File size: 23,808 Bytes
4ba14ad |
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 |
2023-10-17 14:33:40,027 ----------------------------------------------------------------------------------------------------
2023-10-17 14:33:40,029 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=13, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-17 14:33:40,029 ----------------------------------------------------------------------------------------------------
2023-10-17 14:33:40,030 MultiCorpus: 6183 train + 680 dev + 2113 test sentences
- NER_HIPE_2022 Corpus: 6183 train + 680 dev + 2113 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/topres19th/en/with_doc_seperator
2023-10-17 14:33:40,030 ----------------------------------------------------------------------------------------------------
2023-10-17 14:33:40,030 Train: 6183 sentences
2023-10-17 14:33:40,030 (train_with_dev=False, train_with_test=False)
2023-10-17 14:33:40,030 ----------------------------------------------------------------------------------------------------
2023-10-17 14:33:40,030 Training Params:
2023-10-17 14:33:40,030 - learning_rate: "5e-05"
2023-10-17 14:33:40,030 - mini_batch_size: "8"
2023-10-17 14:33:40,030 - max_epochs: "10"
2023-10-17 14:33:40,030 - shuffle: "True"
2023-10-17 14:33:40,030 ----------------------------------------------------------------------------------------------------
2023-10-17 14:33:40,031 Plugins:
2023-10-17 14:33:40,031 - TensorboardLogger
2023-10-17 14:33:40,031 - LinearScheduler | warmup_fraction: '0.1'
2023-10-17 14:33:40,031 ----------------------------------------------------------------------------------------------------
2023-10-17 14:33:40,031 Final evaluation on model from best epoch (best-model.pt)
2023-10-17 14:33:40,031 - metric: "('micro avg', 'f1-score')"
2023-10-17 14:33:40,031 ----------------------------------------------------------------------------------------------------
2023-10-17 14:33:40,031 Computation:
2023-10-17 14:33:40,031 - compute on device: cuda:0
2023-10-17 14:33:40,031 - embedding storage: none
2023-10-17 14:33:40,031 ----------------------------------------------------------------------------------------------------
2023-10-17 14:33:40,031 Model training base path: "hmbench-topres19th/en-hmteams/teams-base-historic-multilingual-discriminator-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-5"
2023-10-17 14:33:40,031 ----------------------------------------------------------------------------------------------------
2023-10-17 14:33:40,031 ----------------------------------------------------------------------------------------------------
2023-10-17 14:33:40,032 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-17 14:33:47,292 epoch 1 - iter 77/773 - loss 2.50923587 - time (sec): 7.26 - samples/sec: 1831.85 - lr: 0.000005 - momentum: 0.000000
2023-10-17 14:33:54,411 epoch 1 - iter 154/773 - loss 1.42130164 - time (sec): 14.38 - samples/sec: 1820.77 - lr: 0.000010 - momentum: 0.000000
2023-10-17 14:34:01,440 epoch 1 - iter 231/773 - loss 1.00481350 - time (sec): 21.41 - samples/sec: 1826.16 - lr: 0.000015 - momentum: 0.000000
2023-10-17 14:34:08,979 epoch 1 - iter 308/773 - loss 0.80200716 - time (sec): 28.95 - samples/sec: 1752.52 - lr: 0.000020 - momentum: 0.000000
2023-10-17 14:34:16,018 epoch 1 - iter 385/773 - loss 0.68374915 - time (sec): 35.98 - samples/sec: 1715.41 - lr: 0.000025 - momentum: 0.000000
2023-10-17 14:34:22,978 epoch 1 - iter 462/773 - loss 0.58561188 - time (sec): 42.95 - samples/sec: 1737.12 - lr: 0.000030 - momentum: 0.000000
2023-10-17 14:34:29,816 epoch 1 - iter 539/773 - loss 0.52024904 - time (sec): 49.78 - samples/sec: 1741.94 - lr: 0.000035 - momentum: 0.000000
2023-10-17 14:34:36,668 epoch 1 - iter 616/773 - loss 0.46975073 - time (sec): 56.63 - samples/sec: 1747.71 - lr: 0.000040 - momentum: 0.000000
2023-10-17 14:34:43,744 epoch 1 - iter 693/773 - loss 0.42724592 - time (sec): 63.71 - samples/sec: 1754.87 - lr: 0.000045 - momentum: 0.000000
2023-10-17 14:34:50,680 epoch 1 - iter 770/773 - loss 0.39511649 - time (sec): 70.65 - samples/sec: 1750.92 - lr: 0.000050 - momentum: 0.000000
2023-10-17 14:34:50,952 ----------------------------------------------------------------------------------------------------
2023-10-17 14:34:50,952 EPOCH 1 done: loss 0.3936 - lr: 0.000050
2023-10-17 14:34:53,277 DEV : loss 0.05638180673122406 - f1-score (micro avg) 0.7427
2023-10-17 14:34:53,307 saving best model
2023-10-17 14:34:53,948 ----------------------------------------------------------------------------------------------------
2023-10-17 14:35:01,084 epoch 2 - iter 77/773 - loss 0.08450516 - time (sec): 7.13 - samples/sec: 1724.49 - lr: 0.000049 - momentum: 0.000000
2023-10-17 14:35:08,104 epoch 2 - iter 154/773 - loss 0.07623715 - time (sec): 14.15 - samples/sec: 1741.82 - lr: 0.000049 - momentum: 0.000000
2023-10-17 14:35:16,129 epoch 2 - iter 231/773 - loss 0.07797626 - time (sec): 22.18 - samples/sec: 1685.46 - lr: 0.000048 - momentum: 0.000000
2023-10-17 14:35:23,452 epoch 2 - iter 308/773 - loss 0.07829088 - time (sec): 29.50 - samples/sec: 1683.48 - lr: 0.000048 - momentum: 0.000000
2023-10-17 14:35:30,778 epoch 2 - iter 385/773 - loss 0.07642468 - time (sec): 36.83 - samples/sec: 1694.33 - lr: 0.000047 - momentum: 0.000000
2023-10-17 14:35:38,282 epoch 2 - iter 462/773 - loss 0.07853226 - time (sec): 44.33 - samples/sec: 1712.25 - lr: 0.000047 - momentum: 0.000000
2023-10-17 14:35:45,196 epoch 2 - iter 539/773 - loss 0.07797723 - time (sec): 51.25 - samples/sec: 1708.62 - lr: 0.000046 - momentum: 0.000000
2023-10-17 14:35:52,242 epoch 2 - iter 616/773 - loss 0.07775728 - time (sec): 58.29 - samples/sec: 1713.08 - lr: 0.000046 - momentum: 0.000000
2023-10-17 14:35:59,251 epoch 2 - iter 693/773 - loss 0.07783386 - time (sec): 65.30 - samples/sec: 1700.02 - lr: 0.000045 - momentum: 0.000000
2023-10-17 14:36:06,466 epoch 2 - iter 770/773 - loss 0.07655030 - time (sec): 72.52 - samples/sec: 1707.08 - lr: 0.000044 - momentum: 0.000000
2023-10-17 14:36:06,742 ----------------------------------------------------------------------------------------------------
2023-10-17 14:36:06,742 EPOCH 2 done: loss 0.0762 - lr: 0.000044
2023-10-17 14:36:09,756 DEV : loss 0.05624998360872269 - f1-score (micro avg) 0.7741
2023-10-17 14:36:09,787 saving best model
2023-10-17 14:36:11,251 ----------------------------------------------------------------------------------------------------
2023-10-17 14:36:18,549 epoch 3 - iter 77/773 - loss 0.04747565 - time (sec): 7.29 - samples/sec: 1682.24 - lr: 0.000044 - momentum: 0.000000
2023-10-17 14:36:26,038 epoch 3 - iter 154/773 - loss 0.05208446 - time (sec): 14.78 - samples/sec: 1713.20 - lr: 0.000043 - momentum: 0.000000
2023-10-17 14:36:33,264 epoch 3 - iter 231/773 - loss 0.05199616 - time (sec): 22.00 - samples/sec: 1689.36 - lr: 0.000043 - momentum: 0.000000
2023-10-17 14:36:40,604 epoch 3 - iter 308/773 - loss 0.04917980 - time (sec): 29.35 - samples/sec: 1690.54 - lr: 0.000042 - momentum: 0.000000
2023-10-17 14:36:47,941 epoch 3 - iter 385/773 - loss 0.04758071 - time (sec): 36.68 - samples/sec: 1678.00 - lr: 0.000042 - momentum: 0.000000
2023-10-17 14:36:55,193 epoch 3 - iter 462/773 - loss 0.04862057 - time (sec): 43.93 - samples/sec: 1685.83 - lr: 0.000041 - momentum: 0.000000
2023-10-17 14:37:02,584 epoch 3 - iter 539/773 - loss 0.04856681 - time (sec): 51.33 - samples/sec: 1684.53 - lr: 0.000041 - momentum: 0.000000
2023-10-17 14:37:09,641 epoch 3 - iter 616/773 - loss 0.05066774 - time (sec): 58.38 - samples/sec: 1694.99 - lr: 0.000040 - momentum: 0.000000
2023-10-17 14:37:16,730 epoch 3 - iter 693/773 - loss 0.05007386 - time (sec): 65.47 - samples/sec: 1706.62 - lr: 0.000039 - momentum: 0.000000
2023-10-17 14:37:23,872 epoch 3 - iter 770/773 - loss 0.05081763 - time (sec): 72.61 - samples/sec: 1704.59 - lr: 0.000039 - momentum: 0.000000
2023-10-17 14:37:24,152 ----------------------------------------------------------------------------------------------------
2023-10-17 14:37:24,152 EPOCH 3 done: loss 0.0507 - lr: 0.000039
2023-10-17 14:37:27,074 DEV : loss 0.0668470710515976 - f1-score (micro avg) 0.7938
2023-10-17 14:37:27,103 saving best model
2023-10-17 14:37:28,513 ----------------------------------------------------------------------------------------------------
2023-10-17 14:37:35,670 epoch 4 - iter 77/773 - loss 0.03373869 - time (sec): 7.15 - samples/sec: 1716.25 - lr: 0.000038 - momentum: 0.000000
2023-10-17 14:37:42,914 epoch 4 - iter 154/773 - loss 0.03223990 - time (sec): 14.40 - samples/sec: 1797.82 - lr: 0.000038 - momentum: 0.000000
2023-10-17 14:37:50,041 epoch 4 - iter 231/773 - loss 0.03664264 - time (sec): 21.52 - samples/sec: 1788.19 - lr: 0.000037 - momentum: 0.000000
2023-10-17 14:37:56,925 epoch 4 - iter 308/773 - loss 0.03580473 - time (sec): 28.41 - samples/sec: 1766.49 - lr: 0.000037 - momentum: 0.000000
2023-10-17 14:38:03,783 epoch 4 - iter 385/773 - loss 0.03747929 - time (sec): 35.27 - samples/sec: 1758.49 - lr: 0.000036 - momentum: 0.000000
2023-10-17 14:38:11,144 epoch 4 - iter 462/773 - loss 0.03752798 - time (sec): 42.63 - samples/sec: 1757.21 - lr: 0.000036 - momentum: 0.000000
2023-10-17 14:38:18,355 epoch 4 - iter 539/773 - loss 0.03661439 - time (sec): 49.84 - samples/sec: 1767.40 - lr: 0.000035 - momentum: 0.000000
2023-10-17 14:38:25,461 epoch 4 - iter 616/773 - loss 0.03508167 - time (sec): 56.94 - samples/sec: 1758.48 - lr: 0.000034 - momentum: 0.000000
2023-10-17 14:38:32,556 epoch 4 - iter 693/773 - loss 0.03524376 - time (sec): 64.04 - samples/sec: 1756.85 - lr: 0.000034 - momentum: 0.000000
2023-10-17 14:38:39,444 epoch 4 - iter 770/773 - loss 0.03569860 - time (sec): 70.93 - samples/sec: 1747.91 - lr: 0.000033 - momentum: 0.000000
2023-10-17 14:38:39,704 ----------------------------------------------------------------------------------------------------
2023-10-17 14:38:39,705 EPOCH 4 done: loss 0.0357 - lr: 0.000033
2023-10-17 14:38:42,665 DEV : loss 0.09208610653877258 - f1-score (micro avg) 0.7724
2023-10-17 14:38:42,696 ----------------------------------------------------------------------------------------------------
2023-10-17 14:38:49,903 epoch 5 - iter 77/773 - loss 0.03208987 - time (sec): 7.20 - samples/sec: 1819.01 - lr: 0.000033 - momentum: 0.000000
2023-10-17 14:38:57,068 epoch 5 - iter 154/773 - loss 0.02963635 - time (sec): 14.37 - samples/sec: 1790.88 - lr: 0.000032 - momentum: 0.000000
2023-10-17 14:39:04,021 epoch 5 - iter 231/773 - loss 0.03156884 - time (sec): 21.32 - samples/sec: 1782.65 - lr: 0.000032 - momentum: 0.000000
2023-10-17 14:39:10,992 epoch 5 - iter 308/773 - loss 0.02957247 - time (sec): 28.29 - samples/sec: 1779.80 - lr: 0.000031 - momentum: 0.000000
2023-10-17 14:39:18,021 epoch 5 - iter 385/773 - loss 0.03043304 - time (sec): 35.32 - samples/sec: 1777.02 - lr: 0.000031 - momentum: 0.000000
2023-10-17 14:39:25,200 epoch 5 - iter 462/773 - loss 0.03088270 - time (sec): 42.50 - samples/sec: 1764.76 - lr: 0.000030 - momentum: 0.000000
2023-10-17 14:39:31,906 epoch 5 - iter 539/773 - loss 0.02923464 - time (sec): 49.21 - samples/sec: 1779.74 - lr: 0.000029 - momentum: 0.000000
2023-10-17 14:39:38,445 epoch 5 - iter 616/773 - loss 0.02915810 - time (sec): 55.75 - samples/sec: 1797.13 - lr: 0.000029 - momentum: 0.000000
2023-10-17 14:39:44,773 epoch 5 - iter 693/773 - loss 0.02815671 - time (sec): 62.07 - samples/sec: 1802.51 - lr: 0.000028 - momentum: 0.000000
2023-10-17 14:39:51,103 epoch 5 - iter 770/773 - loss 0.02806962 - time (sec): 68.41 - samples/sec: 1810.07 - lr: 0.000028 - momentum: 0.000000
2023-10-17 14:39:51,338 ----------------------------------------------------------------------------------------------------
2023-10-17 14:39:51,339 EPOCH 5 done: loss 0.0281 - lr: 0.000028
2023-10-17 14:39:54,281 DEV : loss 0.09081842750310898 - f1-score (micro avg) 0.7808
2023-10-17 14:39:54,311 ----------------------------------------------------------------------------------------------------
2023-10-17 14:40:01,302 epoch 6 - iter 77/773 - loss 0.01833530 - time (sec): 6.99 - samples/sec: 1755.27 - lr: 0.000027 - momentum: 0.000000
2023-10-17 14:40:08,326 epoch 6 - iter 154/773 - loss 0.01654716 - time (sec): 14.01 - samples/sec: 1770.07 - lr: 0.000027 - momentum: 0.000000
2023-10-17 14:40:15,751 epoch 6 - iter 231/773 - loss 0.01693263 - time (sec): 21.44 - samples/sec: 1749.34 - lr: 0.000026 - momentum: 0.000000
2023-10-17 14:40:22,848 epoch 6 - iter 308/773 - loss 0.01631071 - time (sec): 28.54 - samples/sec: 1762.95 - lr: 0.000026 - momentum: 0.000000
2023-10-17 14:40:29,756 epoch 6 - iter 385/773 - loss 0.01560773 - time (sec): 35.44 - samples/sec: 1745.74 - lr: 0.000025 - momentum: 0.000000
2023-10-17 14:40:37,016 epoch 6 - iter 462/773 - loss 0.01553854 - time (sec): 42.70 - samples/sec: 1752.52 - lr: 0.000024 - momentum: 0.000000
2023-10-17 14:40:43,841 epoch 6 - iter 539/773 - loss 0.01650694 - time (sec): 49.53 - samples/sec: 1737.47 - lr: 0.000024 - momentum: 0.000000
2023-10-17 14:40:50,797 epoch 6 - iter 616/773 - loss 0.01698956 - time (sec): 56.48 - samples/sec: 1733.26 - lr: 0.000023 - momentum: 0.000000
2023-10-17 14:40:57,982 epoch 6 - iter 693/773 - loss 0.01685377 - time (sec): 63.67 - samples/sec: 1744.16 - lr: 0.000023 - momentum: 0.000000
2023-10-17 14:41:05,215 epoch 6 - iter 770/773 - loss 0.01627500 - time (sec): 70.90 - samples/sec: 1746.88 - lr: 0.000022 - momentum: 0.000000
2023-10-17 14:41:05,480 ----------------------------------------------------------------------------------------------------
2023-10-17 14:41:05,481 EPOCH 6 done: loss 0.0163 - lr: 0.000022
2023-10-17 14:41:08,499 DEV : loss 0.10279172658920288 - f1-score (micro avg) 0.7918
2023-10-17 14:41:08,530 ----------------------------------------------------------------------------------------------------
2023-10-17 14:41:15,513 epoch 7 - iter 77/773 - loss 0.00452528 - time (sec): 6.98 - samples/sec: 1685.35 - lr: 0.000022 - momentum: 0.000000
2023-10-17 14:41:22,501 epoch 7 - iter 154/773 - loss 0.00876548 - time (sec): 13.97 - samples/sec: 1722.63 - lr: 0.000021 - momentum: 0.000000
2023-10-17 14:41:30,027 epoch 7 - iter 231/773 - loss 0.01243303 - time (sec): 21.49 - samples/sec: 1687.99 - lr: 0.000021 - momentum: 0.000000
2023-10-17 14:41:37,202 epoch 7 - iter 308/773 - loss 0.01319413 - time (sec): 28.67 - samples/sec: 1701.70 - lr: 0.000020 - momentum: 0.000000
2023-10-17 14:41:44,698 epoch 7 - iter 385/773 - loss 0.01232171 - time (sec): 36.17 - samples/sec: 1714.09 - lr: 0.000019 - momentum: 0.000000
2023-10-17 14:41:51,565 epoch 7 - iter 462/773 - loss 0.01118046 - time (sec): 43.03 - samples/sec: 1726.55 - lr: 0.000019 - momentum: 0.000000
2023-10-17 14:41:58,599 epoch 7 - iter 539/773 - loss 0.01102460 - time (sec): 50.07 - samples/sec: 1731.55 - lr: 0.000018 - momentum: 0.000000
2023-10-17 14:42:05,563 epoch 7 - iter 616/773 - loss 0.01052311 - time (sec): 57.03 - samples/sec: 1742.46 - lr: 0.000018 - momentum: 0.000000
2023-10-17 14:42:12,767 epoch 7 - iter 693/773 - loss 0.01166359 - time (sec): 64.23 - samples/sec: 1755.31 - lr: 0.000017 - momentum: 0.000000
2023-10-17 14:42:19,919 epoch 7 - iter 770/773 - loss 0.01166227 - time (sec): 71.39 - samples/sec: 1735.49 - lr: 0.000017 - momentum: 0.000000
2023-10-17 14:42:20,202 ----------------------------------------------------------------------------------------------------
2023-10-17 14:42:20,203 EPOCH 7 done: loss 0.0118 - lr: 0.000017
2023-10-17 14:42:23,314 DEV : loss 0.12683075666427612 - f1-score (micro avg) 0.7672
2023-10-17 14:42:23,351 ----------------------------------------------------------------------------------------------------
2023-10-17 14:42:30,519 epoch 8 - iter 77/773 - loss 0.01092306 - time (sec): 7.17 - samples/sec: 1632.15 - lr: 0.000016 - momentum: 0.000000
2023-10-17 14:42:37,610 epoch 8 - iter 154/773 - loss 0.00834521 - time (sec): 14.26 - samples/sec: 1700.73 - lr: 0.000016 - momentum: 0.000000
2023-10-17 14:42:44,625 epoch 8 - iter 231/773 - loss 0.00763347 - time (sec): 21.27 - samples/sec: 1715.42 - lr: 0.000015 - momentum: 0.000000
2023-10-17 14:42:51,811 epoch 8 - iter 308/773 - loss 0.00734188 - time (sec): 28.46 - samples/sec: 1728.36 - lr: 0.000014 - momentum: 0.000000
2023-10-17 14:42:59,357 epoch 8 - iter 385/773 - loss 0.00782979 - time (sec): 36.00 - samples/sec: 1723.98 - lr: 0.000014 - momentum: 0.000000
2023-10-17 14:43:06,454 epoch 8 - iter 462/773 - loss 0.00716809 - time (sec): 43.10 - samples/sec: 1728.60 - lr: 0.000013 - momentum: 0.000000
2023-10-17 14:43:13,826 epoch 8 - iter 539/773 - loss 0.00773899 - time (sec): 50.47 - samples/sec: 1734.28 - lr: 0.000013 - momentum: 0.000000
2023-10-17 14:43:20,901 epoch 8 - iter 616/773 - loss 0.00783439 - time (sec): 57.55 - samples/sec: 1726.13 - lr: 0.000012 - momentum: 0.000000
2023-10-17 14:43:27,798 epoch 8 - iter 693/773 - loss 0.00774684 - time (sec): 64.45 - samples/sec: 1726.79 - lr: 0.000012 - momentum: 0.000000
2023-10-17 14:43:35,007 epoch 8 - iter 770/773 - loss 0.00791622 - time (sec): 71.65 - samples/sec: 1727.63 - lr: 0.000011 - momentum: 0.000000
2023-10-17 14:43:35,283 ----------------------------------------------------------------------------------------------------
2023-10-17 14:43:35,284 EPOCH 8 done: loss 0.0081 - lr: 0.000011
2023-10-17 14:43:38,140 DEV : loss 0.12362485378980637 - f1-score (micro avg) 0.7778
2023-10-17 14:43:38,169 ----------------------------------------------------------------------------------------------------
2023-10-17 14:43:45,185 epoch 9 - iter 77/773 - loss 0.00194071 - time (sec): 7.01 - samples/sec: 1862.43 - lr: 0.000011 - momentum: 0.000000
2023-10-17 14:43:52,369 epoch 9 - iter 154/773 - loss 0.00273045 - time (sec): 14.20 - samples/sec: 1830.87 - lr: 0.000010 - momentum: 0.000000
2023-10-17 14:43:59,729 epoch 9 - iter 231/773 - loss 0.00364150 - time (sec): 21.56 - samples/sec: 1778.00 - lr: 0.000009 - momentum: 0.000000
2023-10-17 14:44:06,692 epoch 9 - iter 308/773 - loss 0.00314935 - time (sec): 28.52 - samples/sec: 1747.14 - lr: 0.000009 - momentum: 0.000000
2023-10-17 14:44:13,912 epoch 9 - iter 385/773 - loss 0.00381506 - time (sec): 35.74 - samples/sec: 1759.68 - lr: 0.000008 - momentum: 0.000000
2023-10-17 14:44:21,090 epoch 9 - iter 462/773 - loss 0.00476997 - time (sec): 42.92 - samples/sec: 1763.78 - lr: 0.000008 - momentum: 0.000000
2023-10-17 14:44:28,137 epoch 9 - iter 539/773 - loss 0.00427351 - time (sec): 49.97 - samples/sec: 1762.97 - lr: 0.000007 - momentum: 0.000000
2023-10-17 14:44:35,251 epoch 9 - iter 616/773 - loss 0.00437576 - time (sec): 57.08 - samples/sec: 1741.96 - lr: 0.000007 - momentum: 0.000000
2023-10-17 14:44:42,591 epoch 9 - iter 693/773 - loss 0.00480221 - time (sec): 64.42 - samples/sec: 1737.41 - lr: 0.000006 - momentum: 0.000000
2023-10-17 14:44:49,550 epoch 9 - iter 770/773 - loss 0.00516866 - time (sec): 71.38 - samples/sec: 1732.98 - lr: 0.000006 - momentum: 0.000000
2023-10-17 14:44:49,828 ----------------------------------------------------------------------------------------------------
2023-10-17 14:44:49,828 EPOCH 9 done: loss 0.0051 - lr: 0.000006
2023-10-17 14:44:52,751 DEV : loss 0.12392386794090271 - f1-score (micro avg) 0.7699
2023-10-17 14:44:52,780 ----------------------------------------------------------------------------------------------------
2023-10-17 14:45:00,250 epoch 10 - iter 77/773 - loss 0.00532611 - time (sec): 7.47 - samples/sec: 1608.63 - lr: 0.000005 - momentum: 0.000000
2023-10-17 14:45:07,239 epoch 10 - iter 154/773 - loss 0.00378488 - time (sec): 14.46 - samples/sec: 1647.56 - lr: 0.000005 - momentum: 0.000000
2023-10-17 14:45:14,478 epoch 10 - iter 231/773 - loss 0.00382497 - time (sec): 21.70 - samples/sec: 1688.02 - lr: 0.000004 - momentum: 0.000000
2023-10-17 14:45:21,681 epoch 10 - iter 308/773 - loss 0.00350380 - time (sec): 28.90 - samples/sec: 1711.86 - lr: 0.000003 - momentum: 0.000000
2023-10-17 14:45:28,814 epoch 10 - iter 385/773 - loss 0.00387383 - time (sec): 36.03 - samples/sec: 1709.60 - lr: 0.000003 - momentum: 0.000000
2023-10-17 14:45:35,914 epoch 10 - iter 462/773 - loss 0.00390395 - time (sec): 43.13 - samples/sec: 1713.52 - lr: 0.000002 - momentum: 0.000000
2023-10-17 14:45:43,094 epoch 10 - iter 539/773 - loss 0.00380626 - time (sec): 50.31 - samples/sec: 1717.59 - lr: 0.000002 - momentum: 0.000000
2023-10-17 14:45:50,077 epoch 10 - iter 616/773 - loss 0.00372703 - time (sec): 57.29 - samples/sec: 1724.36 - lr: 0.000001 - momentum: 0.000000
2023-10-17 14:45:56,991 epoch 10 - iter 693/773 - loss 0.00367782 - time (sec): 64.21 - samples/sec: 1740.68 - lr: 0.000001 - momentum: 0.000000
2023-10-17 14:46:03,998 epoch 10 - iter 770/773 - loss 0.00357798 - time (sec): 71.22 - samples/sec: 1737.64 - lr: 0.000000 - momentum: 0.000000
2023-10-17 14:46:04,264 ----------------------------------------------------------------------------------------------------
2023-10-17 14:46:04,264 EPOCH 10 done: loss 0.0036 - lr: 0.000000
2023-10-17 14:46:07,181 DEV : loss 0.12464166432619095 - f1-score (micro avg) 0.7835
2023-10-17 14:46:07,879 ----------------------------------------------------------------------------------------------------
2023-10-17 14:46:07,881 Loading model from best epoch ...
2023-10-17 14:46:10,198 SequenceTagger predicts: Dictionary with 13 tags: O, S-LOC, B-LOC, E-LOC, I-LOC, S-BUILDING, B-BUILDING, E-BUILDING, I-BUILDING, S-STREET, B-STREET, E-STREET, I-STREET
2023-10-17 14:46:18,809
Results:
- F-score (micro) 0.7858
- F-score (macro) 0.6784
- Accuracy 0.6705
By class:
precision recall f1-score support
LOC 0.8015 0.8795 0.8387 946
BUILDING 0.5374 0.6216 0.5764 185
STREET 0.5479 0.7143 0.6202 56
micro avg 0.7449 0.8315 0.7858 1187
macro avg 0.6290 0.7385 0.6784 1187
weighted avg 0.7484 0.8315 0.7875 1187
2023-10-17 14:46:18,809 ----------------------------------------------------------------------------------------------------
|