File size: 24,172 Bytes
7e591bc |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 |
2023-10-23 15:51:57,497 ----------------------------------------------------------------------------------------------------
2023-10-23 15:51:57,498 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 15:51:57,498 ----------------------------------------------------------------------------------------------------
2023-10-23 15:51:57,498 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 15:51:57,498 ----------------------------------------------------------------------------------------------------
2023-10-23 15:51:57,499 Train: 1100 sentences
2023-10-23 15:51:57,499 (train_with_dev=False, train_with_test=False)
2023-10-23 15:51:57,499 ----------------------------------------------------------------------------------------------------
2023-10-23 15:51:57,499 Training Params:
2023-10-23 15:51:57,499 - learning_rate: "5e-05"
2023-10-23 15:51:57,499 - mini_batch_size: "4"
2023-10-23 15:51:57,499 - max_epochs: "10"
2023-10-23 15:51:57,499 - shuffle: "True"
2023-10-23 15:51:57,499 ----------------------------------------------------------------------------------------------------
2023-10-23 15:51:57,499 Plugins:
2023-10-23 15:51:57,499 - TensorboardLogger
2023-10-23 15:51:57,499 - LinearScheduler | warmup_fraction: '0.1'
2023-10-23 15:51:57,499 ----------------------------------------------------------------------------------------------------
2023-10-23 15:51:57,499 Final evaluation on model from best epoch (best-model.pt)
2023-10-23 15:51:57,499 - metric: "('micro avg', 'f1-score')"
2023-10-23 15:51:57,499 ----------------------------------------------------------------------------------------------------
2023-10-23 15:51:57,499 Computation:
2023-10-23 15:51:57,499 - compute on device: cuda:0
2023-10-23 15:51:57,499 - embedding storage: none
2023-10-23 15:51:57,499 ----------------------------------------------------------------------------------------------------
2023-10-23 15:51:57,499 Model training base path: "hmbench-ajmc/de-dbmdz/bert-base-historic-multilingual-64k-td-cased-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-5"
2023-10-23 15:51:57,499 ----------------------------------------------------------------------------------------------------
2023-10-23 15:51:57,499 ----------------------------------------------------------------------------------------------------
2023-10-23 15:51:57,499 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-23 15:51:58,821 epoch 1 - iter 27/275 - loss 2.64810433 - time (sec): 1.32 - samples/sec: 1353.57 - lr: 0.000005 - momentum: 0.000000
2023-10-23 15:52:00,167 epoch 1 - iter 54/275 - loss 1.83490599 - time (sec): 2.67 - samples/sec: 1526.70 - lr: 0.000010 - momentum: 0.000000
2023-10-23 15:52:01,507 epoch 1 - iter 81/275 - loss 1.48707717 - time (sec): 4.01 - samples/sec: 1615.59 - lr: 0.000015 - momentum: 0.000000
2023-10-23 15:52:02,867 epoch 1 - iter 108/275 - loss 1.25898807 - time (sec): 5.37 - samples/sec: 1654.13 - lr: 0.000019 - momentum: 0.000000
2023-10-23 15:52:04,220 epoch 1 - iter 135/275 - loss 1.09035369 - time (sec): 6.72 - samples/sec: 1659.42 - lr: 0.000024 - momentum: 0.000000
2023-10-23 15:52:05,547 epoch 1 - iter 162/275 - loss 0.95463488 - time (sec): 8.05 - samples/sec: 1671.29 - lr: 0.000029 - momentum: 0.000000
2023-10-23 15:52:06,938 epoch 1 - iter 189/275 - loss 0.86593875 - time (sec): 9.44 - samples/sec: 1651.35 - lr: 0.000034 - momentum: 0.000000
2023-10-23 15:52:08,332 epoch 1 - iter 216/275 - loss 0.77581041 - time (sec): 10.83 - samples/sec: 1668.19 - lr: 0.000039 - momentum: 0.000000
2023-10-23 15:52:09,731 epoch 1 - iter 243/275 - loss 0.71487665 - time (sec): 12.23 - samples/sec: 1660.58 - lr: 0.000044 - momentum: 0.000000
2023-10-23 15:52:11,121 epoch 1 - iter 270/275 - loss 0.66787877 - time (sec): 13.62 - samples/sec: 1644.92 - lr: 0.000049 - momentum: 0.000000
2023-10-23 15:52:11,374 ----------------------------------------------------------------------------------------------------
2023-10-23 15:52:11,374 EPOCH 1 done: loss 0.6618 - lr: 0.000049
2023-10-23 15:52:11,794 DEV : loss 0.179626926779747 - f1-score (micro avg) 0.7718
2023-10-23 15:52:11,799 saving best model
2023-10-23 15:52:12,196 ----------------------------------------------------------------------------------------------------
2023-10-23 15:52:13,602 epoch 2 - iter 27/275 - loss 0.19354437 - time (sec): 1.40 - samples/sec: 1809.51 - lr: 0.000049 - momentum: 0.000000
2023-10-23 15:52:14,993 epoch 2 - iter 54/275 - loss 0.20163780 - time (sec): 2.80 - samples/sec: 1639.26 - lr: 0.000049 - momentum: 0.000000
2023-10-23 15:52:16,380 epoch 2 - iter 81/275 - loss 0.17039530 - time (sec): 4.18 - samples/sec: 1591.50 - lr: 0.000048 - momentum: 0.000000
2023-10-23 15:52:17,774 epoch 2 - iter 108/275 - loss 0.17475614 - time (sec): 5.58 - samples/sec: 1543.31 - lr: 0.000048 - momentum: 0.000000
2023-10-23 15:52:19,204 epoch 2 - iter 135/275 - loss 0.16444749 - time (sec): 7.01 - samples/sec: 1539.34 - lr: 0.000047 - momentum: 0.000000
2023-10-23 15:52:20,781 epoch 2 - iter 162/275 - loss 0.16649597 - time (sec): 8.58 - samples/sec: 1501.31 - lr: 0.000047 - momentum: 0.000000
2023-10-23 15:52:22,208 epoch 2 - iter 189/275 - loss 0.15541591 - time (sec): 10.01 - samples/sec: 1510.50 - lr: 0.000046 - momentum: 0.000000
2023-10-23 15:52:23,629 epoch 2 - iter 216/275 - loss 0.15842844 - time (sec): 11.43 - samples/sec: 1527.32 - lr: 0.000046 - momentum: 0.000000
2023-10-23 15:52:25,043 epoch 2 - iter 243/275 - loss 0.15889443 - time (sec): 12.85 - samples/sec: 1538.23 - lr: 0.000045 - momentum: 0.000000
2023-10-23 15:52:26,452 epoch 2 - iter 270/275 - loss 0.15806273 - time (sec): 14.25 - samples/sec: 1566.28 - lr: 0.000045 - momentum: 0.000000
2023-10-23 15:52:26,716 ----------------------------------------------------------------------------------------------------
2023-10-23 15:52:26,717 EPOCH 2 done: loss 0.1588 - lr: 0.000045
2023-10-23 15:52:27,252 DEV : loss 0.16758960485458374 - f1-score (micro avg) 0.8019
2023-10-23 15:52:27,257 saving best model
2023-10-23 15:52:27,803 ----------------------------------------------------------------------------------------------------
2023-10-23 15:52:29,170 epoch 3 - iter 27/275 - loss 0.12366499 - time (sec): 1.36 - samples/sec: 1712.37 - lr: 0.000044 - momentum: 0.000000
2023-10-23 15:52:30,511 epoch 3 - iter 54/275 - loss 0.10372222 - time (sec): 2.70 - samples/sec: 1680.75 - lr: 0.000043 - momentum: 0.000000
2023-10-23 15:52:31,880 epoch 3 - iter 81/275 - loss 0.10568938 - time (sec): 4.07 - samples/sec: 1658.05 - lr: 0.000043 - momentum: 0.000000
2023-10-23 15:52:33,251 epoch 3 - iter 108/275 - loss 0.10353497 - time (sec): 5.44 - samples/sec: 1666.27 - lr: 0.000042 - momentum: 0.000000
2023-10-23 15:52:34,612 epoch 3 - iter 135/275 - loss 0.09569370 - time (sec): 6.81 - samples/sec: 1672.37 - lr: 0.000042 - momentum: 0.000000
2023-10-23 15:52:35,955 epoch 3 - iter 162/275 - loss 0.09029367 - time (sec): 8.15 - samples/sec: 1669.38 - lr: 0.000041 - momentum: 0.000000
2023-10-23 15:52:37,322 epoch 3 - iter 189/275 - loss 0.09163925 - time (sec): 9.52 - samples/sec: 1663.19 - lr: 0.000041 - momentum: 0.000000
2023-10-23 15:52:38,707 epoch 3 - iter 216/275 - loss 0.09751995 - time (sec): 10.90 - samples/sec: 1648.00 - lr: 0.000040 - momentum: 0.000000
2023-10-23 15:52:40,073 epoch 3 - iter 243/275 - loss 0.10012864 - time (sec): 12.27 - samples/sec: 1656.28 - lr: 0.000040 - momentum: 0.000000
2023-10-23 15:52:41,385 epoch 3 - iter 270/275 - loss 0.09615579 - time (sec): 13.58 - samples/sec: 1640.19 - lr: 0.000039 - momentum: 0.000000
2023-10-23 15:52:41,630 ----------------------------------------------------------------------------------------------------
2023-10-23 15:52:41,630 EPOCH 3 done: loss 0.0982 - lr: 0.000039
2023-10-23 15:52:42,169 DEV : loss 0.1806371957063675 - f1-score (micro avg) 0.84
2023-10-23 15:52:42,174 saving best model
2023-10-23 15:52:42,723 ----------------------------------------------------------------------------------------------------
2023-10-23 15:52:44,124 epoch 4 - iter 27/275 - loss 0.11872071 - time (sec): 1.40 - samples/sec: 1659.62 - lr: 0.000038 - momentum: 0.000000
2023-10-23 15:52:45,466 epoch 4 - iter 54/275 - loss 0.08762465 - time (sec): 2.74 - samples/sec: 1721.30 - lr: 0.000038 - momentum: 0.000000
2023-10-23 15:52:46,772 epoch 4 - iter 81/275 - loss 0.08315257 - time (sec): 4.04 - samples/sec: 1659.82 - lr: 0.000037 - momentum: 0.000000
2023-10-23 15:52:48,068 epoch 4 - iter 108/275 - loss 0.07259559 - time (sec): 5.34 - samples/sec: 1646.58 - lr: 0.000037 - momentum: 0.000000
2023-10-23 15:52:49,413 epoch 4 - iter 135/275 - loss 0.06971542 - time (sec): 6.69 - samples/sec: 1682.02 - lr: 0.000036 - momentum: 0.000000
2023-10-23 15:52:50,757 epoch 4 - iter 162/275 - loss 0.07403488 - time (sec): 8.03 - samples/sec: 1701.06 - lr: 0.000036 - momentum: 0.000000
2023-10-23 15:52:52,181 epoch 4 - iter 189/275 - loss 0.06823022 - time (sec): 9.45 - samples/sec: 1682.76 - lr: 0.000035 - momentum: 0.000000
2023-10-23 15:52:53,576 epoch 4 - iter 216/275 - loss 0.07045628 - time (sec): 10.85 - samples/sec: 1662.31 - lr: 0.000035 - momentum: 0.000000
2023-10-23 15:52:54,990 epoch 4 - iter 243/275 - loss 0.06840953 - time (sec): 12.26 - samples/sec: 1624.01 - lr: 0.000034 - momentum: 0.000000
2023-10-23 15:52:56,380 epoch 4 - iter 270/275 - loss 0.06924410 - time (sec): 13.65 - samples/sec: 1638.06 - lr: 0.000034 - momentum: 0.000000
2023-10-23 15:52:56,636 ----------------------------------------------------------------------------------------------------
2023-10-23 15:52:56,636 EPOCH 4 done: loss 0.0694 - lr: 0.000034
2023-10-23 15:52:57,168 DEV : loss 0.14905601739883423 - f1-score (micro avg) 0.8714
2023-10-23 15:52:57,173 saving best model
2023-10-23 15:52:57,715 ----------------------------------------------------------------------------------------------------
2023-10-23 15:52:59,088 epoch 5 - iter 27/275 - loss 0.04803966 - time (sec): 1.37 - samples/sec: 1701.69 - lr: 0.000033 - momentum: 0.000000
2023-10-23 15:53:00,481 epoch 5 - iter 54/275 - loss 0.06866180 - time (sec): 2.76 - samples/sec: 1631.46 - lr: 0.000032 - momentum: 0.000000
2023-10-23 15:53:01,893 epoch 5 - iter 81/275 - loss 0.05926000 - time (sec): 4.17 - samples/sec: 1619.86 - lr: 0.000032 - momentum: 0.000000
2023-10-23 15:53:03,281 epoch 5 - iter 108/275 - loss 0.04963370 - time (sec): 5.56 - samples/sec: 1603.80 - lr: 0.000031 - momentum: 0.000000
2023-10-23 15:53:04,688 epoch 5 - iter 135/275 - loss 0.05658941 - time (sec): 6.97 - samples/sec: 1621.04 - lr: 0.000031 - momentum: 0.000000
2023-10-23 15:53:06,093 epoch 5 - iter 162/275 - loss 0.05275609 - time (sec): 8.37 - samples/sec: 1599.71 - lr: 0.000030 - momentum: 0.000000
2023-10-23 15:53:07,500 epoch 5 - iter 189/275 - loss 0.05027437 - time (sec): 9.78 - samples/sec: 1599.91 - lr: 0.000030 - momentum: 0.000000
2023-10-23 15:53:08,924 epoch 5 - iter 216/275 - loss 0.04932611 - time (sec): 11.21 - samples/sec: 1596.52 - lr: 0.000029 - momentum: 0.000000
2023-10-23 15:53:10,325 epoch 5 - iter 243/275 - loss 0.05382881 - time (sec): 12.61 - samples/sec: 1578.66 - lr: 0.000029 - momentum: 0.000000
2023-10-23 15:53:11,739 epoch 5 - iter 270/275 - loss 0.05098833 - time (sec): 14.02 - samples/sec: 1587.32 - lr: 0.000028 - momentum: 0.000000
2023-10-23 15:53:11,992 ----------------------------------------------------------------------------------------------------
2023-10-23 15:53:11,992 EPOCH 5 done: loss 0.0507 - lr: 0.000028
2023-10-23 15:53:12,527 DEV : loss 0.1650022566318512 - f1-score (micro avg) 0.8724
2023-10-23 15:53:12,533 saving best model
2023-10-23 15:53:13,088 ----------------------------------------------------------------------------------------------------
2023-10-23 15:53:14,537 epoch 6 - iter 27/275 - loss 0.00692271 - time (sec): 1.45 - samples/sec: 1743.64 - lr: 0.000027 - momentum: 0.000000
2023-10-23 15:53:15,926 epoch 6 - iter 54/275 - loss 0.02989654 - time (sec): 2.83 - samples/sec: 1571.81 - lr: 0.000027 - momentum: 0.000000
2023-10-23 15:53:17,328 epoch 6 - iter 81/275 - loss 0.02792755 - time (sec): 4.24 - samples/sec: 1564.99 - lr: 0.000026 - momentum: 0.000000
2023-10-23 15:53:18,738 epoch 6 - iter 108/275 - loss 0.02765114 - time (sec): 5.65 - samples/sec: 1597.13 - lr: 0.000026 - momentum: 0.000000
2023-10-23 15:53:20,147 epoch 6 - iter 135/275 - loss 0.02524505 - time (sec): 7.06 - samples/sec: 1587.87 - lr: 0.000025 - momentum: 0.000000
2023-10-23 15:53:21,557 epoch 6 - iter 162/275 - loss 0.02776897 - time (sec): 8.47 - samples/sec: 1566.30 - lr: 0.000025 - momentum: 0.000000
2023-10-23 15:53:22,900 epoch 6 - iter 189/275 - loss 0.02839678 - time (sec): 9.81 - samples/sec: 1571.02 - lr: 0.000024 - momentum: 0.000000
2023-10-23 15:53:24,269 epoch 6 - iter 216/275 - loss 0.02881164 - time (sec): 11.18 - samples/sec: 1576.51 - lr: 0.000024 - momentum: 0.000000
2023-10-23 15:53:25,664 epoch 6 - iter 243/275 - loss 0.03172496 - time (sec): 12.57 - samples/sec: 1587.99 - lr: 0.000023 - momentum: 0.000000
2023-10-23 15:53:27,075 epoch 6 - iter 270/275 - loss 0.03269819 - time (sec): 13.98 - samples/sec: 1597.39 - lr: 0.000022 - momentum: 0.000000
2023-10-23 15:53:27,343 ----------------------------------------------------------------------------------------------------
2023-10-23 15:53:27,343 EPOCH 6 done: loss 0.0332 - lr: 0.000022
2023-10-23 15:53:27,884 DEV : loss 0.165008082985878 - f1-score (micro avg) 0.8825
2023-10-23 15:53:27,890 saving best model
2023-10-23 15:53:28,432 ----------------------------------------------------------------------------------------------------
2023-10-23 15:53:29,759 epoch 7 - iter 27/275 - loss 0.03004289 - time (sec): 1.32 - samples/sec: 1768.40 - lr: 0.000022 - momentum: 0.000000
2023-10-23 15:53:31,121 epoch 7 - iter 54/275 - loss 0.02755990 - time (sec): 2.69 - samples/sec: 1634.52 - lr: 0.000021 - momentum: 0.000000
2023-10-23 15:53:32,468 epoch 7 - iter 81/275 - loss 0.02623236 - time (sec): 4.03 - samples/sec: 1606.70 - lr: 0.000021 - momentum: 0.000000
2023-10-23 15:53:33,820 epoch 7 - iter 108/275 - loss 0.02598619 - time (sec): 5.38 - samples/sec: 1648.23 - lr: 0.000020 - momentum: 0.000000
2023-10-23 15:53:35,203 epoch 7 - iter 135/275 - loss 0.02416610 - time (sec): 6.77 - samples/sec: 1642.59 - lr: 0.000020 - momentum: 0.000000
2023-10-23 15:53:36,540 epoch 7 - iter 162/275 - loss 0.02452176 - time (sec): 8.10 - samples/sec: 1628.01 - lr: 0.000019 - momentum: 0.000000
2023-10-23 15:53:37,892 epoch 7 - iter 189/275 - loss 0.02358188 - time (sec): 9.46 - samples/sec: 1642.33 - lr: 0.000019 - momentum: 0.000000
2023-10-23 15:53:39,212 epoch 7 - iter 216/275 - loss 0.02191696 - time (sec): 10.78 - samples/sec: 1659.22 - lr: 0.000018 - momentum: 0.000000
2023-10-23 15:53:40,556 epoch 7 - iter 243/275 - loss 0.02403767 - time (sec): 12.12 - samples/sec: 1667.16 - lr: 0.000017 - momentum: 0.000000
2023-10-23 15:53:41,896 epoch 7 - iter 270/275 - loss 0.02337190 - time (sec): 13.46 - samples/sec: 1658.48 - lr: 0.000017 - momentum: 0.000000
2023-10-23 15:53:42,151 ----------------------------------------------------------------------------------------------------
2023-10-23 15:53:42,152 EPOCH 7 done: loss 0.0230 - lr: 0.000017
2023-10-23 15:53:42,689 DEV : loss 0.16212137043476105 - f1-score (micro avg) 0.8764
2023-10-23 15:53:42,694 ----------------------------------------------------------------------------------------------------
2023-10-23 15:53:44,021 epoch 8 - iter 27/275 - loss 0.03973778 - time (sec): 1.33 - samples/sec: 1641.32 - lr: 0.000016 - momentum: 0.000000
2023-10-23 15:53:45,394 epoch 8 - iter 54/275 - loss 0.02648907 - time (sec): 2.70 - samples/sec: 1695.49 - lr: 0.000016 - momentum: 0.000000
2023-10-23 15:53:46,757 epoch 8 - iter 81/275 - loss 0.02530064 - time (sec): 4.06 - samples/sec: 1649.28 - lr: 0.000015 - momentum: 0.000000
2023-10-23 15:53:48,129 epoch 8 - iter 108/275 - loss 0.02187305 - time (sec): 5.43 - samples/sec: 1687.18 - lr: 0.000015 - momentum: 0.000000
2023-10-23 15:53:49,477 epoch 8 - iter 135/275 - loss 0.01955675 - time (sec): 6.78 - samples/sec: 1696.35 - lr: 0.000014 - momentum: 0.000000
2023-10-23 15:53:50,832 epoch 8 - iter 162/275 - loss 0.01740010 - time (sec): 8.14 - samples/sec: 1717.17 - lr: 0.000014 - momentum: 0.000000
2023-10-23 15:53:52,186 epoch 8 - iter 189/275 - loss 0.01676100 - time (sec): 9.49 - samples/sec: 1688.69 - lr: 0.000013 - momentum: 0.000000
2023-10-23 15:53:53,535 epoch 8 - iter 216/275 - loss 0.01505490 - time (sec): 10.84 - samples/sec: 1664.98 - lr: 0.000012 - momentum: 0.000000
2023-10-23 15:53:54,906 epoch 8 - iter 243/275 - loss 0.01536436 - time (sec): 12.21 - samples/sec: 1652.09 - lr: 0.000012 - momentum: 0.000000
2023-10-23 15:53:56,238 epoch 8 - iter 270/275 - loss 0.01475368 - time (sec): 13.54 - samples/sec: 1647.04 - lr: 0.000011 - momentum: 0.000000
2023-10-23 15:53:56,480 ----------------------------------------------------------------------------------------------------
2023-10-23 15:53:56,480 EPOCH 8 done: loss 0.0154 - lr: 0.000011
2023-10-23 15:53:57,013 DEV : loss 0.17422381043434143 - f1-score (micro avg) 0.8873
2023-10-23 15:53:57,019 saving best model
2023-10-23 15:53:57,566 ----------------------------------------------------------------------------------------------------
2023-10-23 15:53:58,903 epoch 9 - iter 27/275 - loss 0.01095488 - time (sec): 1.34 - samples/sec: 1702.86 - lr: 0.000011 - momentum: 0.000000
2023-10-23 15:54:00,288 epoch 9 - iter 54/275 - loss 0.00940144 - time (sec): 2.72 - samples/sec: 1683.98 - lr: 0.000010 - momentum: 0.000000
2023-10-23 15:54:01,674 epoch 9 - iter 81/275 - loss 0.01506953 - time (sec): 4.11 - samples/sec: 1654.32 - lr: 0.000010 - momentum: 0.000000
2023-10-23 15:54:03,064 epoch 9 - iter 108/275 - loss 0.01534731 - time (sec): 5.50 - samples/sec: 1634.29 - lr: 0.000009 - momentum: 0.000000
2023-10-23 15:54:04,465 epoch 9 - iter 135/275 - loss 0.01222164 - time (sec): 6.90 - samples/sec: 1639.34 - lr: 0.000009 - momentum: 0.000000
2023-10-23 15:54:05,835 epoch 9 - iter 162/275 - loss 0.01073052 - time (sec): 8.27 - samples/sec: 1655.88 - lr: 0.000008 - momentum: 0.000000
2023-10-23 15:54:07,212 epoch 9 - iter 189/275 - loss 0.00980372 - time (sec): 9.64 - samples/sec: 1639.50 - lr: 0.000007 - momentum: 0.000000
2023-10-23 15:54:08,577 epoch 9 - iter 216/275 - loss 0.00910948 - time (sec): 11.01 - samples/sec: 1644.32 - lr: 0.000007 - momentum: 0.000000
2023-10-23 15:54:09,952 epoch 9 - iter 243/275 - loss 0.00861179 - time (sec): 12.38 - samples/sec: 1636.73 - lr: 0.000006 - momentum: 0.000000
2023-10-23 15:54:11,336 epoch 9 - iter 270/275 - loss 0.00873483 - time (sec): 13.77 - samples/sec: 1632.00 - lr: 0.000006 - momentum: 0.000000
2023-10-23 15:54:11,596 ----------------------------------------------------------------------------------------------------
2023-10-23 15:54:11,596 EPOCH 9 done: loss 0.0086 - lr: 0.000006
2023-10-23 15:54:12,138 DEV : loss 0.17060527205467224 - f1-score (micro avg) 0.8862
2023-10-23 15:54:12,144 ----------------------------------------------------------------------------------------------------
2023-10-23 15:54:13,506 epoch 10 - iter 27/275 - loss 0.00040091 - time (sec): 1.36 - samples/sec: 1548.35 - lr: 0.000005 - momentum: 0.000000
2023-10-23 15:54:14,844 epoch 10 - iter 54/275 - loss 0.00059839 - time (sec): 2.70 - samples/sec: 1612.29 - lr: 0.000005 - momentum: 0.000000
2023-10-23 15:54:16,170 epoch 10 - iter 81/275 - loss 0.00106300 - time (sec): 4.02 - samples/sec: 1670.09 - lr: 0.000004 - momentum: 0.000000
2023-10-23 15:54:17,527 epoch 10 - iter 108/275 - loss 0.00081409 - time (sec): 5.38 - samples/sec: 1681.74 - lr: 0.000004 - momentum: 0.000000
2023-10-23 15:54:18,898 epoch 10 - iter 135/275 - loss 0.00325952 - time (sec): 6.75 - samples/sec: 1647.07 - lr: 0.000003 - momentum: 0.000000
2023-10-23 15:54:20,253 epoch 10 - iter 162/275 - loss 0.00476398 - time (sec): 8.11 - samples/sec: 1619.34 - lr: 0.000002 - momentum: 0.000000
2023-10-23 15:54:21,624 epoch 10 - iter 189/275 - loss 0.00493019 - time (sec): 9.48 - samples/sec: 1611.15 - lr: 0.000002 - momentum: 0.000000
2023-10-23 15:54:22,987 epoch 10 - iter 216/275 - loss 0.00603623 - time (sec): 10.84 - samples/sec: 1626.70 - lr: 0.000001 - momentum: 0.000000
2023-10-23 15:54:24,336 epoch 10 - iter 243/275 - loss 0.00625175 - time (sec): 12.19 - samples/sec: 1636.88 - lr: 0.000001 - momentum: 0.000000
2023-10-23 15:54:25,709 epoch 10 - iter 270/275 - loss 0.00581465 - time (sec): 13.56 - samples/sec: 1646.57 - lr: 0.000000 - momentum: 0.000000
2023-10-23 15:54:25,960 ----------------------------------------------------------------------------------------------------
2023-10-23 15:54:25,960 EPOCH 10 done: loss 0.0057 - lr: 0.000000
2023-10-23 15:54:26,504 DEV : loss 0.17152057588100433 - f1-score (micro avg) 0.8809
2023-10-23 15:54:26,910 ----------------------------------------------------------------------------------------------------
2023-10-23 15:54:26,912 Loading model from best epoch ...
2023-10-23 15:54:28,582 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 15:54:29,253
Results:
- F-score (micro) 0.8976
- F-score (macro) 0.7692
- Accuracy 0.8281
By class:
precision recall f1-score support
scope 0.8864 0.8864 0.8864 176
pers 0.9837 0.9453 0.9641 128
work 0.8077 0.8514 0.8289 74
object 0.5000 0.5000 0.5000 2
loc 1.0000 0.5000 0.6667 2
micro avg 0.9000 0.8953 0.8976 382
macro avg 0.8356 0.7366 0.7692 382
weighted avg 0.9023 0.8953 0.8981 382
2023-10-23 15:54:29,253 ----------------------------------------------------------------------------------------------------
|