File size: 23,866 Bytes
73730f6 |
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 |
2023-10-25 21:27:49,476 ----------------------------------------------------------------------------------------------------
2023-10-25 21:27:49,477 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=17, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-25 21:27:49,477 ----------------------------------------------------------------------------------------------------
2023-10-25 21:27:49,477 MultiCorpus: 1166 train + 165 dev + 415 test sentences
- NER_HIPE_2022 Corpus: 1166 train + 165 dev + 415 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/newseye/fi/with_doc_seperator
2023-10-25 21:27:49,477 ----------------------------------------------------------------------------------------------------
2023-10-25 21:27:49,477 Train: 1166 sentences
2023-10-25 21:27:49,477 (train_with_dev=False, train_with_test=False)
2023-10-25 21:27:49,477 ----------------------------------------------------------------------------------------------------
2023-10-25 21:27:49,477 Training Params:
2023-10-25 21:27:49,477 - learning_rate: "5e-05"
2023-10-25 21:27:49,477 - mini_batch_size: "8"
2023-10-25 21:27:49,477 - max_epochs: "10"
2023-10-25 21:27:49,477 - shuffle: "True"
2023-10-25 21:27:49,477 ----------------------------------------------------------------------------------------------------
2023-10-25 21:27:49,477 Plugins:
2023-10-25 21:27:49,477 - TensorboardLogger
2023-10-25 21:27:49,477 - LinearScheduler | warmup_fraction: '0.1'
2023-10-25 21:27:49,477 ----------------------------------------------------------------------------------------------------
2023-10-25 21:27:49,477 Final evaluation on model from best epoch (best-model.pt)
2023-10-25 21:27:49,477 - metric: "('micro avg', 'f1-score')"
2023-10-25 21:27:49,477 ----------------------------------------------------------------------------------------------------
2023-10-25 21:27:49,477 Computation:
2023-10-25 21:27:49,477 - compute on device: cuda:0
2023-10-25 21:27:49,477 - embedding storage: none
2023-10-25 21:27:49,477 ----------------------------------------------------------------------------------------------------
2023-10-25 21:27:49,477 Model training base path: "hmbench-newseye/fi-dbmdz/bert-base-historic-multilingual-64k-td-cased-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-5"
2023-10-25 21:27:49,477 ----------------------------------------------------------------------------------------------------
2023-10-25 21:27:49,477 ----------------------------------------------------------------------------------------------------
2023-10-25 21:27:49,478 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-25 21:27:50,374 epoch 1 - iter 14/146 - loss 2.52526717 - time (sec): 0.90 - samples/sec: 4334.26 - lr: 0.000004 - momentum: 0.000000
2023-10-25 21:27:51,165 epoch 1 - iter 28/146 - loss 1.96710579 - time (sec): 1.69 - samples/sec: 4519.61 - lr: 0.000009 - momentum: 0.000000
2023-10-25 21:27:52,148 epoch 1 - iter 42/146 - loss 1.59185410 - time (sec): 2.67 - samples/sec: 4566.94 - lr: 0.000014 - momentum: 0.000000
2023-10-25 21:27:53,087 epoch 1 - iter 56/146 - loss 1.35500246 - time (sec): 3.61 - samples/sec: 4550.27 - lr: 0.000019 - momentum: 0.000000
2023-10-25 21:27:53,966 epoch 1 - iter 70/146 - loss 1.14468229 - time (sec): 4.49 - samples/sec: 4665.80 - lr: 0.000024 - momentum: 0.000000
2023-10-25 21:27:54,730 epoch 1 - iter 84/146 - loss 1.03230977 - time (sec): 5.25 - samples/sec: 4651.66 - lr: 0.000028 - momentum: 0.000000
2023-10-25 21:27:55,831 epoch 1 - iter 98/146 - loss 0.91852887 - time (sec): 6.35 - samples/sec: 4594.65 - lr: 0.000033 - momentum: 0.000000
2023-10-25 21:27:56,790 epoch 1 - iter 112/146 - loss 0.82179505 - time (sec): 7.31 - samples/sec: 4648.55 - lr: 0.000038 - momentum: 0.000000
2023-10-25 21:27:57,868 epoch 1 - iter 126/146 - loss 0.74630898 - time (sec): 8.39 - samples/sec: 4676.66 - lr: 0.000043 - momentum: 0.000000
2023-10-25 21:27:58,648 epoch 1 - iter 140/146 - loss 0.70541709 - time (sec): 9.17 - samples/sec: 4657.20 - lr: 0.000048 - momentum: 0.000000
2023-10-25 21:27:59,025 ----------------------------------------------------------------------------------------------------
2023-10-25 21:27:59,025 EPOCH 1 done: loss 0.6862 - lr: 0.000048
2023-10-25 21:27:59,692 DEV : loss 0.1474415361881256 - f1-score (micro avg) 0.5733
2023-10-25 21:27:59,697 saving best model
2023-10-25 21:28:00,164 ----------------------------------------------------------------------------------------------------
2023-10-25 21:28:01,070 epoch 2 - iter 14/146 - loss 0.17089233 - time (sec): 0.90 - samples/sec: 4814.35 - lr: 0.000050 - momentum: 0.000000
2023-10-25 21:28:02,016 epoch 2 - iter 28/146 - loss 0.21428820 - time (sec): 1.85 - samples/sec: 4654.61 - lr: 0.000049 - momentum: 0.000000
2023-10-25 21:28:02,882 epoch 2 - iter 42/146 - loss 0.20237772 - time (sec): 2.72 - samples/sec: 4512.63 - lr: 0.000048 - momentum: 0.000000
2023-10-25 21:28:03,920 epoch 2 - iter 56/146 - loss 0.18853700 - time (sec): 3.75 - samples/sec: 4493.86 - lr: 0.000048 - momentum: 0.000000
2023-10-25 21:28:04,814 epoch 2 - iter 70/146 - loss 0.18382748 - time (sec): 4.65 - samples/sec: 4572.10 - lr: 0.000047 - momentum: 0.000000
2023-10-25 21:28:05,612 epoch 2 - iter 84/146 - loss 0.18071178 - time (sec): 5.45 - samples/sec: 4642.87 - lr: 0.000047 - momentum: 0.000000
2023-10-25 21:28:06,448 epoch 2 - iter 98/146 - loss 0.17992603 - time (sec): 6.28 - samples/sec: 4685.36 - lr: 0.000046 - momentum: 0.000000
2023-10-25 21:28:07,296 epoch 2 - iter 112/146 - loss 0.17844954 - time (sec): 7.13 - samples/sec: 4705.85 - lr: 0.000046 - momentum: 0.000000
2023-10-25 21:28:08,253 epoch 2 - iter 126/146 - loss 0.17872038 - time (sec): 8.09 - samples/sec: 4685.94 - lr: 0.000045 - momentum: 0.000000
2023-10-25 21:28:09,246 epoch 2 - iter 140/146 - loss 0.17249572 - time (sec): 9.08 - samples/sec: 4687.55 - lr: 0.000045 - momentum: 0.000000
2023-10-25 21:28:09,584 ----------------------------------------------------------------------------------------------------
2023-10-25 21:28:09,584 EPOCH 2 done: loss 0.1707 - lr: 0.000045
2023-10-25 21:28:10,498 DEV : loss 0.1254296749830246 - f1-score (micro avg) 0.6564
2023-10-25 21:28:10,503 saving best model
2023-10-25 21:28:11,122 ----------------------------------------------------------------------------------------------------
2023-10-25 21:28:12,433 epoch 3 - iter 14/146 - loss 0.08260202 - time (sec): 1.31 - samples/sec: 4377.23 - lr: 0.000044 - momentum: 0.000000
2023-10-25 21:28:13,378 epoch 3 - iter 28/146 - loss 0.09990362 - time (sec): 2.25 - samples/sec: 4505.97 - lr: 0.000043 - momentum: 0.000000
2023-10-25 21:28:14,194 epoch 3 - iter 42/146 - loss 0.10051524 - time (sec): 3.07 - samples/sec: 4619.88 - lr: 0.000043 - momentum: 0.000000
2023-10-25 21:28:15,072 epoch 3 - iter 56/146 - loss 0.09789885 - time (sec): 3.95 - samples/sec: 4535.60 - lr: 0.000042 - momentum: 0.000000
2023-10-25 21:28:15,991 epoch 3 - iter 70/146 - loss 0.09781657 - time (sec): 4.87 - samples/sec: 4595.00 - lr: 0.000042 - momentum: 0.000000
2023-10-25 21:28:16,812 epoch 3 - iter 84/146 - loss 0.09773301 - time (sec): 5.69 - samples/sec: 4587.72 - lr: 0.000041 - momentum: 0.000000
2023-10-25 21:28:17,705 epoch 3 - iter 98/146 - loss 0.09125920 - time (sec): 6.58 - samples/sec: 4628.73 - lr: 0.000041 - momentum: 0.000000
2023-10-25 21:28:18,551 epoch 3 - iter 112/146 - loss 0.09081431 - time (sec): 7.43 - samples/sec: 4588.91 - lr: 0.000040 - momentum: 0.000000
2023-10-25 21:28:19,452 epoch 3 - iter 126/146 - loss 0.09242276 - time (sec): 8.33 - samples/sec: 4629.85 - lr: 0.000040 - momentum: 0.000000
2023-10-25 21:28:20,231 epoch 3 - iter 140/146 - loss 0.08938489 - time (sec): 9.11 - samples/sec: 4708.37 - lr: 0.000039 - momentum: 0.000000
2023-10-25 21:28:20,593 ----------------------------------------------------------------------------------------------------
2023-10-25 21:28:20,593 EPOCH 3 done: loss 0.0898 - lr: 0.000039
2023-10-25 21:28:21,508 DEV : loss 0.11051346361637115 - f1-score (micro avg) 0.7417
2023-10-25 21:28:21,513 saving best model
2023-10-25 21:28:21,994 ----------------------------------------------------------------------------------------------------
2023-10-25 21:28:22,771 epoch 4 - iter 14/146 - loss 0.06118331 - time (sec): 0.78 - samples/sec: 4971.16 - lr: 0.000038 - momentum: 0.000000
2023-10-25 21:28:23,745 epoch 4 - iter 28/146 - loss 0.04975956 - time (sec): 1.75 - samples/sec: 4691.92 - lr: 0.000038 - momentum: 0.000000
2023-10-25 21:28:24,628 epoch 4 - iter 42/146 - loss 0.04676491 - time (sec): 2.63 - samples/sec: 4792.58 - lr: 0.000037 - momentum: 0.000000
2023-10-25 21:28:25,468 epoch 4 - iter 56/146 - loss 0.04813187 - time (sec): 3.47 - samples/sec: 4634.02 - lr: 0.000037 - momentum: 0.000000
2023-10-25 21:28:26,397 epoch 4 - iter 70/146 - loss 0.04806710 - time (sec): 4.40 - samples/sec: 4840.63 - lr: 0.000036 - momentum: 0.000000
2023-10-25 21:28:27,403 epoch 4 - iter 84/146 - loss 0.05168396 - time (sec): 5.41 - samples/sec: 4870.21 - lr: 0.000036 - momentum: 0.000000
2023-10-25 21:28:28,249 epoch 4 - iter 98/146 - loss 0.05023341 - time (sec): 6.25 - samples/sec: 4935.40 - lr: 0.000035 - momentum: 0.000000
2023-10-25 21:28:29,084 epoch 4 - iter 112/146 - loss 0.05330836 - time (sec): 7.09 - samples/sec: 4889.49 - lr: 0.000035 - momentum: 0.000000
2023-10-25 21:28:30,183 epoch 4 - iter 126/146 - loss 0.05565502 - time (sec): 8.19 - samples/sec: 4793.60 - lr: 0.000034 - momentum: 0.000000
2023-10-25 21:28:30,966 epoch 4 - iter 140/146 - loss 0.05422907 - time (sec): 8.97 - samples/sec: 4782.18 - lr: 0.000034 - momentum: 0.000000
2023-10-25 21:28:31,283 ----------------------------------------------------------------------------------------------------
2023-10-25 21:28:31,283 EPOCH 4 done: loss 0.0546 - lr: 0.000034
2023-10-25 21:28:32,200 DEV : loss 0.09850870817899704 - f1-score (micro avg) 0.7521
2023-10-25 21:28:32,205 saving best model
2023-10-25 21:28:32,815 ----------------------------------------------------------------------------------------------------
2023-10-25 21:28:33,722 epoch 5 - iter 14/146 - loss 0.02659804 - time (sec): 0.91 - samples/sec: 4906.64 - lr: 0.000033 - momentum: 0.000000
2023-10-25 21:28:34,641 epoch 5 - iter 28/146 - loss 0.03160875 - time (sec): 1.82 - samples/sec: 4673.76 - lr: 0.000032 - momentum: 0.000000
2023-10-25 21:28:35,524 epoch 5 - iter 42/146 - loss 0.02704574 - time (sec): 2.71 - samples/sec: 4687.55 - lr: 0.000032 - momentum: 0.000000
2023-10-25 21:28:36,385 epoch 5 - iter 56/146 - loss 0.02872717 - time (sec): 3.57 - samples/sec: 4740.65 - lr: 0.000031 - momentum: 0.000000
2023-10-25 21:28:37,247 epoch 5 - iter 70/146 - loss 0.02814581 - time (sec): 4.43 - samples/sec: 4780.67 - lr: 0.000031 - momentum: 0.000000
2023-10-25 21:28:38,017 epoch 5 - iter 84/146 - loss 0.02973212 - time (sec): 5.20 - samples/sec: 4792.43 - lr: 0.000030 - momentum: 0.000000
2023-10-25 21:28:39,141 epoch 5 - iter 98/146 - loss 0.03124285 - time (sec): 6.32 - samples/sec: 4699.64 - lr: 0.000030 - momentum: 0.000000
2023-10-25 21:28:39,990 epoch 5 - iter 112/146 - loss 0.03154566 - time (sec): 7.17 - samples/sec: 4751.94 - lr: 0.000029 - momentum: 0.000000
2023-10-25 21:28:40,965 epoch 5 - iter 126/146 - loss 0.03179183 - time (sec): 8.15 - samples/sec: 4757.02 - lr: 0.000029 - momentum: 0.000000
2023-10-25 21:28:41,808 epoch 5 - iter 140/146 - loss 0.03049933 - time (sec): 8.99 - samples/sec: 4780.25 - lr: 0.000028 - momentum: 0.000000
2023-10-25 21:28:42,164 ----------------------------------------------------------------------------------------------------
2023-10-25 21:28:42,165 EPOCH 5 done: loss 0.0306 - lr: 0.000028
2023-10-25 21:28:43,083 DEV : loss 0.11159916967153549 - f1-score (micro avg) 0.7301
2023-10-25 21:28:43,088 ----------------------------------------------------------------------------------------------------
2023-10-25 21:28:44,021 epoch 6 - iter 14/146 - loss 0.02453557 - time (sec): 0.93 - samples/sec: 5113.56 - lr: 0.000027 - momentum: 0.000000
2023-10-25 21:28:44,918 epoch 6 - iter 28/146 - loss 0.02735209 - time (sec): 1.83 - samples/sec: 4855.99 - lr: 0.000027 - momentum: 0.000000
2023-10-25 21:28:45,785 epoch 6 - iter 42/146 - loss 0.02225382 - time (sec): 2.70 - samples/sec: 4893.55 - lr: 0.000026 - momentum: 0.000000
2023-10-25 21:28:46,824 epoch 6 - iter 56/146 - loss 0.03298141 - time (sec): 3.73 - samples/sec: 4702.92 - lr: 0.000026 - momentum: 0.000000
2023-10-25 21:28:47,652 epoch 6 - iter 70/146 - loss 0.03018167 - time (sec): 4.56 - samples/sec: 4819.79 - lr: 0.000025 - momentum: 0.000000
2023-10-25 21:28:48,464 epoch 6 - iter 84/146 - loss 0.02855897 - time (sec): 5.38 - samples/sec: 4798.07 - lr: 0.000025 - momentum: 0.000000
2023-10-25 21:28:49,401 epoch 6 - iter 98/146 - loss 0.02624982 - time (sec): 6.31 - samples/sec: 4759.70 - lr: 0.000024 - momentum: 0.000000
2023-10-25 21:28:50,309 epoch 6 - iter 112/146 - loss 0.02634661 - time (sec): 7.22 - samples/sec: 4728.86 - lr: 0.000024 - momentum: 0.000000
2023-10-25 21:28:51,351 epoch 6 - iter 126/146 - loss 0.02455669 - time (sec): 8.26 - samples/sec: 4696.35 - lr: 0.000023 - momentum: 0.000000
2023-10-25 21:28:52,156 epoch 6 - iter 140/146 - loss 0.02413772 - time (sec): 9.07 - samples/sec: 4692.77 - lr: 0.000023 - momentum: 0.000000
2023-10-25 21:28:52,526 ----------------------------------------------------------------------------------------------------
2023-10-25 21:28:52,526 EPOCH 6 done: loss 0.0243 - lr: 0.000023
2023-10-25 21:28:53,437 DEV : loss 0.12900103628635406 - f1-score (micro avg) 0.7484
2023-10-25 21:28:53,442 ----------------------------------------------------------------------------------------------------
2023-10-25 21:28:54,337 epoch 7 - iter 14/146 - loss 0.01901767 - time (sec): 0.89 - samples/sec: 5157.57 - lr: 0.000022 - momentum: 0.000000
2023-10-25 21:28:55,345 epoch 7 - iter 28/146 - loss 0.02538527 - time (sec): 1.90 - samples/sec: 4964.20 - lr: 0.000021 - momentum: 0.000000
2023-10-25 21:28:56,194 epoch 7 - iter 42/146 - loss 0.02265917 - time (sec): 2.75 - samples/sec: 4868.84 - lr: 0.000021 - momentum: 0.000000
2023-10-25 21:28:57,056 epoch 7 - iter 56/146 - loss 0.01889249 - time (sec): 3.61 - samples/sec: 4781.15 - lr: 0.000020 - momentum: 0.000000
2023-10-25 21:28:57,849 epoch 7 - iter 70/146 - loss 0.01700472 - time (sec): 4.41 - samples/sec: 4763.45 - lr: 0.000020 - momentum: 0.000000
2023-10-25 21:28:58,890 epoch 7 - iter 84/146 - loss 0.01723572 - time (sec): 5.45 - samples/sec: 4727.82 - lr: 0.000019 - momentum: 0.000000
2023-10-25 21:28:59,804 epoch 7 - iter 98/146 - loss 0.01698293 - time (sec): 6.36 - samples/sec: 4774.84 - lr: 0.000019 - momentum: 0.000000
2023-10-25 21:29:00,640 epoch 7 - iter 112/146 - loss 0.01540217 - time (sec): 7.20 - samples/sec: 4755.44 - lr: 0.000018 - momentum: 0.000000
2023-10-25 21:29:01,566 epoch 7 - iter 126/146 - loss 0.01515024 - time (sec): 8.12 - samples/sec: 4737.44 - lr: 0.000018 - momentum: 0.000000
2023-10-25 21:29:02,435 epoch 7 - iter 140/146 - loss 0.01548718 - time (sec): 8.99 - samples/sec: 4711.76 - lr: 0.000017 - momentum: 0.000000
2023-10-25 21:29:02,899 ----------------------------------------------------------------------------------------------------
2023-10-25 21:29:02,899 EPOCH 7 done: loss 0.0149 - lr: 0.000017
2023-10-25 21:29:03,821 DEV : loss 0.16139209270477295 - f1-score (micro avg) 0.719
2023-10-25 21:29:03,825 ----------------------------------------------------------------------------------------------------
2023-10-25 21:29:04,777 epoch 8 - iter 14/146 - loss 0.01190115 - time (sec): 0.95 - samples/sec: 4467.82 - lr: 0.000016 - momentum: 0.000000
2023-10-25 21:29:05,705 epoch 8 - iter 28/146 - loss 0.01702526 - time (sec): 1.88 - samples/sec: 4501.00 - lr: 0.000016 - momentum: 0.000000
2023-10-25 21:29:06,547 epoch 8 - iter 42/146 - loss 0.01300344 - time (sec): 2.72 - samples/sec: 4594.14 - lr: 0.000015 - momentum: 0.000000
2023-10-25 21:29:07,500 epoch 8 - iter 56/146 - loss 0.01363888 - time (sec): 3.67 - samples/sec: 4669.93 - lr: 0.000015 - momentum: 0.000000
2023-10-25 21:29:08,334 epoch 8 - iter 70/146 - loss 0.01399870 - time (sec): 4.51 - samples/sec: 4657.71 - lr: 0.000014 - momentum: 0.000000
2023-10-25 21:29:09,169 epoch 8 - iter 84/146 - loss 0.01463633 - time (sec): 5.34 - samples/sec: 4716.96 - lr: 0.000014 - momentum: 0.000000
2023-10-25 21:29:10,006 epoch 8 - iter 98/146 - loss 0.01386733 - time (sec): 6.18 - samples/sec: 4689.65 - lr: 0.000013 - momentum: 0.000000
2023-10-25 21:29:10,942 epoch 8 - iter 112/146 - loss 0.01351418 - time (sec): 7.12 - samples/sec: 4656.21 - lr: 0.000013 - momentum: 0.000000
2023-10-25 21:29:11,872 epoch 8 - iter 126/146 - loss 0.01318788 - time (sec): 8.05 - samples/sec: 4666.02 - lr: 0.000012 - momentum: 0.000000
2023-10-25 21:29:12,874 epoch 8 - iter 140/146 - loss 0.01256842 - time (sec): 9.05 - samples/sec: 4703.36 - lr: 0.000012 - momentum: 0.000000
2023-10-25 21:29:13,263 ----------------------------------------------------------------------------------------------------
2023-10-25 21:29:13,264 EPOCH 8 done: loss 0.0127 - lr: 0.000012
2023-10-25 21:29:14,203 DEV : loss 0.16343142092227936 - f1-score (micro avg) 0.7269
2023-10-25 21:29:14,208 ----------------------------------------------------------------------------------------------------
2023-10-25 21:29:15,182 epoch 9 - iter 14/146 - loss 0.00051275 - time (sec): 0.97 - samples/sec: 5085.26 - lr: 0.000011 - momentum: 0.000000
2023-10-25 21:29:16,024 epoch 9 - iter 28/146 - loss 0.00491903 - time (sec): 1.82 - samples/sec: 5033.56 - lr: 0.000010 - momentum: 0.000000
2023-10-25 21:29:16,859 epoch 9 - iter 42/146 - loss 0.00934029 - time (sec): 2.65 - samples/sec: 4919.50 - lr: 0.000010 - momentum: 0.000000
2023-10-25 21:29:17,799 epoch 9 - iter 56/146 - loss 0.00934967 - time (sec): 3.59 - samples/sec: 4914.97 - lr: 0.000009 - momentum: 0.000000
2023-10-25 21:29:18,748 epoch 9 - iter 70/146 - loss 0.01185195 - time (sec): 4.54 - samples/sec: 4825.98 - lr: 0.000009 - momentum: 0.000000
2023-10-25 21:29:19,899 epoch 9 - iter 84/146 - loss 0.01258572 - time (sec): 5.69 - samples/sec: 4605.52 - lr: 0.000008 - momentum: 0.000000
2023-10-25 21:29:20,792 epoch 9 - iter 98/146 - loss 0.01171045 - time (sec): 6.58 - samples/sec: 4635.70 - lr: 0.000008 - momentum: 0.000000
2023-10-25 21:29:21,692 epoch 9 - iter 112/146 - loss 0.01098300 - time (sec): 7.48 - samples/sec: 4640.80 - lr: 0.000007 - momentum: 0.000000
2023-10-25 21:29:22,569 epoch 9 - iter 126/146 - loss 0.01080796 - time (sec): 8.36 - samples/sec: 4615.82 - lr: 0.000007 - momentum: 0.000000
2023-10-25 21:29:23,467 epoch 9 - iter 140/146 - loss 0.01040519 - time (sec): 9.26 - samples/sec: 4609.56 - lr: 0.000006 - momentum: 0.000000
2023-10-25 21:29:23,818 ----------------------------------------------------------------------------------------------------
2023-10-25 21:29:23,818 EPOCH 9 done: loss 0.0101 - lr: 0.000006
2023-10-25 21:29:24,734 DEV : loss 0.16184502840042114 - f1-score (micro avg) 0.7387
2023-10-25 21:29:24,738 ----------------------------------------------------------------------------------------------------
2023-10-25 21:29:25,626 epoch 10 - iter 14/146 - loss 0.00435948 - time (sec): 0.89 - samples/sec: 4844.74 - lr: 0.000005 - momentum: 0.000000
2023-10-25 21:29:26,494 epoch 10 - iter 28/146 - loss 0.00294552 - time (sec): 1.75 - samples/sec: 4538.68 - lr: 0.000005 - momentum: 0.000000
2023-10-25 21:29:27,388 epoch 10 - iter 42/146 - loss 0.00866389 - time (sec): 2.65 - samples/sec: 4561.33 - lr: 0.000004 - momentum: 0.000000
2023-10-25 21:29:28,184 epoch 10 - iter 56/146 - loss 0.00811777 - time (sec): 3.44 - samples/sec: 4587.61 - lr: 0.000004 - momentum: 0.000000
2023-10-25 21:29:29,168 epoch 10 - iter 70/146 - loss 0.00765822 - time (sec): 4.43 - samples/sec: 4645.83 - lr: 0.000003 - momentum: 0.000000
2023-10-25 21:29:30,062 epoch 10 - iter 84/146 - loss 0.00758295 - time (sec): 5.32 - samples/sec: 4648.84 - lr: 0.000003 - momentum: 0.000000
2023-10-25 21:29:30,892 epoch 10 - iter 98/146 - loss 0.00682628 - time (sec): 6.15 - samples/sec: 4734.92 - lr: 0.000002 - momentum: 0.000000
2023-10-25 21:29:31,880 epoch 10 - iter 112/146 - loss 0.00744500 - time (sec): 7.14 - samples/sec: 4766.54 - lr: 0.000002 - momentum: 0.000000
2023-10-25 21:29:32,788 epoch 10 - iter 126/146 - loss 0.00674236 - time (sec): 8.05 - samples/sec: 4750.00 - lr: 0.000001 - momentum: 0.000000
2023-10-25 21:29:33,723 epoch 10 - iter 140/146 - loss 0.00623926 - time (sec): 8.98 - samples/sec: 4789.04 - lr: 0.000000 - momentum: 0.000000
2023-10-25 21:29:34,035 ----------------------------------------------------------------------------------------------------
2023-10-25 21:29:34,035 EPOCH 10 done: loss 0.0060 - lr: 0.000000
2023-10-25 21:29:34,945 DEV : loss 0.164560928940773 - f1-score (micro avg) 0.7473
2023-10-25 21:29:35,418 ----------------------------------------------------------------------------------------------------
2023-10-25 21:29:35,419 Loading model from best epoch ...
2023-10-25 21:29:37,026 SequenceTagger predicts: Dictionary with 17 tags: O, S-LOC, B-LOC, E-LOC, I-LOC, S-PER, B-PER, E-PER, I-PER, S-ORG, B-ORG, E-ORG, I-ORG, S-HumanProd, B-HumanProd, E-HumanProd, I-HumanProd
2023-10-25 21:29:38,560
Results:
- F-score (micro) 0.7441
- F-score (macro) 0.6586
- Accuracy 0.6156
By class:
precision recall f1-score support
PER 0.7978 0.8506 0.8234 348
LOC 0.6375 0.7816 0.7022 261
ORG 0.4884 0.4038 0.4421 52
HumanProd 0.5862 0.7727 0.6667 22
micro avg 0.7051 0.7877 0.7441 683
macro avg 0.6275 0.7022 0.6586 683
weighted avg 0.7062 0.7877 0.7430 683
2023-10-25 21:29:38,560 ----------------------------------------------------------------------------------------------------
|