:31:12,497 epoch 5 - iter 270/275 - loss 0.05727806 - time (sec): 14.30 - samples/sec: 1563.51 - lr: 0.000017 - momentum: 0.000000
2023-10-13 08:31:12,765 ----------------------------------------------------------------------------------------------------
2023-10-13 08:31:12,765 EPOCH 5 done: loss 0.0604 - lr: 0.000017
2023-10-13 08:31:13,425 DEV : loss 0.15178388357162476 - f1-score (micro avg) 0.8668
2023-10-13 08:31:13,430 ----------------------------------------------------------------------------------------------------
2023-10-13 08:31:14,809 epoch 6 - iter 27/275 - loss 0.04081421 - time (sec): 1.38 - samples/sec: 1709.43 - lr: 0.000016 - momentum: 0.000000
2023-10-13 08:31:16,072 epoch 6 - iter 54/275 - loss 0.04499500 - time (sec): 2.64 - samples/sec: 1795.16 - lr: 0.000016 - momentum: 0.000000
2023-10-13 08:31:17,259 epoch 6 - iter 81/275 - loss 0.04264265 - time (sec): 3.83 - samples/sec: 1755.59 - lr: 0.000016 - momentum: 0.000000
2023-10-13 08:31:18,451 epoch 6 - iter 108/275 - loss 0.04066899 - time (sec): 5.02 - samples/sec: 1764.48 - lr: 0.000015 - momentum: 0.000000
2023-10-13 08:31:19,876 epoch 6 - iter 135/275 - loss 0.04266846 - time (sec): 6.44 - samples/sec: 1728.64 - lr: 0.000015 - momentum: 0.000000
2023-10-13 08:31:21,313 epoch 6 - iter 162/275 - loss 0.03859168 - time (sec): 7.88 - samples/sec: 1690.96 - lr: 0.000015 - momentum: 0.000000
2023-10-13 08:31:22,664 epoch 6 - iter 189/275 - loss 0.04050303 - time (sec): 9.23 - samples/sec: 1669.95 - lr: 0.000014 - momentum: 0.000000
2023-10-13 08:31:23,900 epoch 6 - iter 216/275 - loss 0.03833768 - time (sec): 10.47 - samples/sec: 1692.11 - lr: 0.000014 - momentum: 0.000000
2023-10-13 08:31:25,180 epoch 6 - iter 243/275 - loss 0.04402705 - time (sec): 11.75 - samples/sec: 1718.16 - lr: 0.000014 - momentum: 0.000000
2023-10-13 08:31:26,464 epoch 6 - iter 270/275 - loss 0.04151386 - time (sec): 13.03 - samples/sec: 1721.61 - lr: 0.000013 - momentum: 0.000000
2023-10-13 08:31:26,680 ----------------------------------------------------------------------------------------------------
2023-10-13 08:31:26,680 EPOCH 6 done: loss 0.0422 - lr: 0.000013
2023-10-13 08:31:27,425 DEV : loss 0.15648043155670166 - f1-score (micro avg) 0.8775
2023-10-13 08:31:27,431 saving best model
2023-10-13 08:31:28,034 ----------------------------------------------------------------------------------------------------
2023-10-13 08:31:29,450 epoch 7 - iter 27/275 - loss 0.04169935 - time (sec): 1.41 - samples/sec: 1607.86 - lr: 0.000013 - momentum: 0.000000
2023-10-13 08:31:30,721 epoch 7 - iter 54/275 - loss 0.02824331 - time (sec): 2.68 - samples/sec: 1592.78 - lr: 0.000013 - momentum: 0.000000
2023-10-13 08:31:31,988 epoch 7 - iter 81/275 - loss 0.03815050 - time (sec): 3.95 - samples/sec: 1679.31 - lr: 0.000012 - momentum: 0.000000
2023-10-13 08:31:33,215 epoch 7 - iter 108/275 - loss 0.02987254 - time (sec): 5.18 - samples/sec: 1722.11 - lr: 0.000012 - momentum: 0.000000
2023-10-13 08:31:34,512 epoch 7 - iter 135/275 - loss 0.03424007 - time (sec): 6.47 - samples/sec: 1730.27 - lr: 0.000012 - momentum: 0.000000
2023-10-13 08:31:35,896 epoch 7 - iter 162/275 - loss 0.03450026 - time (sec): 7.86 - samples/sec: 1708.45 - lr: 0.000011 - momentum: 0.000000
2023-10-13 08:31:37,308 epoch 7 - iter 189/275 - loss 0.03417843 - time (sec): 9.27 - samples/sec: 1706.92 - lr: 0.000011 - momentum: 0.000000
2023-10-13 08:31:38,693 epoch 7 - iter 216/275 - loss 0.03320135 - time (sec): 10.66 - samples/sec: 1678.75 - lr: 0.000011 - momentum: 0.000000
2023-10-13 08:31:40,145 epoch 7 - iter 243/275 - loss 0.02986744 - time (sec): 12.11 - samples/sec: 1667.99 - lr: 0.000010 - momentum: 0.000000
2023-10-13 08:31:41,528 epoch 7 - iter 270/275 - loss 0.03255562 - time (sec): 13.49 - samples/sec: 1657.39 - lr: 0.000010 - momentum: 0.000000
2023-10-13 08:31:41,742 ----------------------------------------------------------------------------------------------------
2023-10-13 08:31:41,743 EPOCH 7 done: loss 0.0320 - lr: 0.000010
2023-10-13 08:31:42,446 DEV : loss 0.1622074544429779 - f1-score (micro avg) 0.8645
2023-10-13 08:31:42,453 ----------------------------------------------------------------------------------------------------
2023-10-13 08:31:43,828 epoch 8 - iter 27/275 - loss 0.01838056 - time (sec): 1.37 - samples/sec: 1668.51 - lr: 0.000010 - momentum: 0.000000
2023-10-13 08:31:45,182 epoch 8 - iter 54/275 - loss 0.02235674 - time (sec): 2.73 - samples/sec: 1620.96 - lr: 0.000009 - momentum: 0.000000
2023-10-13 08:31:46,523 epoch 8 - iter 81/275 - loss 0.01827360 - time (sec): 4.07 - samples/sec: 1652.60 - lr: 0.000009 - momentum: 0.000000
2023-10-13 08:31:47,933 epoch 8 - iter 108/275 - loss 0.02429007 - time (sec): 5.48 - samples/sec: 1669.54 - lr: 0.000009 - momentum: 0.000000
2023-10-13 08:31:49,349 epoch 8 - iter 135/275 - loss 0.02657182 - time (sec): 6.89 - samples/sec: 1646.59 - lr: 0.000008 - momentum: 0.000000
2023-10-13 08:31:50,782 epoch 8 - iter 162/275 - loss 0.02556321 - time (sec): 8.33 - samples/sec: 1634.43 - lr: 0.000008 - momentum: 0.000000
2023-10-13 08:31:52,192 epoch 8 - iter 189/275 - loss 0.02914361 - time (sec): 9.74 - samples/sec: 1624.79 - lr: 0.000008 - momentum: 0.000000
2023-10-13 08:31:53,585 epoch 8 - iter 216/275 - loss 0.02779016 - time (sec): 11.13 - samples/sec: 1596.03 - lr: 0.000007 - momentum: 0.000000
2023-10-13 08:31:55,003 epoch 8 - iter 243/275 - loss 0.02659038 - time (sec): 12.55 - samples/sec: 1597.71 - lr: 0.000007 - momentum: 0.000000
2023-10-13 08:31:56,388 epoch 8 - iter 270/275 - loss 0.02472247 - time (sec): 13.93 - samples/sec: 1599.82 - lr: 0.000007 - momentum: 0.000000
2023-10-13 08:31:56,638 ----------------------------------------------------------------------------------------------------
2023-10-13 08:31:56,638 EPOCH 8 done: loss 0.0265 - lr: 0.000007
2023-10-13 08:31:57,338 DEV : loss 0.15169711410999298 - f1-score (micro avg) 0.894
2023-10-13 08:31:57,344 saving best model
2023-10-13 08:31:57,896 ----------------------------------------------------------------------------------------------------
2023-10-13 08:31:59,298 epoch 9 - iter 27/275 - loss 0.02854229 - time (sec): 1.40 - samples/sec: 1622.74 - lr: 0.000006 - momentum: 0.000000
2023-10-13 08:32:00,683 epoch 9 - iter 54/275 - loss 0.01793449 - time (sec): 2.79 - samples/sec: 1561.20 - lr: 0.000006 - momentum: 0.000000
2023-10-13 08:32:02,087 epoch 9 - iter 81/275 - loss 0.01751264 - time (sec): 4.19 - samples/sec: 1572.81 - lr: 0.000006 - momentum: 0.000000
2023-10-13 08:32:03,475 epoch 9 - iter 108/275 - loss 0.01705168 - time (sec): 5.58 - samples/sec: 1531.77 - lr: 0.000005 - momentum: 0.000000
2023-10-13 08:32:04,888 epoch 9 - iter 135/275 - loss 0.02420730 - time (sec): 6.99 - samples/sec: 1593.00 - lr: 0.000005 - momentum: 0.000000
2023-10-13 08:32:06,284 epoch 9 - iter 162/275 - loss 0.02315029 - time (sec): 8.39 - samples/sec: 1623.26 - lr: 0.000005 - momentum: 0.000000
2023-10-13 08:32:07,954 epoch 9 - iter 189/275 - loss 0.02030702 - time (sec): 10.06 - samples/sec: 1574.07 - lr: 0.000004 - momentum: 0.000000
2023-10-13 08:32:09,750 epoch 9 - iter 216/275 - loss 0.02223492 - time (sec): 11.85 - samples/sec: 1521.39 - lr: 0.000004 - momentum: 0.000000
2023-10-13 08:32:11,140 epoch 9 - iter 243/275 - loss 0.02034742 - time (sec): 13.24 - samples/sec: 1520.43 - lr: 0.000004 - momentum: 0.000000
2023-10-13 08:32:12,486 epoch 9 - iter 270/275 - loss 0.02192502 - time (sec): 14.59 - samples/sec: 1530.07 - lr: 0.000003 - momentum: 0.000000
2023-10-13 08:32:12,733 ----------------------------------------------------------------------------------------------------
2023-10-13 08:32:12,733 EPOCH 9 done: loss 0.0216 - lr: 0.000003
2023-10-13 08:32:13,467 DEV : loss 0.14888478815555573 - f1-score (micro avg) 0.8951
2023-10-13 08:32:13,474 saving best model
2023-10-13 08:32:14,259 ----------------------------------------------------------------------------------------------------
2023-10-13 08:32:15,685 epoch 10 - iter 27/275 - loss 0.03725164 - time (sec): 1.42 - samples/sec: 1607.61 - lr: 0.000003 - momentum: 0.000000
2023-10-13 08:32:17,128 epoch 10 - iter 54/275 - loss 0.04843116 - time (sec): 2.87 - samples/sec: 1594.73 - lr: 0.000003 - momentum: 0.000000
2023-10-13 08:32:18,538 epoch 10 - iter 81/275 - loss 0.03243751 - time (sec): 4.28 - samples/sec: 1633.39 - lr: 0.000002 - momentum: 0.000000
2023-10-13 08:32:20,014 epoch 10 - iter 108/275 - loss 0.03059973 - time (sec): 5.75 - samples/sec: 1584.55 - lr: 0.000002 - momentum: 0.000000
2023-10-13 08:32:21,378 epoch 10 - iter 135/275 - loss 0.02487169 - time (sec): 7.12 - samples/sec: 1582.45 - lr: 0.000002 - momentum: 0.000000
2023-10-13 08:32:22,789 epoch 10 - iter 162/275 - loss 0.02418583 - time (sec): 8.53 - samples/sec: 1616.61 - lr: 0.000001 - momentum: 0.000000
2023-10-13 08:32:24,210 epoch 10 - iter 189/275 - loss 0.02168156 - time (sec): 9.95 - samples/sec: 1595.54 - lr: 0.000001 - momentum: 0.000000
2023-10-13 08:32:25,575 epoch 10 - iter 216/275 - loss 0.01915487 - time (sec): 11.31 - samples/sec: 1591.14 - lr: 0.000001 - momentum: 0.000000
2023-10-13 08:32:26,927 epoch 10 - iter 243/275 - loss 0.01800133 - time (sec): 12.67 - samples/sec: 1585.16 - lr: 0.000000 - momentum: 0.000000
2023-10-13 08:32:28,275 epoch 10 - iter 270/275 - loss 0.01827690 - time (sec): 14.01 - samples/sec: 1589.32 - lr: 0.000000 - momentum: 0.000000
2023-10-13 08:32:28,531 ----------------------------------------------------------------------------------------------------
2023-10-13 08:32:28,531 EPOCH 10 done: loss 0.0179 - lr: 0.000000
2023-10-13 08:32:29,219 DEV : loss 0.15245532989501953 - f1-score (micro avg) 0.894
2023-10-13 08:32:29,686 ----------------------------------------------------------------------------------------------------
2023-10-13 08:32:29,687 Loading model from best epoch ...
2023-10-13 08:32:31,411 SequenceTagger predicts: Dictionary with 25 tags: O, S-scope, B-scope, E-scope, I-scope, S-pers, B-pers, E-pers, I-pers, S-work, B-work, E-work, I-work, S-loc, B-loc, E-loc, I-loc, S-object, B-object, E-object, I-object, S-date, B-date, E-date, I-date
2023-10-13 08:32:32,043
Results:
- F-score (micro) 0.9217
- F-score (macro) 0.8466
- Accuracy 0.8652
By class:
precision recall f1-score support
scope 0.8950 0.9205 0.9076 176
pers 0.9683 0.9531 0.9606 128
work 0.9041 0.8919 0.8980 74
loc 0.6667 1.0000 0.8000 2
object 1.0000 0.5000 0.6667 2
micro avg 0.9193 0.9241 0.9217 382
macro avg 0.8868 0.8531 0.8466 382
weighted avg 0.9207 0.9241 0.9217 382
2023-10-13 08:32:32,043 ----------------------------------------------------------------------------------------------------
|