File size: 24,066 Bytes
e1394b3 |
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 |
2023-10-17 08:46:12,766 ----------------------------------------------------------------------------------------------------
2023-10-17 08:46:12,767 Model: "SequenceTagger(
(embeddings): TransformerWordEmbeddings(
(model): ElectraModel(
(embeddings): ElectraEmbeddings(
(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): ElectraEncoder(
(layer): ModuleList(
(0-11): 12 x ElectraLayer(
(attention): ElectraAttention(
(self): ElectraSelfAttention(
(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): ElectraSelfOutput(
(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): ElectraIntermediate(
(dense): Linear(in_features=768, out_features=3072, bias=True)
(intermediate_act_fn): GELUActivation()
)
(output): ElectraOutput(
(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)
)
)
)
)
)
)
(locked_dropout): LockedDropout(p=0.5)
(linear): Linear(in_features=768, out_features=25, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-17 08:46:12,767 ----------------------------------------------------------------------------------------------------
2023-10-17 08:46:12,767 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
2023-10-17 08:46:12,767 ----------------------------------------------------------------------------------------------------
2023-10-17 08:46:12,767 Train: 1100 sentences
2023-10-17 08:46:12,768 (train_with_dev=False, train_with_test=False)
2023-10-17 08:46:12,768 ----------------------------------------------------------------------------------------------------
2023-10-17 08:46:12,768 Training Params:
2023-10-17 08:46:12,768 - learning_rate: "3e-05"
2023-10-17 08:46:12,768 - mini_batch_size: "4"
2023-10-17 08:46:12,768 - max_epochs: "10"
2023-10-17 08:46:12,768 - shuffle: "True"
2023-10-17 08:46:12,768 ----------------------------------------------------------------------------------------------------
2023-10-17 08:46:12,768 Plugins:
2023-10-17 08:46:12,768 - TensorboardLogger
2023-10-17 08:46:12,768 - LinearScheduler | warmup_fraction: '0.1'
2023-10-17 08:46:12,768 ----------------------------------------------------------------------------------------------------
2023-10-17 08:46:12,768 Final evaluation on model from best epoch (best-model.pt)
2023-10-17 08:46:12,768 - metric: "('micro avg', 'f1-score')"
2023-10-17 08:46:12,768 ----------------------------------------------------------------------------------------------------
2023-10-17 08:46:12,768 Computation:
2023-10-17 08:46:12,768 - compute on device: cuda:0
2023-10-17 08:46:12,768 - embedding storage: none
2023-10-17 08:46:12,768 ----------------------------------------------------------------------------------------------------
2023-10-17 08:46:12,768 Model training base path: "hmbench-ajmc/de-hmteams/teams-base-historic-multilingual-discriminator-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-4"
2023-10-17 08:46:12,768 ----------------------------------------------------------------------------------------------------
2023-10-17 08:46:12,768 ----------------------------------------------------------------------------------------------------
2023-10-17 08:46:12,768 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-17 08:46:14,067 epoch 1 - iter 27/275 - loss 3.97024551 - time (sec): 1.30 - samples/sec: 1816.98 - lr: 0.000003 - momentum: 0.000000
2023-10-17 08:46:15,306 epoch 1 - iter 54/275 - loss 3.41470820 - time (sec): 2.54 - samples/sec: 1694.36 - lr: 0.000006 - momentum: 0.000000
2023-10-17 08:46:16,558 epoch 1 - iter 81/275 - loss 2.68789197 - time (sec): 3.79 - samples/sec: 1760.77 - lr: 0.000009 - momentum: 0.000000
2023-10-17 08:46:17,774 epoch 1 - iter 108/275 - loss 2.21443759 - time (sec): 5.01 - samples/sec: 1747.17 - lr: 0.000012 - momentum: 0.000000
2023-10-17 08:46:19,034 epoch 1 - iter 135/275 - loss 1.87610511 - time (sec): 6.26 - samples/sec: 1745.65 - lr: 0.000015 - momentum: 0.000000
2023-10-17 08:46:20,265 epoch 1 - iter 162/275 - loss 1.63283108 - time (sec): 7.50 - samples/sec: 1762.93 - lr: 0.000018 - momentum: 0.000000
2023-10-17 08:46:21,520 epoch 1 - iter 189/275 - loss 1.44276987 - time (sec): 8.75 - samples/sec: 1797.17 - lr: 0.000021 - momentum: 0.000000
2023-10-17 08:46:22,795 epoch 1 - iter 216/275 - loss 1.28758424 - time (sec): 10.03 - samples/sec: 1819.62 - lr: 0.000023 - momentum: 0.000000
2023-10-17 08:46:24,026 epoch 1 - iter 243/275 - loss 1.19494444 - time (sec): 11.26 - samples/sec: 1791.18 - lr: 0.000026 - momentum: 0.000000
2023-10-17 08:46:25,261 epoch 1 - iter 270/275 - loss 1.10556620 - time (sec): 12.49 - samples/sec: 1790.23 - lr: 0.000029 - momentum: 0.000000
2023-10-17 08:46:25,489 ----------------------------------------------------------------------------------------------------
2023-10-17 08:46:25,489 EPOCH 1 done: loss 1.0898 - lr: 0.000029
2023-10-17 08:46:26,237 DEV : loss 0.2100459337234497 - f1-score (micro avg) 0.7394
2023-10-17 08:46:26,242 saving best model
2023-10-17 08:46:26,590 ----------------------------------------------------------------------------------------------------
2023-10-17 08:46:27,824 epoch 2 - iter 27/275 - loss 0.17970721 - time (sec): 1.23 - samples/sec: 1835.24 - lr: 0.000030 - momentum: 0.000000
2023-10-17 08:46:29,079 epoch 2 - iter 54/275 - loss 0.18007533 - time (sec): 2.49 - samples/sec: 1801.59 - lr: 0.000029 - momentum: 0.000000
2023-10-17 08:46:30,296 epoch 2 - iter 81/275 - loss 0.18204146 - time (sec): 3.70 - samples/sec: 1799.96 - lr: 0.000029 - momentum: 0.000000
2023-10-17 08:46:31,517 epoch 2 - iter 108/275 - loss 0.18586872 - time (sec): 4.93 - samples/sec: 1836.50 - lr: 0.000029 - momentum: 0.000000
2023-10-17 08:46:32,804 epoch 2 - iter 135/275 - loss 0.18371927 - time (sec): 6.21 - samples/sec: 1805.97 - lr: 0.000028 - momentum: 0.000000
2023-10-17 08:46:34,029 epoch 2 - iter 162/275 - loss 0.18284554 - time (sec): 7.44 - samples/sec: 1819.86 - lr: 0.000028 - momentum: 0.000000
2023-10-17 08:46:35,293 epoch 2 - iter 189/275 - loss 0.17931458 - time (sec): 8.70 - samples/sec: 1810.73 - lr: 0.000028 - momentum: 0.000000
2023-10-17 08:46:36,602 epoch 2 - iter 216/275 - loss 0.18162776 - time (sec): 10.01 - samples/sec: 1831.89 - lr: 0.000027 - momentum: 0.000000
2023-10-17 08:46:37,845 epoch 2 - iter 243/275 - loss 0.17965500 - time (sec): 11.25 - samples/sec: 1822.02 - lr: 0.000027 - momentum: 0.000000
2023-10-17 08:46:39,081 epoch 2 - iter 270/275 - loss 0.17595326 - time (sec): 12.49 - samples/sec: 1797.58 - lr: 0.000027 - momentum: 0.000000
2023-10-17 08:46:39,300 ----------------------------------------------------------------------------------------------------
2023-10-17 08:46:39,300 EPOCH 2 done: loss 0.1765 - lr: 0.000027
2023-10-17 08:46:39,946 DEV : loss 0.20042622089385986 - f1-score (micro avg) 0.7796
2023-10-17 08:46:39,951 saving best model
2023-10-17 08:46:40,414 ----------------------------------------------------------------------------------------------------
2023-10-17 08:46:41,662 epoch 3 - iter 27/275 - loss 0.12635593 - time (sec): 1.25 - samples/sec: 1690.00 - lr: 0.000026 - momentum: 0.000000
2023-10-17 08:46:42,881 epoch 3 - iter 54/275 - loss 0.09902783 - time (sec): 2.46 - samples/sec: 1597.56 - lr: 0.000026 - momentum: 0.000000
2023-10-17 08:46:44,101 epoch 3 - iter 81/275 - loss 0.09432280 - time (sec): 3.68 - samples/sec: 1680.69 - lr: 0.000026 - momentum: 0.000000
2023-10-17 08:46:45,347 epoch 3 - iter 108/275 - loss 0.10335773 - time (sec): 4.93 - samples/sec: 1757.59 - lr: 0.000025 - momentum: 0.000000
2023-10-17 08:46:46,572 epoch 3 - iter 135/275 - loss 0.09804934 - time (sec): 6.16 - samples/sec: 1747.46 - lr: 0.000025 - momentum: 0.000000
2023-10-17 08:46:47,787 epoch 3 - iter 162/275 - loss 0.09737870 - time (sec): 7.37 - samples/sec: 1775.22 - lr: 0.000025 - momentum: 0.000000
2023-10-17 08:46:49,052 epoch 3 - iter 189/275 - loss 0.09737990 - time (sec): 8.64 - samples/sec: 1765.38 - lr: 0.000024 - momentum: 0.000000
2023-10-17 08:46:50,289 epoch 3 - iter 216/275 - loss 0.10009286 - time (sec): 9.87 - samples/sec: 1777.82 - lr: 0.000024 - momentum: 0.000000
2023-10-17 08:46:51,552 epoch 3 - iter 243/275 - loss 0.10060457 - time (sec): 11.14 - samples/sec: 1789.43 - lr: 0.000024 - momentum: 0.000000
2023-10-17 08:46:52,790 epoch 3 - iter 270/275 - loss 0.10670080 - time (sec): 12.37 - samples/sec: 1800.99 - lr: 0.000023 - momentum: 0.000000
2023-10-17 08:46:53,026 ----------------------------------------------------------------------------------------------------
2023-10-17 08:46:53,026 EPOCH 3 done: loss 0.1055 - lr: 0.000023
2023-10-17 08:46:53,665 DEV : loss 0.14525996148586273 - f1-score (micro avg) 0.852
2023-10-17 08:46:53,670 saving best model
2023-10-17 08:46:54,106 ----------------------------------------------------------------------------------------------------
2023-10-17 08:46:55,432 epoch 4 - iter 27/275 - loss 0.07663780 - time (sec): 1.32 - samples/sec: 1604.50 - lr: 0.000023 - momentum: 0.000000
2023-10-17 08:46:56,710 epoch 4 - iter 54/275 - loss 0.05673337 - time (sec): 2.60 - samples/sec: 1664.79 - lr: 0.000023 - momentum: 0.000000
2023-10-17 08:46:57,963 epoch 4 - iter 81/275 - loss 0.08993934 - time (sec): 3.85 - samples/sec: 1680.69 - lr: 0.000022 - momentum: 0.000000
2023-10-17 08:46:59,202 epoch 4 - iter 108/275 - loss 0.08785841 - time (sec): 5.09 - samples/sec: 1712.71 - lr: 0.000022 - momentum: 0.000000
2023-10-17 08:47:00,555 epoch 4 - iter 135/275 - loss 0.08309543 - time (sec): 6.45 - samples/sec: 1682.77 - lr: 0.000022 - momentum: 0.000000
2023-10-17 08:47:01,879 epoch 4 - iter 162/275 - loss 0.08392456 - time (sec): 7.77 - samples/sec: 1709.62 - lr: 0.000021 - momentum: 0.000000
2023-10-17 08:47:03,101 epoch 4 - iter 189/275 - loss 0.08150298 - time (sec): 8.99 - samples/sec: 1713.53 - lr: 0.000021 - momentum: 0.000000
2023-10-17 08:47:04,325 epoch 4 - iter 216/275 - loss 0.08090320 - time (sec): 10.22 - samples/sec: 1750.66 - lr: 0.000021 - momentum: 0.000000
2023-10-17 08:47:05,557 epoch 4 - iter 243/275 - loss 0.08222903 - time (sec): 11.45 - samples/sec: 1756.15 - lr: 0.000020 - momentum: 0.000000
2023-10-17 08:47:06,785 epoch 4 - iter 270/275 - loss 0.08283375 - time (sec): 12.68 - samples/sec: 1757.16 - lr: 0.000020 - momentum: 0.000000
2023-10-17 08:47:07,008 ----------------------------------------------------------------------------------------------------
2023-10-17 08:47:07,008 EPOCH 4 done: loss 0.0815 - lr: 0.000020
2023-10-17 08:47:07,649 DEV : loss 0.16326691210269928 - f1-score (micro avg) 0.8735
2023-10-17 08:47:07,654 saving best model
2023-10-17 08:47:08,086 ----------------------------------------------------------------------------------------------------
2023-10-17 08:47:09,299 epoch 5 - iter 27/275 - loss 0.11167857 - time (sec): 1.21 - samples/sec: 2006.63 - lr: 0.000020 - momentum: 0.000000
2023-10-17 08:47:10,524 epoch 5 - iter 54/275 - loss 0.08355736 - time (sec): 2.44 - samples/sec: 1883.54 - lr: 0.000019 - momentum: 0.000000
2023-10-17 08:47:11,747 epoch 5 - iter 81/275 - loss 0.06609019 - time (sec): 3.66 - samples/sec: 1807.76 - lr: 0.000019 - momentum: 0.000000
2023-10-17 08:47:13,030 epoch 5 - iter 108/275 - loss 0.07976555 - time (sec): 4.94 - samples/sec: 1735.78 - lr: 0.000019 - momentum: 0.000000
2023-10-17 08:47:14,249 epoch 5 - iter 135/275 - loss 0.07909386 - time (sec): 6.16 - samples/sec: 1767.66 - lr: 0.000018 - momentum: 0.000000
2023-10-17 08:47:15,546 epoch 5 - iter 162/275 - loss 0.07437644 - time (sec): 7.46 - samples/sec: 1760.74 - lr: 0.000018 - momentum: 0.000000
2023-10-17 08:47:16,766 epoch 5 - iter 189/275 - loss 0.07040829 - time (sec): 8.68 - samples/sec: 1780.58 - lr: 0.000018 - momentum: 0.000000
2023-10-17 08:47:17,986 epoch 5 - iter 216/275 - loss 0.06799222 - time (sec): 9.90 - samples/sec: 1811.00 - lr: 0.000017 - momentum: 0.000000
2023-10-17 08:47:19,246 epoch 5 - iter 243/275 - loss 0.06299278 - time (sec): 11.16 - samples/sec: 1805.20 - lr: 0.000017 - momentum: 0.000000
2023-10-17 08:47:20,482 epoch 5 - iter 270/275 - loss 0.06246345 - time (sec): 12.40 - samples/sec: 1806.75 - lr: 0.000017 - momentum: 0.000000
2023-10-17 08:47:20,713 ----------------------------------------------------------------------------------------------------
2023-10-17 08:47:20,713 EPOCH 5 done: loss 0.0624 - lr: 0.000017
2023-10-17 08:47:21,345 DEV : loss 0.16908515989780426 - f1-score (micro avg) 0.8803
2023-10-17 08:47:21,350 saving best model
2023-10-17 08:47:21,779 ----------------------------------------------------------------------------------------------------
2023-10-17 08:47:22,983 epoch 6 - iter 27/275 - loss 0.05658757 - time (sec): 1.20 - samples/sec: 1666.25 - lr: 0.000016 - momentum: 0.000000
2023-10-17 08:47:24,261 epoch 6 - iter 54/275 - loss 0.05408551 - time (sec): 2.48 - samples/sec: 1693.91 - lr: 0.000016 - momentum: 0.000000
2023-10-17 08:47:25,553 epoch 6 - iter 81/275 - loss 0.06296102 - time (sec): 3.77 - samples/sec: 1725.26 - lr: 0.000016 - momentum: 0.000000
2023-10-17 08:47:26,766 epoch 6 - iter 108/275 - loss 0.06062024 - time (sec): 4.98 - samples/sec: 1734.48 - lr: 0.000015 - momentum: 0.000000
2023-10-17 08:47:27,990 epoch 6 - iter 135/275 - loss 0.05487807 - time (sec): 6.21 - samples/sec: 1761.68 - lr: 0.000015 - momentum: 0.000000
2023-10-17 08:47:29,247 epoch 6 - iter 162/275 - loss 0.04790647 - time (sec): 7.46 - samples/sec: 1758.75 - lr: 0.000015 - momentum: 0.000000
2023-10-17 08:47:30,503 epoch 6 - iter 189/275 - loss 0.04791047 - time (sec): 8.72 - samples/sec: 1759.19 - lr: 0.000014 - momentum: 0.000000
2023-10-17 08:47:31,762 epoch 6 - iter 216/275 - loss 0.04903596 - time (sec): 9.98 - samples/sec: 1765.27 - lr: 0.000014 - momentum: 0.000000
2023-10-17 08:47:33,033 epoch 6 - iter 243/275 - loss 0.05332217 - time (sec): 11.25 - samples/sec: 1784.97 - lr: 0.000014 - momentum: 0.000000
2023-10-17 08:47:34,255 epoch 6 - iter 270/275 - loss 0.05091613 - time (sec): 12.47 - samples/sec: 1787.09 - lr: 0.000013 - momentum: 0.000000
2023-10-17 08:47:34,480 ----------------------------------------------------------------------------------------------------
2023-10-17 08:47:34,481 EPOCH 6 done: loss 0.0501 - lr: 0.000013
2023-10-17 08:47:35,157 DEV : loss 0.17602381110191345 - f1-score (micro avg) 0.8822
2023-10-17 08:47:35,167 saving best model
2023-10-17 08:47:35,710 ----------------------------------------------------------------------------------------------------
2023-10-17 08:47:37,152 epoch 7 - iter 27/275 - loss 0.08149249 - time (sec): 1.44 - samples/sec: 1431.14 - lr: 0.000013 - momentum: 0.000000
2023-10-17 08:47:38,645 epoch 7 - iter 54/275 - loss 0.05422051 - time (sec): 2.93 - samples/sec: 1503.81 - lr: 0.000013 - momentum: 0.000000
2023-10-17 08:47:40,059 epoch 7 - iter 81/275 - loss 0.06668234 - time (sec): 4.35 - samples/sec: 1541.04 - lr: 0.000012 - momentum: 0.000000
2023-10-17 08:47:41,325 epoch 7 - iter 108/275 - loss 0.05693708 - time (sec): 5.61 - samples/sec: 1591.18 - lr: 0.000012 - momentum: 0.000000
2023-10-17 08:47:42,586 epoch 7 - iter 135/275 - loss 0.04821359 - time (sec): 6.87 - samples/sec: 1607.07 - lr: 0.000012 - momentum: 0.000000
2023-10-17 08:47:43,887 epoch 7 - iter 162/275 - loss 0.04377735 - time (sec): 8.18 - samples/sec: 1638.57 - lr: 0.000011 - momentum: 0.000000
2023-10-17 08:47:45,191 epoch 7 - iter 189/275 - loss 0.04185787 - time (sec): 9.48 - samples/sec: 1673.78 - lr: 0.000011 - momentum: 0.000000
2023-10-17 08:47:46,422 epoch 7 - iter 216/275 - loss 0.04208472 - time (sec): 10.71 - samples/sec: 1678.53 - lr: 0.000011 - momentum: 0.000000
2023-10-17 08:47:47,681 epoch 7 - iter 243/275 - loss 0.04099767 - time (sec): 11.97 - samples/sec: 1674.21 - lr: 0.000010 - momentum: 0.000000
2023-10-17 08:47:49,017 epoch 7 - iter 270/275 - loss 0.04134562 - time (sec): 13.31 - samples/sec: 1678.74 - lr: 0.000010 - momentum: 0.000000
2023-10-17 08:47:49,256 ----------------------------------------------------------------------------------------------------
2023-10-17 08:47:49,256 EPOCH 7 done: loss 0.0406 - lr: 0.000010
2023-10-17 08:47:49,919 DEV : loss 0.18893340229988098 - f1-score (micro avg) 0.8723
2023-10-17 08:47:49,923 ----------------------------------------------------------------------------------------------------
2023-10-17 08:47:51,156 epoch 8 - iter 27/275 - loss 0.00906963 - time (sec): 1.23 - samples/sec: 1960.66 - lr: 0.000010 - momentum: 0.000000
2023-10-17 08:47:52,384 epoch 8 - iter 54/275 - loss 0.02783979 - time (sec): 2.46 - samples/sec: 1887.13 - lr: 0.000009 - momentum: 0.000000
2023-10-17 08:47:53,638 epoch 8 - iter 81/275 - loss 0.02990417 - time (sec): 3.71 - samples/sec: 1848.33 - lr: 0.000009 - momentum: 0.000000
2023-10-17 08:47:54,853 epoch 8 - iter 108/275 - loss 0.02627080 - time (sec): 4.93 - samples/sec: 1826.83 - lr: 0.000009 - momentum: 0.000000
2023-10-17 08:47:56,119 epoch 8 - iter 135/275 - loss 0.02263615 - time (sec): 6.19 - samples/sec: 1866.59 - lr: 0.000008 - momentum: 0.000000
2023-10-17 08:47:57,325 epoch 8 - iter 162/275 - loss 0.02311608 - time (sec): 7.40 - samples/sec: 1842.73 - lr: 0.000008 - momentum: 0.000000
2023-10-17 08:47:58,559 epoch 8 - iter 189/275 - loss 0.02151264 - time (sec): 8.63 - samples/sec: 1843.38 - lr: 0.000008 - momentum: 0.000000
2023-10-17 08:47:59,779 epoch 8 - iter 216/275 - loss 0.02896352 - time (sec): 9.85 - samples/sec: 1837.29 - lr: 0.000007 - momentum: 0.000000
2023-10-17 08:48:01,054 epoch 8 - iter 243/275 - loss 0.02972546 - time (sec): 11.13 - samples/sec: 1825.46 - lr: 0.000007 - momentum: 0.000000
2023-10-17 08:48:02,272 epoch 8 - iter 270/275 - loss 0.02940927 - time (sec): 12.35 - samples/sec: 1816.05 - lr: 0.000007 - momentum: 0.000000
2023-10-17 08:48:02,503 ----------------------------------------------------------------------------------------------------
2023-10-17 08:48:02,503 EPOCH 8 done: loss 0.0301 - lr: 0.000007
2023-10-17 08:48:03,194 DEV : loss 0.17884737253189087 - f1-score (micro avg) 0.8905
2023-10-17 08:48:03,198 saving best model
2023-10-17 08:48:03,654 ----------------------------------------------------------------------------------------------------
2023-10-17 08:48:04,846 epoch 9 - iter 27/275 - loss 0.03662375 - time (sec): 1.19 - samples/sec: 2005.60 - lr: 0.000006 - momentum: 0.000000
2023-10-17 08:48:06,020 epoch 9 - iter 54/275 - loss 0.02152339 - time (sec): 2.36 - samples/sec: 2035.85 - lr: 0.000006 - momentum: 0.000000
2023-10-17 08:48:07,185 epoch 9 - iter 81/275 - loss 0.02197956 - time (sec): 3.53 - samples/sec: 1949.37 - lr: 0.000006 - momentum: 0.000000
2023-10-17 08:48:08,371 epoch 9 - iter 108/275 - loss 0.02212008 - time (sec): 4.72 - samples/sec: 1896.48 - lr: 0.000005 - momentum: 0.000000
2023-10-17 08:48:09,590 epoch 9 - iter 135/275 - loss 0.02084789 - time (sec): 5.93 - samples/sec: 1851.36 - lr: 0.000005 - momentum: 0.000000
2023-10-17 08:48:10,800 epoch 9 - iter 162/275 - loss 0.02603522 - time (sec): 7.14 - samples/sec: 1858.08 - lr: 0.000005 - momentum: 0.000000
2023-10-17 08:48:12,005 epoch 9 - iter 189/275 - loss 0.02658731 - time (sec): 8.35 - samples/sec: 1857.87 - lr: 0.000004 - momentum: 0.000000
2023-10-17 08:48:13,260 epoch 9 - iter 216/275 - loss 0.02480973 - time (sec): 9.60 - samples/sec: 1835.23 - lr: 0.000004 - momentum: 0.000000
2023-10-17 08:48:14,480 epoch 9 - iter 243/275 - loss 0.02327959 - time (sec): 10.82 - samples/sec: 1840.65 - lr: 0.000004 - momentum: 0.000000
2023-10-17 08:48:15,704 epoch 9 - iter 270/275 - loss 0.02433170 - time (sec): 12.05 - samples/sec: 1849.42 - lr: 0.000003 - momentum: 0.000000
2023-10-17 08:48:15,935 ----------------------------------------------------------------------------------------------------
2023-10-17 08:48:15,935 EPOCH 9 done: loss 0.0239 - lr: 0.000003
2023-10-17 08:48:16,663 DEV : loss 0.1809743344783783 - f1-score (micro avg) 0.8884
2023-10-17 08:48:16,668 ----------------------------------------------------------------------------------------------------
2023-10-17 08:48:17,828 epoch 10 - iter 27/275 - loss 0.01640093 - time (sec): 1.16 - samples/sec: 1842.79 - lr: 0.000003 - momentum: 0.000000
2023-10-17 08:48:19,001 epoch 10 - iter 54/275 - loss 0.01996468 - time (sec): 2.33 - samples/sec: 1952.99 - lr: 0.000003 - momentum: 0.000000
2023-10-17 08:48:20,159 epoch 10 - iter 81/275 - loss 0.02308692 - time (sec): 3.49 - samples/sec: 1934.94 - lr: 0.000002 - momentum: 0.000000
2023-10-17 08:48:21,308 epoch 10 - iter 108/275 - loss 0.01854556 - time (sec): 4.64 - samples/sec: 1864.98 - lr: 0.000002 - momentum: 0.000000
2023-10-17 08:48:22,472 epoch 10 - iter 135/275 - loss 0.01695391 - time (sec): 5.80 - samples/sec: 1868.82 - lr: 0.000002 - momentum: 0.000000
2023-10-17 08:48:23,639 epoch 10 - iter 162/275 - loss 0.02042256 - time (sec): 6.97 - samples/sec: 1872.87 - lr: 0.000001 - momentum: 0.000000
2023-10-17 08:48:24,861 epoch 10 - iter 189/275 - loss 0.01872876 - time (sec): 8.19 - samples/sec: 1894.03 - lr: 0.000001 - momentum: 0.000000
2023-10-17 08:48:26,090 epoch 10 - iter 216/275 - loss 0.02260026 - time (sec): 9.42 - samples/sec: 1892.11 - lr: 0.000001 - momentum: 0.000000
2023-10-17 08:48:27,338 epoch 10 - iter 243/275 - loss 0.02105999 - time (sec): 10.67 - samples/sec: 1887.12 - lr: 0.000000 - momentum: 0.000000
2023-10-17 08:48:28,558 epoch 10 - iter 270/275 - loss 0.01967608 - time (sec): 11.89 - samples/sec: 1879.11 - lr: 0.000000 - momentum: 0.000000
2023-10-17 08:48:28,795 ----------------------------------------------------------------------------------------------------
2023-10-17 08:48:28,795 EPOCH 10 done: loss 0.0199 - lr: 0.000000
2023-10-17 08:48:29,444 DEV : loss 0.18379689753055573 - f1-score (micro avg) 0.8828
2023-10-17 08:48:29,797 ----------------------------------------------------------------------------------------------------
2023-10-17 08:48:29,799 Loading model from best epoch ...
2023-10-17 08:48:31,311 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
2023-10-17 08:48:31,953
Results:
- F-score (micro) 0.9062
- F-score (macro) 0.6759
- Accuracy 0.8469
By class:
precision recall f1-score support
scope 0.9023 0.8920 0.8971 176
pers 0.9680 0.9453 0.9565 128
work 0.8533 0.8649 0.8591 74
loc 1.0000 0.5000 0.6667 2
object 0.0000 0.0000 0.0000 2
micro avg 0.9147 0.8979 0.9062 382
macro avg 0.7447 0.6404 0.6759 382
weighted avg 0.9106 0.8979 0.9038 382
2023-10-17 08:48:31,954 ----------------------------------------------------------------------------------------------------
|