File size: 24,219 Bytes
2c7e1dc |
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 |
2023-10-23 15:01:13,895 ----------------------------------------------------------------------------------------------------
2023-10-23 15:01:13,896 Model: "SequenceTagger(
(embeddings): TransformerWordEmbeddings(
(model): BertModel(
(embeddings): BertEmbeddings(
(word_embeddings): Embedding(64001, 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()
)"
2023-10-23 15:01:13,896 ----------------------------------------------------------------------------------------------------
2023-10-23 15:01:13,896 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
2023-10-23 15:01:13,896 ----------------------------------------------------------------------------------------------------
2023-10-23 15:01:13,896 Train: 1100 sentences
2023-10-23 15:01:13,896 (train_with_dev=False, train_with_test=False)
2023-10-23 15:01:13,896 ----------------------------------------------------------------------------------------------------
2023-10-23 15:01:13,896 Training Params:
2023-10-23 15:01:13,896 - learning_rate: "3e-05"
2023-10-23 15:01:13,896 - mini_batch_size: "4"
2023-10-23 15:01:13,896 - max_epochs: "10"
2023-10-23 15:01:13,896 - shuffle: "True"
2023-10-23 15:01:13,896 ----------------------------------------------------------------------------------------------------
2023-10-23 15:01:13,896 Plugins:
2023-10-23 15:01:13,896 - TensorboardLogger
2023-10-23 15:01:13,896 - LinearScheduler | warmup_fraction: '0.1'
2023-10-23 15:01:13,896 ----------------------------------------------------------------------------------------------------
2023-10-23 15:01:13,896 Final evaluation on model from best epoch (best-model.pt)
2023-10-23 15:01:13,896 - metric: "('micro avg', 'f1-score')"
2023-10-23 15:01:13,896 ----------------------------------------------------------------------------------------------------
2023-10-23 15:01:13,896 Computation:
2023-10-23 15:01:13,897 - compute on device: cuda:0
2023-10-23 15:01:13,897 - embedding storage: none
2023-10-23 15:01:13,897 ----------------------------------------------------------------------------------------------------
2023-10-23 15:01:13,897 Model training base path: "hmbench-ajmc/de-dbmdz/bert-base-historic-multilingual-64k-td-cased-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-2"
2023-10-23 15:01:13,897 ----------------------------------------------------------------------------------------------------
2023-10-23 15:01:13,897 ----------------------------------------------------------------------------------------------------
2023-10-23 15:01:13,897 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-23 15:01:15,272 epoch 1 - iter 27/275 - loss 3.25393908 - time (sec): 1.37 - samples/sec: 1669.52 - lr: 0.000003 - momentum: 0.000000
2023-10-23 15:01:16,649 epoch 1 - iter 54/275 - loss 2.57464096 - time (sec): 2.75 - samples/sec: 1607.22 - lr: 0.000006 - momentum: 0.000000
2023-10-23 15:01:18,015 epoch 1 - iter 81/275 - loss 2.06080207 - time (sec): 4.12 - samples/sec: 1546.21 - lr: 0.000009 - momentum: 0.000000
2023-10-23 15:01:19,383 epoch 1 - iter 108/275 - loss 1.71478848 - time (sec): 5.48 - samples/sec: 1528.00 - lr: 0.000012 - momentum: 0.000000
2023-10-23 15:01:20,785 epoch 1 - iter 135/275 - loss 1.47161691 - time (sec): 6.89 - samples/sec: 1569.93 - lr: 0.000015 - momentum: 0.000000
2023-10-23 15:01:22,187 epoch 1 - iter 162/275 - loss 1.28629244 - time (sec): 8.29 - samples/sec: 1588.83 - lr: 0.000018 - momentum: 0.000000
2023-10-23 15:01:23,575 epoch 1 - iter 189/275 - loss 1.14942489 - time (sec): 9.68 - samples/sec: 1602.90 - lr: 0.000021 - momentum: 0.000000
2023-10-23 15:01:24,996 epoch 1 - iter 216/275 - loss 1.03789197 - time (sec): 11.10 - samples/sec: 1607.47 - lr: 0.000023 - momentum: 0.000000
2023-10-23 15:01:26,380 epoch 1 - iter 243/275 - loss 0.94916015 - time (sec): 12.48 - samples/sec: 1602.47 - lr: 0.000026 - momentum: 0.000000
2023-10-23 15:01:27,757 epoch 1 - iter 270/275 - loss 0.87843050 - time (sec): 13.86 - samples/sec: 1610.25 - lr: 0.000029 - momentum: 0.000000
2023-10-23 15:01:28,016 ----------------------------------------------------------------------------------------------------
2023-10-23 15:01:28,016 EPOCH 1 done: loss 0.8687 - lr: 0.000029
2023-10-23 15:01:28,433 DEV : loss 0.1798924207687378 - f1-score (micro avg) 0.7726
2023-10-23 15:01:28,438 saving best model
2023-10-23 15:01:28,837 ----------------------------------------------------------------------------------------------------
2023-10-23 15:01:30,227 epoch 2 - iter 27/275 - loss 0.13016473 - time (sec): 1.39 - samples/sec: 1562.72 - lr: 0.000030 - momentum: 0.000000
2023-10-23 15:01:31,622 epoch 2 - iter 54/275 - loss 0.15754978 - time (sec): 2.78 - samples/sec: 1547.47 - lr: 0.000029 - momentum: 0.000000
2023-10-23 15:01:33,017 epoch 2 - iter 81/275 - loss 0.17446397 - time (sec): 4.18 - samples/sec: 1542.82 - lr: 0.000029 - momentum: 0.000000
2023-10-23 15:01:34,403 epoch 2 - iter 108/275 - loss 0.18375447 - time (sec): 5.56 - samples/sec: 1577.39 - lr: 0.000029 - momentum: 0.000000
2023-10-23 15:01:35,796 epoch 2 - iter 135/275 - loss 0.18527314 - time (sec): 6.96 - samples/sec: 1565.96 - lr: 0.000028 - momentum: 0.000000
2023-10-23 15:01:37,217 epoch 2 - iter 162/275 - loss 0.17131923 - time (sec): 8.38 - samples/sec: 1552.03 - lr: 0.000028 - momentum: 0.000000
2023-10-23 15:01:38,613 epoch 2 - iter 189/275 - loss 0.16272828 - time (sec): 9.77 - samples/sec: 1548.36 - lr: 0.000028 - momentum: 0.000000
2023-10-23 15:01:40,029 epoch 2 - iter 216/275 - loss 0.16178130 - time (sec): 11.19 - samples/sec: 1585.99 - lr: 0.000027 - momentum: 0.000000
2023-10-23 15:01:41,427 epoch 2 - iter 243/275 - loss 0.15174993 - time (sec): 12.59 - samples/sec: 1580.54 - lr: 0.000027 - momentum: 0.000000
2023-10-23 15:01:42,820 epoch 2 - iter 270/275 - loss 0.15643905 - time (sec): 13.98 - samples/sec: 1597.35 - lr: 0.000027 - momentum: 0.000000
2023-10-23 15:01:43,079 ----------------------------------------------------------------------------------------------------
2023-10-23 15:01:43,080 EPOCH 2 done: loss 0.1571 - lr: 0.000027
2023-10-23 15:01:43,634 DEV : loss 0.12782859802246094 - f1-score (micro avg) 0.8339
2023-10-23 15:01:43,639 saving best model
2023-10-23 15:01:44,198 ----------------------------------------------------------------------------------------------------
2023-10-23 15:01:45,588 epoch 3 - iter 27/275 - loss 0.07572565 - time (sec): 1.39 - samples/sec: 1631.46 - lr: 0.000026 - momentum: 0.000000
2023-10-23 15:01:46,995 epoch 3 - iter 54/275 - loss 0.08926542 - time (sec): 2.79 - samples/sec: 1557.60 - lr: 0.000026 - momentum: 0.000000
2023-10-23 15:01:48,380 epoch 3 - iter 81/275 - loss 0.08635936 - time (sec): 4.18 - samples/sec: 1504.36 - lr: 0.000026 - momentum: 0.000000
2023-10-23 15:01:49,820 epoch 3 - iter 108/275 - loss 0.09552644 - time (sec): 5.62 - samples/sec: 1554.91 - lr: 0.000025 - momentum: 0.000000
2023-10-23 15:01:51,327 epoch 3 - iter 135/275 - loss 0.09493698 - time (sec): 7.12 - samples/sec: 1562.41 - lr: 0.000025 - momentum: 0.000000
2023-10-23 15:01:52,836 epoch 3 - iter 162/275 - loss 0.10197681 - time (sec): 8.63 - samples/sec: 1562.70 - lr: 0.000025 - momentum: 0.000000
2023-10-23 15:01:54,320 epoch 3 - iter 189/275 - loss 0.10219325 - time (sec): 10.12 - samples/sec: 1546.45 - lr: 0.000024 - momentum: 0.000000
2023-10-23 15:01:55,807 epoch 3 - iter 216/275 - loss 0.10502848 - time (sec): 11.60 - samples/sec: 1523.56 - lr: 0.000024 - momentum: 0.000000
2023-10-23 15:01:57,305 epoch 3 - iter 243/275 - loss 0.09950956 - time (sec): 13.10 - samples/sec: 1518.30 - lr: 0.000024 - momentum: 0.000000
2023-10-23 15:01:58,832 epoch 3 - iter 270/275 - loss 0.09846269 - time (sec): 14.63 - samples/sec: 1528.72 - lr: 0.000023 - momentum: 0.000000
2023-10-23 15:01:59,105 ----------------------------------------------------------------------------------------------------
2023-10-23 15:01:59,105 EPOCH 3 done: loss 0.0985 - lr: 0.000023
2023-10-23 15:01:59,645 DEV : loss 0.12585023045539856 - f1-score (micro avg) 0.8524
2023-10-23 15:01:59,652 saving best model
2023-10-23 15:02:00,197 ----------------------------------------------------------------------------------------------------
2023-10-23 15:02:01,708 epoch 4 - iter 27/275 - loss 0.07581123 - time (sec): 1.51 - samples/sec: 1483.21 - lr: 0.000023 - momentum: 0.000000
2023-10-23 15:02:03,216 epoch 4 - iter 54/275 - loss 0.06572517 - time (sec): 3.01 - samples/sec: 1458.76 - lr: 0.000023 - momentum: 0.000000
2023-10-23 15:02:04,687 epoch 4 - iter 81/275 - loss 0.07992442 - time (sec): 4.49 - samples/sec: 1548.84 - lr: 0.000022 - momentum: 0.000000
2023-10-23 15:02:06,103 epoch 4 - iter 108/275 - loss 0.07802940 - time (sec): 5.90 - samples/sec: 1560.87 - lr: 0.000022 - momentum: 0.000000
2023-10-23 15:02:07,653 epoch 4 - iter 135/275 - loss 0.08316745 - time (sec): 7.45 - samples/sec: 1514.72 - lr: 0.000022 - momentum: 0.000000
2023-10-23 15:02:09,083 epoch 4 - iter 162/275 - loss 0.07750285 - time (sec): 8.88 - samples/sec: 1539.09 - lr: 0.000021 - momentum: 0.000000
2023-10-23 15:02:10,505 epoch 4 - iter 189/275 - loss 0.07107305 - time (sec): 10.30 - samples/sec: 1524.37 - lr: 0.000021 - momentum: 0.000000
2023-10-23 15:02:11,936 epoch 4 - iter 216/275 - loss 0.06532731 - time (sec): 11.73 - samples/sec: 1537.98 - lr: 0.000021 - momentum: 0.000000
2023-10-23 15:02:13,349 epoch 4 - iter 243/275 - loss 0.06597475 - time (sec): 13.15 - samples/sec: 1533.71 - lr: 0.000020 - momentum: 0.000000
2023-10-23 15:02:14,760 epoch 4 - iter 270/275 - loss 0.06189829 - time (sec): 14.56 - samples/sec: 1539.77 - lr: 0.000020 - momentum: 0.000000
2023-10-23 15:02:15,016 ----------------------------------------------------------------------------------------------------
2023-10-23 15:02:15,016 EPOCH 4 done: loss 0.0609 - lr: 0.000020
2023-10-23 15:02:15,560 DEV : loss 0.1704767942428589 - f1-score (micro avg) 0.8595
2023-10-23 15:02:15,566 saving best model
2023-10-23 15:02:16,119 ----------------------------------------------------------------------------------------------------
2023-10-23 15:02:17,500 epoch 5 - iter 27/275 - loss 0.08075043 - time (sec): 1.38 - samples/sec: 1537.85 - lr: 0.000020 - momentum: 0.000000
2023-10-23 15:02:18,937 epoch 5 - iter 54/275 - loss 0.04771189 - time (sec): 2.81 - samples/sec: 1599.62 - lr: 0.000019 - momentum: 0.000000
2023-10-23 15:02:20,408 epoch 5 - iter 81/275 - loss 0.04589936 - time (sec): 4.28 - samples/sec: 1553.83 - lr: 0.000019 - momentum: 0.000000
2023-10-23 15:02:21,827 epoch 5 - iter 108/275 - loss 0.04625470 - time (sec): 5.70 - samples/sec: 1557.53 - lr: 0.000019 - momentum: 0.000000
2023-10-23 15:02:23,256 epoch 5 - iter 135/275 - loss 0.05076116 - time (sec): 7.13 - samples/sec: 1542.98 - lr: 0.000018 - momentum: 0.000000
2023-10-23 15:02:24,653 epoch 5 - iter 162/275 - loss 0.05107033 - time (sec): 8.53 - samples/sec: 1580.55 - lr: 0.000018 - momentum: 0.000000
2023-10-23 15:02:26,053 epoch 5 - iter 189/275 - loss 0.05185483 - time (sec): 9.93 - samples/sec: 1584.96 - lr: 0.000018 - momentum: 0.000000
2023-10-23 15:02:27,463 epoch 5 - iter 216/275 - loss 0.04858783 - time (sec): 11.34 - samples/sec: 1574.70 - lr: 0.000017 - momentum: 0.000000
2023-10-23 15:02:28,833 epoch 5 - iter 243/275 - loss 0.04990404 - time (sec): 12.71 - samples/sec: 1583.66 - lr: 0.000017 - momentum: 0.000000
2023-10-23 15:02:30,264 epoch 5 - iter 270/275 - loss 0.05114650 - time (sec): 14.14 - samples/sec: 1586.53 - lr: 0.000017 - momentum: 0.000000
2023-10-23 15:02:30,522 ----------------------------------------------------------------------------------------------------
2023-10-23 15:02:30,522 EPOCH 5 done: loss 0.0509 - lr: 0.000017
2023-10-23 15:02:31,058 DEV : loss 0.14342500269412994 - f1-score (micro avg) 0.8789
2023-10-23 15:02:31,064 saving best model
2023-10-23 15:02:31,594 ----------------------------------------------------------------------------------------------------
2023-10-23 15:02:33,008 epoch 6 - iter 27/275 - loss 0.05416127 - time (sec): 1.41 - samples/sec: 1747.19 - lr: 0.000016 - momentum: 0.000000
2023-10-23 15:02:34,424 epoch 6 - iter 54/275 - loss 0.03374837 - time (sec): 2.83 - samples/sec: 1743.42 - lr: 0.000016 - momentum: 0.000000
2023-10-23 15:02:35,844 epoch 6 - iter 81/275 - loss 0.04839805 - time (sec): 4.25 - samples/sec: 1687.75 - lr: 0.000016 - momentum: 0.000000
2023-10-23 15:02:37,247 epoch 6 - iter 108/275 - loss 0.04789575 - time (sec): 5.65 - samples/sec: 1596.63 - lr: 0.000015 - momentum: 0.000000
2023-10-23 15:02:38,672 epoch 6 - iter 135/275 - loss 0.04182854 - time (sec): 7.08 - samples/sec: 1586.27 - lr: 0.000015 - momentum: 0.000000
2023-10-23 15:02:40,121 epoch 6 - iter 162/275 - loss 0.04156850 - time (sec): 8.52 - samples/sec: 1599.60 - lr: 0.000015 - momentum: 0.000000
2023-10-23 15:02:41,575 epoch 6 - iter 189/275 - loss 0.03833196 - time (sec): 9.98 - samples/sec: 1604.29 - lr: 0.000014 - momentum: 0.000000
2023-10-23 15:02:42,992 epoch 6 - iter 216/275 - loss 0.03681823 - time (sec): 11.40 - samples/sec: 1591.12 - lr: 0.000014 - momentum: 0.000000
2023-10-23 15:02:44,434 epoch 6 - iter 243/275 - loss 0.03594981 - time (sec): 12.84 - samples/sec: 1586.74 - lr: 0.000014 - momentum: 0.000000
2023-10-23 15:02:45,847 epoch 6 - iter 270/275 - loss 0.03433456 - time (sec): 14.25 - samples/sec: 1574.51 - lr: 0.000013 - momentum: 0.000000
2023-10-23 15:02:46,104 ----------------------------------------------------------------------------------------------------
2023-10-23 15:02:46,104 EPOCH 6 done: loss 0.0346 - lr: 0.000013
2023-10-23 15:02:46,637 DEV : loss 0.13642658293247223 - f1-score (micro avg) 0.8814
2023-10-23 15:02:46,643 saving best model
2023-10-23 15:02:47,200 ----------------------------------------------------------------------------------------------------
2023-10-23 15:02:48,608 epoch 7 - iter 27/275 - loss 0.02650342 - time (sec): 1.40 - samples/sec: 1417.21 - lr: 0.000013 - momentum: 0.000000
2023-10-23 15:02:50,006 epoch 7 - iter 54/275 - loss 0.01385811 - time (sec): 2.80 - samples/sec: 1553.85 - lr: 0.000013 - momentum: 0.000000
2023-10-23 15:02:51,397 epoch 7 - iter 81/275 - loss 0.01617586 - time (sec): 4.19 - samples/sec: 1604.95 - lr: 0.000012 - momentum: 0.000000
2023-10-23 15:02:52,792 epoch 7 - iter 108/275 - loss 0.02170695 - time (sec): 5.59 - samples/sec: 1559.56 - lr: 0.000012 - momentum: 0.000000
2023-10-23 15:02:54,192 epoch 7 - iter 135/275 - loss 0.02142902 - time (sec): 6.99 - samples/sec: 1615.02 - lr: 0.000012 - momentum: 0.000000
2023-10-23 15:02:55,580 epoch 7 - iter 162/275 - loss 0.02338113 - time (sec): 8.38 - samples/sec: 1616.77 - lr: 0.000011 - momentum: 0.000000
2023-10-23 15:02:56,984 epoch 7 - iter 189/275 - loss 0.02495879 - time (sec): 9.78 - samples/sec: 1603.29 - lr: 0.000011 - momentum: 0.000000
2023-10-23 15:02:58,384 epoch 7 - iter 216/275 - loss 0.02648994 - time (sec): 11.18 - samples/sec: 1621.75 - lr: 0.000011 - momentum: 0.000000
2023-10-23 15:02:59,784 epoch 7 - iter 243/275 - loss 0.02714924 - time (sec): 12.58 - samples/sec: 1620.14 - lr: 0.000010 - momentum: 0.000000
2023-10-23 15:03:01,203 epoch 7 - iter 270/275 - loss 0.02572334 - time (sec): 14.00 - samples/sec: 1600.82 - lr: 0.000010 - momentum: 0.000000
2023-10-23 15:03:01,459 ----------------------------------------------------------------------------------------------------
2023-10-23 15:03:01,459 EPOCH 7 done: loss 0.0261 - lr: 0.000010
2023-10-23 15:03:02,002 DEV : loss 0.14091593027114868 - f1-score (micro avg) 0.8889
2023-10-23 15:03:02,008 saving best model
2023-10-23 15:03:02,559 ----------------------------------------------------------------------------------------------------
2023-10-23 15:03:03,964 epoch 8 - iter 27/275 - loss 0.00849142 - time (sec): 1.40 - samples/sec: 1639.27 - lr: 0.000010 - momentum: 0.000000
2023-10-23 15:03:05,368 epoch 8 - iter 54/275 - loss 0.01191158 - time (sec): 2.80 - samples/sec: 1653.99 - lr: 0.000009 - momentum: 0.000000
2023-10-23 15:03:06,755 epoch 8 - iter 81/275 - loss 0.01732017 - time (sec): 4.19 - samples/sec: 1601.75 - lr: 0.000009 - momentum: 0.000000
2023-10-23 15:03:08,151 epoch 8 - iter 108/275 - loss 0.01442721 - time (sec): 5.59 - samples/sec: 1617.09 - lr: 0.000009 - momentum: 0.000000
2023-10-23 15:03:09,545 epoch 8 - iter 135/275 - loss 0.01458848 - time (sec): 6.98 - samples/sec: 1643.43 - lr: 0.000008 - momentum: 0.000000
2023-10-23 15:03:10,940 epoch 8 - iter 162/275 - loss 0.01509681 - time (sec): 8.38 - samples/sec: 1649.02 - lr: 0.000008 - momentum: 0.000000
2023-10-23 15:03:12,328 epoch 8 - iter 189/275 - loss 0.01342641 - time (sec): 9.76 - samples/sec: 1631.85 - lr: 0.000008 - momentum: 0.000000
2023-10-23 15:03:13,716 epoch 8 - iter 216/275 - loss 0.01227183 - time (sec): 11.15 - samples/sec: 1622.71 - lr: 0.000007 - momentum: 0.000000
2023-10-23 15:03:15,113 epoch 8 - iter 243/275 - loss 0.01648427 - time (sec): 12.55 - samples/sec: 1612.23 - lr: 0.000007 - momentum: 0.000000
2023-10-23 15:03:16,499 epoch 8 - iter 270/275 - loss 0.01889030 - time (sec): 13.94 - samples/sec: 1599.89 - lr: 0.000007 - momentum: 0.000000
2023-10-23 15:03:16,766 ----------------------------------------------------------------------------------------------------
2023-10-23 15:03:16,766 EPOCH 8 done: loss 0.0199 - lr: 0.000007
2023-10-23 15:03:17,321 DEV : loss 0.14507058262825012 - f1-score (micro avg) 0.8988
2023-10-23 15:03:17,327 saving best model
2023-10-23 15:03:17,877 ----------------------------------------------------------------------------------------------------
2023-10-23 15:03:19,270 epoch 9 - iter 27/275 - loss 0.01029226 - time (sec): 1.39 - samples/sec: 1423.42 - lr: 0.000006 - momentum: 0.000000
2023-10-23 15:03:20,672 epoch 9 - iter 54/275 - loss 0.01788155 - time (sec): 2.79 - samples/sec: 1558.13 - lr: 0.000006 - momentum: 0.000000
2023-10-23 15:03:22,096 epoch 9 - iter 81/275 - loss 0.01234753 - time (sec): 4.21 - samples/sec: 1601.89 - lr: 0.000006 - momentum: 0.000000
2023-10-23 15:03:23,487 epoch 9 - iter 108/275 - loss 0.01125364 - time (sec): 5.60 - samples/sec: 1573.81 - lr: 0.000005 - momentum: 0.000000
2023-10-23 15:03:24,889 epoch 9 - iter 135/275 - loss 0.01157687 - time (sec): 7.01 - samples/sec: 1580.38 - lr: 0.000005 - momentum: 0.000000
2023-10-23 15:03:26,273 epoch 9 - iter 162/275 - loss 0.01077598 - time (sec): 8.39 - samples/sec: 1573.70 - lr: 0.000005 - momentum: 0.000000
2023-10-23 15:03:27,605 epoch 9 - iter 189/275 - loss 0.01072423 - time (sec): 9.72 - samples/sec: 1611.81 - lr: 0.000004 - momentum: 0.000000
2023-10-23 15:03:28,911 epoch 9 - iter 216/275 - loss 0.01415302 - time (sec): 11.03 - samples/sec: 1622.47 - lr: 0.000004 - momentum: 0.000000
2023-10-23 15:03:30,231 epoch 9 - iter 243/275 - loss 0.01322181 - time (sec): 12.35 - samples/sec: 1621.65 - lr: 0.000004 - momentum: 0.000000
2023-10-23 15:03:31,542 epoch 9 - iter 270/275 - loss 0.01419018 - time (sec): 13.66 - samples/sec: 1635.47 - lr: 0.000003 - momentum: 0.000000
2023-10-23 15:03:31,788 ----------------------------------------------------------------------------------------------------
2023-10-23 15:03:31,788 EPOCH 9 done: loss 0.0141 - lr: 0.000003
2023-10-23 15:03:32,327 DEV : loss 0.15191467106342316 - f1-score (micro avg) 0.89
2023-10-23 15:03:32,333 ----------------------------------------------------------------------------------------------------
2023-10-23 15:03:33,635 epoch 10 - iter 27/275 - loss 0.00481745 - time (sec): 1.30 - samples/sec: 1678.25 - lr: 0.000003 - momentum: 0.000000
2023-10-23 15:03:34,957 epoch 10 - iter 54/275 - loss 0.00245864 - time (sec): 2.62 - samples/sec: 1662.82 - lr: 0.000003 - momentum: 0.000000
2023-10-23 15:03:36,282 epoch 10 - iter 81/275 - loss 0.00464933 - time (sec): 3.95 - samples/sec: 1678.84 - lr: 0.000002 - momentum: 0.000000
2023-10-23 15:03:37,576 epoch 10 - iter 108/275 - loss 0.00510011 - time (sec): 5.24 - samples/sec: 1693.47 - lr: 0.000002 - momentum: 0.000000
2023-10-23 15:03:38,869 epoch 10 - iter 135/275 - loss 0.00565368 - time (sec): 6.53 - samples/sec: 1698.88 - lr: 0.000002 - momentum: 0.000000
2023-10-23 15:03:40,197 epoch 10 - iter 162/275 - loss 0.00574751 - time (sec): 7.86 - samples/sec: 1748.46 - lr: 0.000001 - momentum: 0.000000
2023-10-23 15:03:41,524 epoch 10 - iter 189/275 - loss 0.00588541 - time (sec): 9.19 - samples/sec: 1771.29 - lr: 0.000001 - momentum: 0.000000
2023-10-23 15:03:42,825 epoch 10 - iter 216/275 - loss 0.00763093 - time (sec): 10.49 - samples/sec: 1752.62 - lr: 0.000001 - momentum: 0.000000
2023-10-23 15:03:44,145 epoch 10 - iter 243/275 - loss 0.00980050 - time (sec): 11.81 - samples/sec: 1729.03 - lr: 0.000000 - momentum: 0.000000
2023-10-23 15:03:45,443 epoch 10 - iter 270/275 - loss 0.01148583 - time (sec): 13.11 - samples/sec: 1709.20 - lr: 0.000000 - momentum: 0.000000
2023-10-23 15:03:45,696 ----------------------------------------------------------------------------------------------------
2023-10-23 15:03:45,696 EPOCH 10 done: loss 0.0113 - lr: 0.000000
2023-10-23 15:03:46,229 DEV : loss 0.15140938758850098 - f1-score (micro avg) 0.8935
2023-10-23 15:03:46,636 ----------------------------------------------------------------------------------------------------
2023-10-23 15:03:46,637 Loading model from best epoch ...
2023-10-23 15:03:48,383 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
2023-10-23 15:03:49,046
Results:
- F-score (micro) 0.9227
- F-score (macro) 0.7845
- Accuracy 0.8691
By class:
precision recall f1-score support
scope 0.9157 0.9261 0.9209 176
pers 0.9839 0.9531 0.9683 128
work 0.8553 0.8784 0.8667 74
object 0.5000 0.5000 0.5000 2
loc 1.0000 0.5000 0.6667 2
micro avg 0.9239 0.9215 0.9227 382
macro avg 0.8510 0.7515 0.7845 382
weighted avg 0.9251 0.9215 0.9227 382
2023-10-23 15:03:49,046 ----------------------------------------------------------------------------------------------------
|