File size: 23,834 Bytes
92c7a42 |
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 |
2023-10-16 17:53:10,355 ----------------------------------------------------------------------------------------------------
2023-10-16 17:53:10,356 Model: "SequenceTagger(
(embeddings): TransformerWordEmbeddings(
(model): BertModel(
(embeddings): BertEmbeddings(
(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): 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()
)"
2023-10-16 17:53:10,356 ----------------------------------------------------------------------------------------------------
2023-10-16 17:53:10,356 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
2023-10-16 17:53:10,356 ----------------------------------------------------------------------------------------------------
2023-10-16 17:53:10,356 Train: 1166 sentences
2023-10-16 17:53:10,356 (train_with_dev=False, train_with_test=False)
2023-10-16 17:53:10,356 ----------------------------------------------------------------------------------------------------
2023-10-16 17:53:10,356 Training Params:
2023-10-16 17:53:10,356 - learning_rate: "3e-05"
2023-10-16 17:53:10,356 - mini_batch_size: "4"
2023-10-16 17:53:10,356 - max_epochs: "10"
2023-10-16 17:53:10,356 - shuffle: "True"
2023-10-16 17:53:10,356 ----------------------------------------------------------------------------------------------------
2023-10-16 17:53:10,356 Plugins:
2023-10-16 17:53:10,356 - LinearScheduler | warmup_fraction: '0.1'
2023-10-16 17:53:10,356 ----------------------------------------------------------------------------------------------------
2023-10-16 17:53:10,356 Final evaluation on model from best epoch (best-model.pt)
2023-10-16 17:53:10,356 - metric: "('micro avg', 'f1-score')"
2023-10-16 17:53:10,356 ----------------------------------------------------------------------------------------------------
2023-10-16 17:53:10,356 Computation:
2023-10-16 17:53:10,356 - compute on device: cuda:0
2023-10-16 17:53:10,356 - embedding storage: none
2023-10-16 17:53:10,356 ----------------------------------------------------------------------------------------------------
2023-10-16 17:53:10,356 Model training base path: "hmbench-newseye/fi-dbmdz/bert-base-historic-multilingual-cased-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-1"
2023-10-16 17:53:10,356 ----------------------------------------------------------------------------------------------------
2023-10-16 17:53:10,357 ----------------------------------------------------------------------------------------------------
2023-10-16 17:53:13,018 epoch 1 - iter 29/292 - loss 3.01670138 - time (sec): 2.66 - samples/sec: 1590.03 - lr: 0.000003 - momentum: 0.000000
2023-10-16 17:53:14,948 epoch 1 - iter 58/292 - loss 2.54580984 - time (sec): 4.59 - samples/sec: 2158.32 - lr: 0.000006 - momentum: 0.000000
2023-10-16 17:53:16,645 epoch 1 - iter 87/292 - loss 2.07822180 - time (sec): 6.29 - samples/sec: 2183.16 - lr: 0.000009 - momentum: 0.000000
2023-10-16 17:53:18,313 epoch 1 - iter 116/292 - loss 1.77670253 - time (sec): 7.96 - samples/sec: 2206.58 - lr: 0.000012 - momentum: 0.000000
2023-10-16 17:53:20,188 epoch 1 - iter 145/292 - loss 1.48355404 - time (sec): 9.83 - samples/sec: 2309.09 - lr: 0.000015 - momentum: 0.000000
2023-10-16 17:53:21,905 epoch 1 - iter 174/292 - loss 1.31635425 - time (sec): 11.55 - samples/sec: 2320.51 - lr: 0.000018 - momentum: 0.000000
2023-10-16 17:53:23,540 epoch 1 - iter 203/292 - loss 1.20996793 - time (sec): 13.18 - samples/sec: 2379.01 - lr: 0.000021 - momentum: 0.000000
2023-10-16 17:53:25,115 epoch 1 - iter 232/292 - loss 1.10738448 - time (sec): 14.76 - samples/sec: 2389.64 - lr: 0.000024 - momentum: 0.000000
2023-10-16 17:53:26,860 epoch 1 - iter 261/292 - loss 1.01693990 - time (sec): 16.50 - samples/sec: 2405.80 - lr: 0.000027 - momentum: 0.000000
2023-10-16 17:53:28,671 epoch 1 - iter 290/292 - loss 0.94433148 - time (sec): 18.31 - samples/sec: 2415.89 - lr: 0.000030 - momentum: 0.000000
2023-10-16 17:53:28,769 ----------------------------------------------------------------------------------------------------
2023-10-16 17:53:28,769 EPOCH 1 done: loss 0.9414 - lr: 0.000030
2023-10-16 17:53:29,849 DEV : loss 0.2120908498764038 - f1-score (micro avg) 0.4492
2023-10-16 17:53:29,853 saving best model
2023-10-16 17:53:30,340 ----------------------------------------------------------------------------------------------------
2023-10-16 17:53:32,028 epoch 2 - iter 29/292 - loss 0.18672523 - time (sec): 1.69 - samples/sec: 2637.53 - lr: 0.000030 - momentum: 0.000000
2023-10-16 17:53:33,606 epoch 2 - iter 58/292 - loss 0.22053055 - time (sec): 3.27 - samples/sec: 2719.56 - lr: 0.000029 - momentum: 0.000000
2023-10-16 17:53:35,118 epoch 2 - iter 87/292 - loss 0.23027135 - time (sec): 4.78 - samples/sec: 2749.72 - lr: 0.000029 - momentum: 0.000000
2023-10-16 17:53:36,675 epoch 2 - iter 116/292 - loss 0.20606513 - time (sec): 6.33 - samples/sec: 2752.78 - lr: 0.000029 - momentum: 0.000000
2023-10-16 17:53:38,235 epoch 2 - iter 145/292 - loss 0.20734738 - time (sec): 7.89 - samples/sec: 2671.83 - lr: 0.000028 - momentum: 0.000000
2023-10-16 17:53:39,802 epoch 2 - iter 174/292 - loss 0.20460916 - time (sec): 9.46 - samples/sec: 2668.26 - lr: 0.000028 - momentum: 0.000000
2023-10-16 17:53:41,754 epoch 2 - iter 203/292 - loss 0.21184347 - time (sec): 11.41 - samples/sec: 2680.86 - lr: 0.000028 - momentum: 0.000000
2023-10-16 17:53:43,491 epoch 2 - iter 232/292 - loss 0.20796678 - time (sec): 13.15 - samples/sec: 2667.97 - lr: 0.000027 - momentum: 0.000000
2023-10-16 17:53:45,327 epoch 2 - iter 261/292 - loss 0.20176642 - time (sec): 14.99 - samples/sec: 2667.45 - lr: 0.000027 - momentum: 0.000000
2023-10-16 17:53:46,957 epoch 2 - iter 290/292 - loss 0.19939226 - time (sec): 16.62 - samples/sec: 2657.51 - lr: 0.000027 - momentum: 0.000000
2023-10-16 17:53:47,055 ----------------------------------------------------------------------------------------------------
2023-10-16 17:53:47,056 EPOCH 2 done: loss 0.1997 - lr: 0.000027
2023-10-16 17:53:48,273 DEV : loss 0.13168571889400482 - f1-score (micro avg) 0.6189
2023-10-16 17:53:48,277 saving best model
2023-10-16 17:53:48,826 ----------------------------------------------------------------------------------------------------
2023-10-16 17:53:50,649 epoch 3 - iter 29/292 - loss 0.15474519 - time (sec): 1.82 - samples/sec: 3032.01 - lr: 0.000026 - momentum: 0.000000
2023-10-16 17:53:52,245 epoch 3 - iter 58/292 - loss 0.13264738 - time (sec): 3.41 - samples/sec: 2593.52 - lr: 0.000026 - momentum: 0.000000
2023-10-16 17:53:53,805 epoch 3 - iter 87/292 - loss 0.13055608 - time (sec): 4.97 - samples/sec: 2655.91 - lr: 0.000026 - momentum: 0.000000
2023-10-16 17:53:55,758 epoch 3 - iter 116/292 - loss 0.12609661 - time (sec): 6.93 - samples/sec: 2574.92 - lr: 0.000025 - momentum: 0.000000
2023-10-16 17:53:57,465 epoch 3 - iter 145/292 - loss 0.11999825 - time (sec): 8.63 - samples/sec: 2669.39 - lr: 0.000025 - momentum: 0.000000
2023-10-16 17:53:59,053 epoch 3 - iter 174/292 - loss 0.11593147 - time (sec): 10.22 - samples/sec: 2668.14 - lr: 0.000025 - momentum: 0.000000
2023-10-16 17:54:00,553 epoch 3 - iter 203/292 - loss 0.11595612 - time (sec): 11.72 - samples/sec: 2624.30 - lr: 0.000024 - momentum: 0.000000
2023-10-16 17:54:02,144 epoch 3 - iter 232/292 - loss 0.11529465 - time (sec): 13.31 - samples/sec: 2613.51 - lr: 0.000024 - momentum: 0.000000
2023-10-16 17:54:03,755 epoch 3 - iter 261/292 - loss 0.11497819 - time (sec): 14.92 - samples/sec: 2629.55 - lr: 0.000024 - momentum: 0.000000
2023-10-16 17:54:05,520 epoch 3 - iter 290/292 - loss 0.11247568 - time (sec): 16.69 - samples/sec: 2648.07 - lr: 0.000023 - momentum: 0.000000
2023-10-16 17:54:05,615 ----------------------------------------------------------------------------------------------------
2023-10-16 17:54:05,615 EPOCH 3 done: loss 0.1132 - lr: 0.000023
2023-10-16 17:54:06,825 DEV : loss 0.12188015133142471 - f1-score (micro avg) 0.7049
2023-10-16 17:54:06,829 saving best model
2023-10-16 17:54:07,368 ----------------------------------------------------------------------------------------------------
2023-10-16 17:54:09,077 epoch 4 - iter 29/292 - loss 0.07659714 - time (sec): 1.71 - samples/sec: 2724.79 - lr: 0.000023 - momentum: 0.000000
2023-10-16 17:54:10,758 epoch 4 - iter 58/292 - loss 0.08198423 - time (sec): 3.39 - samples/sec: 2706.29 - lr: 0.000023 - momentum: 0.000000
2023-10-16 17:54:12,342 epoch 4 - iter 87/292 - loss 0.07174156 - time (sec): 4.97 - samples/sec: 2704.35 - lr: 0.000022 - momentum: 0.000000
2023-10-16 17:54:14,000 epoch 4 - iter 116/292 - loss 0.07292734 - time (sec): 6.63 - samples/sec: 2706.79 - lr: 0.000022 - momentum: 0.000000
2023-10-16 17:54:15,587 epoch 4 - iter 145/292 - loss 0.07248743 - time (sec): 8.22 - samples/sec: 2666.65 - lr: 0.000022 - momentum: 0.000000
2023-10-16 17:54:17,256 epoch 4 - iter 174/292 - loss 0.07455389 - time (sec): 9.88 - samples/sec: 2672.34 - lr: 0.000021 - momentum: 0.000000
2023-10-16 17:54:18,864 epoch 4 - iter 203/292 - loss 0.07414268 - time (sec): 11.49 - samples/sec: 2648.26 - lr: 0.000021 - momentum: 0.000000
2023-10-16 17:54:20,561 epoch 4 - iter 232/292 - loss 0.07838926 - time (sec): 13.19 - samples/sec: 2687.21 - lr: 0.000021 - momentum: 0.000000
2023-10-16 17:54:22,083 epoch 4 - iter 261/292 - loss 0.07356868 - time (sec): 14.71 - samples/sec: 2688.42 - lr: 0.000020 - momentum: 0.000000
2023-10-16 17:54:23,717 epoch 4 - iter 290/292 - loss 0.07163331 - time (sec): 16.35 - samples/sec: 2709.19 - lr: 0.000020 - momentum: 0.000000
2023-10-16 17:54:23,801 ----------------------------------------------------------------------------------------------------
2023-10-16 17:54:23,801 EPOCH 4 done: loss 0.0721 - lr: 0.000020
2023-10-16 17:54:24,988 DEV : loss 0.11146893352270126 - f1-score (micro avg) 0.7448
2023-10-16 17:54:24,992 saving best model
2023-10-16 17:54:25,594 ----------------------------------------------------------------------------------------------------
2023-10-16 17:54:27,306 epoch 5 - iter 29/292 - loss 0.05024086 - time (sec): 1.71 - samples/sec: 2953.50 - lr: 0.000020 - momentum: 0.000000
2023-10-16 17:54:28,952 epoch 5 - iter 58/292 - loss 0.05493652 - time (sec): 3.36 - samples/sec: 2757.32 - lr: 0.000019 - momentum: 0.000000
2023-10-16 17:54:30,633 epoch 5 - iter 87/292 - loss 0.05660584 - time (sec): 5.04 - samples/sec: 2735.59 - lr: 0.000019 - momentum: 0.000000
2023-10-16 17:54:32,295 epoch 5 - iter 116/292 - loss 0.05164633 - time (sec): 6.70 - samples/sec: 2713.03 - lr: 0.000019 - momentum: 0.000000
2023-10-16 17:54:34,049 epoch 5 - iter 145/292 - loss 0.04760992 - time (sec): 8.45 - samples/sec: 2700.32 - lr: 0.000018 - momentum: 0.000000
2023-10-16 17:54:35,733 epoch 5 - iter 174/292 - loss 0.05141896 - time (sec): 10.14 - samples/sec: 2731.30 - lr: 0.000018 - momentum: 0.000000
2023-10-16 17:54:37,256 epoch 5 - iter 203/292 - loss 0.05190871 - time (sec): 11.66 - samples/sec: 2715.80 - lr: 0.000018 - momentum: 0.000000
2023-10-16 17:54:38,859 epoch 5 - iter 232/292 - loss 0.05203336 - time (sec): 13.26 - samples/sec: 2699.38 - lr: 0.000017 - momentum: 0.000000
2023-10-16 17:54:40,414 epoch 5 - iter 261/292 - loss 0.05031939 - time (sec): 14.82 - samples/sec: 2694.27 - lr: 0.000017 - momentum: 0.000000
2023-10-16 17:54:42,108 epoch 5 - iter 290/292 - loss 0.05098912 - time (sec): 16.51 - samples/sec: 2682.83 - lr: 0.000017 - momentum: 0.000000
2023-10-16 17:54:42,199 ----------------------------------------------------------------------------------------------------
2023-10-16 17:54:42,200 EPOCH 5 done: loss 0.0516 - lr: 0.000017
2023-10-16 17:54:43,417 DEV : loss 0.12832774221897125 - f1-score (micro avg) 0.73
2023-10-16 17:54:43,422 ----------------------------------------------------------------------------------------------------
2023-10-16 17:54:45,097 epoch 6 - iter 29/292 - loss 0.05950145 - time (sec): 1.67 - samples/sec: 2499.26 - lr: 0.000016 - momentum: 0.000000
2023-10-16 17:54:46,754 epoch 6 - iter 58/292 - loss 0.04628346 - time (sec): 3.33 - samples/sec: 2599.83 - lr: 0.000016 - momentum: 0.000000
2023-10-16 17:54:48,212 epoch 6 - iter 87/292 - loss 0.04890591 - time (sec): 4.79 - samples/sec: 2578.40 - lr: 0.000016 - momentum: 0.000000
2023-10-16 17:54:49,916 epoch 6 - iter 116/292 - loss 0.04392787 - time (sec): 6.49 - samples/sec: 2612.30 - lr: 0.000015 - momentum: 0.000000
2023-10-16 17:54:51,424 epoch 6 - iter 145/292 - loss 0.04177698 - time (sec): 8.00 - samples/sec: 2569.81 - lr: 0.000015 - momentum: 0.000000
2023-10-16 17:54:53,231 epoch 6 - iter 174/292 - loss 0.04185515 - time (sec): 9.81 - samples/sec: 2600.73 - lr: 0.000015 - momentum: 0.000000
2023-10-16 17:54:54,950 epoch 6 - iter 203/292 - loss 0.04627777 - time (sec): 11.53 - samples/sec: 2669.00 - lr: 0.000014 - momentum: 0.000000
2023-10-16 17:54:56,490 epoch 6 - iter 232/292 - loss 0.04466345 - time (sec): 13.07 - samples/sec: 2675.35 - lr: 0.000014 - momentum: 0.000000
2023-10-16 17:54:58,193 epoch 6 - iter 261/292 - loss 0.04324020 - time (sec): 14.77 - samples/sec: 2697.60 - lr: 0.000014 - momentum: 0.000000
2023-10-16 17:54:59,804 epoch 6 - iter 290/292 - loss 0.04093259 - time (sec): 16.38 - samples/sec: 2700.29 - lr: 0.000013 - momentum: 0.000000
2023-10-16 17:54:59,896 ----------------------------------------------------------------------------------------------------
2023-10-16 17:54:59,897 EPOCH 6 done: loss 0.0408 - lr: 0.000013
2023-10-16 17:55:01,125 DEV : loss 0.14976172149181366 - f1-score (micro avg) 0.74
2023-10-16 17:55:01,129 ----------------------------------------------------------------------------------------------------
2023-10-16 17:55:02,678 epoch 7 - iter 29/292 - loss 0.02344868 - time (sec): 1.55 - samples/sec: 2823.98 - lr: 0.000013 - momentum: 0.000000
2023-10-16 17:55:04,236 epoch 7 - iter 58/292 - loss 0.01915777 - time (sec): 3.11 - samples/sec: 2719.21 - lr: 0.000013 - momentum: 0.000000
2023-10-16 17:55:05,773 epoch 7 - iter 87/292 - loss 0.03000931 - time (sec): 4.64 - samples/sec: 2680.51 - lr: 0.000012 - momentum: 0.000000
2023-10-16 17:55:07,689 epoch 7 - iter 116/292 - loss 0.03479154 - time (sec): 6.56 - samples/sec: 2718.57 - lr: 0.000012 - momentum: 0.000000
2023-10-16 17:55:09,461 epoch 7 - iter 145/292 - loss 0.03189663 - time (sec): 8.33 - samples/sec: 2703.43 - lr: 0.000012 - momentum: 0.000000
2023-10-16 17:55:11,006 epoch 7 - iter 174/292 - loss 0.03017231 - time (sec): 9.88 - samples/sec: 2650.72 - lr: 0.000011 - momentum: 0.000000
2023-10-16 17:55:12,695 epoch 7 - iter 203/292 - loss 0.02853514 - time (sec): 11.56 - samples/sec: 2617.59 - lr: 0.000011 - momentum: 0.000000
2023-10-16 17:55:14,460 epoch 7 - iter 232/292 - loss 0.03262212 - time (sec): 13.33 - samples/sec: 2637.76 - lr: 0.000011 - momentum: 0.000000
2023-10-16 17:55:16,144 epoch 7 - iter 261/292 - loss 0.03145605 - time (sec): 15.01 - samples/sec: 2643.78 - lr: 0.000010 - momentum: 0.000000
2023-10-16 17:55:17,765 epoch 7 - iter 290/292 - loss 0.03161094 - time (sec): 16.63 - samples/sec: 2662.24 - lr: 0.000010 - momentum: 0.000000
2023-10-16 17:55:17,864 ----------------------------------------------------------------------------------------------------
2023-10-16 17:55:17,864 EPOCH 7 done: loss 0.0317 - lr: 0.000010
2023-10-16 17:55:19,109 DEV : loss 0.15802569687366486 - f1-score (micro avg) 0.7669
2023-10-16 17:55:19,115 saving best model
2023-10-16 17:55:19,736 ----------------------------------------------------------------------------------------------------
2023-10-16 17:55:21,251 epoch 8 - iter 29/292 - loss 0.02308704 - time (sec): 1.51 - samples/sec: 2423.33 - lr: 0.000010 - momentum: 0.000000
2023-10-16 17:55:23,003 epoch 8 - iter 58/292 - loss 0.01751107 - time (sec): 3.27 - samples/sec: 2645.43 - lr: 0.000009 - momentum: 0.000000
2023-10-16 17:55:24,636 epoch 8 - iter 87/292 - loss 0.01868607 - time (sec): 4.90 - samples/sec: 2645.19 - lr: 0.000009 - momentum: 0.000000
2023-10-16 17:55:26,347 epoch 8 - iter 116/292 - loss 0.01794405 - time (sec): 6.61 - samples/sec: 2669.32 - lr: 0.000009 - momentum: 0.000000
2023-10-16 17:55:28,103 epoch 8 - iter 145/292 - loss 0.02452178 - time (sec): 8.37 - samples/sec: 2670.62 - lr: 0.000008 - momentum: 0.000000
2023-10-16 17:55:29,621 epoch 8 - iter 174/292 - loss 0.02379010 - time (sec): 9.88 - samples/sec: 2633.50 - lr: 0.000008 - momentum: 0.000000
2023-10-16 17:55:31,201 epoch 8 - iter 203/292 - loss 0.02414199 - time (sec): 11.46 - samples/sec: 2615.66 - lr: 0.000008 - momentum: 0.000000
2023-10-16 17:55:32,900 epoch 8 - iter 232/292 - loss 0.02725617 - time (sec): 13.16 - samples/sec: 2616.60 - lr: 0.000007 - momentum: 0.000000
2023-10-16 17:55:34,529 epoch 8 - iter 261/292 - loss 0.02585742 - time (sec): 14.79 - samples/sec: 2621.20 - lr: 0.000007 - momentum: 0.000000
2023-10-16 17:55:36,363 epoch 8 - iter 290/292 - loss 0.02434377 - time (sec): 16.63 - samples/sec: 2657.48 - lr: 0.000007 - momentum: 0.000000
2023-10-16 17:55:36,485 ----------------------------------------------------------------------------------------------------
2023-10-16 17:55:36,485 EPOCH 8 done: loss 0.0242 - lr: 0.000007
2023-10-16 17:55:37,720 DEV : loss 0.16654258966445923 - f1-score (micro avg) 0.742
2023-10-16 17:55:37,725 ----------------------------------------------------------------------------------------------------
2023-10-16 17:55:39,403 epoch 9 - iter 29/292 - loss 0.01417823 - time (sec): 1.68 - samples/sec: 2715.08 - lr: 0.000006 - momentum: 0.000000
2023-10-16 17:55:41,051 epoch 9 - iter 58/292 - loss 0.01232230 - time (sec): 3.33 - samples/sec: 2616.24 - lr: 0.000006 - momentum: 0.000000
2023-10-16 17:55:42,622 epoch 9 - iter 87/292 - loss 0.01376009 - time (sec): 4.90 - samples/sec: 2518.21 - lr: 0.000006 - momentum: 0.000000
2023-10-16 17:55:44,385 epoch 9 - iter 116/292 - loss 0.02555969 - time (sec): 6.66 - samples/sec: 2645.87 - lr: 0.000005 - momentum: 0.000000
2023-10-16 17:55:46,074 epoch 9 - iter 145/292 - loss 0.02158313 - time (sec): 8.35 - samples/sec: 2653.79 - lr: 0.000005 - momentum: 0.000000
2023-10-16 17:55:47,637 epoch 9 - iter 174/292 - loss 0.02019502 - time (sec): 9.91 - samples/sec: 2641.31 - lr: 0.000005 - momentum: 0.000000
2023-10-16 17:55:49,335 epoch 9 - iter 203/292 - loss 0.02252725 - time (sec): 11.61 - samples/sec: 2620.54 - lr: 0.000004 - momentum: 0.000000
2023-10-16 17:55:50,943 epoch 9 - iter 232/292 - loss 0.02172381 - time (sec): 13.22 - samples/sec: 2623.35 - lr: 0.000004 - momentum: 0.000000
2023-10-16 17:55:52,771 epoch 9 - iter 261/292 - loss 0.01959173 - time (sec): 15.05 - samples/sec: 2637.08 - lr: 0.000004 - momentum: 0.000000
2023-10-16 17:55:54,506 epoch 9 - iter 290/292 - loss 0.01884339 - time (sec): 16.78 - samples/sec: 2641.24 - lr: 0.000003 - momentum: 0.000000
2023-10-16 17:55:54,588 ----------------------------------------------------------------------------------------------------
2023-10-16 17:55:54,588 EPOCH 9 done: loss 0.0188 - lr: 0.000003
2023-10-16 17:55:55,813 DEV : loss 0.1669674515724182 - f1-score (micro avg) 0.7447
2023-10-16 17:55:55,819 ----------------------------------------------------------------------------------------------------
2023-10-16 17:55:57,419 epoch 10 - iter 29/292 - loss 0.00817807 - time (sec): 1.60 - samples/sec: 2510.45 - lr: 0.000003 - momentum: 0.000000
2023-10-16 17:55:59,136 epoch 10 - iter 58/292 - loss 0.00727258 - time (sec): 3.32 - samples/sec: 2788.28 - lr: 0.000003 - momentum: 0.000000
2023-10-16 17:56:01,145 epoch 10 - iter 87/292 - loss 0.01631169 - time (sec): 5.33 - samples/sec: 2808.60 - lr: 0.000002 - momentum: 0.000000
2023-10-16 17:56:02,812 epoch 10 - iter 116/292 - loss 0.01409848 - time (sec): 6.99 - samples/sec: 2845.86 - lr: 0.000002 - momentum: 0.000000
2023-10-16 17:56:04,508 epoch 10 - iter 145/292 - loss 0.01324680 - time (sec): 8.69 - samples/sec: 2765.70 - lr: 0.000002 - momentum: 0.000000
2023-10-16 17:56:06,061 epoch 10 - iter 174/292 - loss 0.01866032 - time (sec): 10.24 - samples/sec: 2731.65 - lr: 0.000001 - momentum: 0.000000
2023-10-16 17:56:07,607 epoch 10 - iter 203/292 - loss 0.01831901 - time (sec): 11.79 - samples/sec: 2690.65 - lr: 0.000001 - momentum: 0.000000
2023-10-16 17:56:09,171 epoch 10 - iter 232/292 - loss 0.01714936 - time (sec): 13.35 - samples/sec: 2679.21 - lr: 0.000001 - momentum: 0.000000
2023-10-16 17:56:10,757 epoch 10 - iter 261/292 - loss 0.01673991 - time (sec): 14.94 - samples/sec: 2656.88 - lr: 0.000000 - momentum: 0.000000
2023-10-16 17:56:12,321 epoch 10 - iter 290/292 - loss 0.01661350 - time (sec): 16.50 - samples/sec: 2666.97 - lr: 0.000000 - momentum: 0.000000
2023-10-16 17:56:12,466 ----------------------------------------------------------------------------------------------------
2023-10-16 17:56:12,466 EPOCH 10 done: loss 0.0165 - lr: 0.000000
2023-10-16 17:56:13,697 DEV : loss 0.1702689379453659 - f1-score (micro avg) 0.7368
2023-10-16 17:56:14,198 ----------------------------------------------------------------------------------------------------
2023-10-16 17:56:14,199 Loading model from best epoch ...
2023-10-16 17:56:16,142 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
2023-10-16 17:56:18,744
Results:
- F-score (micro) 0.7574
- F-score (macro) 0.6866
- Accuracy 0.6335
By class:
precision recall f1-score support
PER 0.7936 0.8506 0.8211 348
LOC 0.6677 0.8238 0.7376 261
ORG 0.4130 0.3654 0.3878 52
HumanProd 0.7826 0.8182 0.8000 22
micro avg 0.7173 0.8023 0.7574 683
macro avg 0.6642 0.7145 0.6866 683
weighted avg 0.7161 0.8023 0.7555 683
2023-10-16 17:56:18,744 ----------------------------------------------------------------------------------------------------
|