File size: 23,736 Bytes
f2e9927 |
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 |
2023-10-13 21:37:23,966 ----------------------------------------------------------------------------------------------------
2023-10-13 21:37:23,967 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=13, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-13 21:37:23,967 ----------------------------------------------------------------------------------------------------
2023-10-13 21:37:23,967 MultiCorpus: 7936 train + 992 dev + 992 test sentences
- NER_ICDAR_EUROPEANA Corpus: 7936 train + 992 dev + 992 test sentences - /root/.flair/datasets/ner_icdar_europeana/fr
2023-10-13 21:37:23,967 ----------------------------------------------------------------------------------------------------
2023-10-13 21:37:23,967 Train: 7936 sentences
2023-10-13 21:37:23,967 (train_with_dev=False, train_with_test=False)
2023-10-13 21:37:23,967 ----------------------------------------------------------------------------------------------------
2023-10-13 21:37:23,967 Training Params:
2023-10-13 21:37:23,967 - learning_rate: "5e-05"
2023-10-13 21:37:23,968 - mini_batch_size: "8"
2023-10-13 21:37:23,968 - max_epochs: "10"
2023-10-13 21:37:23,968 - shuffle: "True"
2023-10-13 21:37:23,968 ----------------------------------------------------------------------------------------------------
2023-10-13 21:37:23,968 Plugins:
2023-10-13 21:37:23,968 - LinearScheduler | warmup_fraction: '0.1'
2023-10-13 21:37:23,968 ----------------------------------------------------------------------------------------------------
2023-10-13 21:37:23,968 Final evaluation on model from best epoch (best-model.pt)
2023-10-13 21:37:23,968 - metric: "('micro avg', 'f1-score')"
2023-10-13 21:37:23,968 ----------------------------------------------------------------------------------------------------
2023-10-13 21:37:23,968 Computation:
2023-10-13 21:37:23,968 - compute on device: cuda:0
2023-10-13 21:37:23,968 - embedding storage: none
2023-10-13 21:37:23,968 ----------------------------------------------------------------------------------------------------
2023-10-13 21:37:23,968 Model training base path: "hmbench-icdar/fr-dbmdz/bert-base-historic-multilingual-cased-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-1"
2023-10-13 21:37:23,968 ----------------------------------------------------------------------------------------------------
2023-10-13 21:37:23,968 ----------------------------------------------------------------------------------------------------
2023-10-13 21:37:29,890 epoch 1 - iter 99/992 - loss 1.91261312 - time (sec): 5.92 - samples/sec: 2715.74 - lr: 0.000005 - momentum: 0.000000
2023-10-13 21:37:35,908 epoch 1 - iter 198/992 - loss 1.13445665 - time (sec): 11.94 - samples/sec: 2727.88 - lr: 0.000010 - momentum: 0.000000
2023-10-13 21:37:41,742 epoch 1 - iter 297/992 - loss 0.83788670 - time (sec): 17.77 - samples/sec: 2767.01 - lr: 0.000015 - momentum: 0.000000
2023-10-13 21:37:47,441 epoch 1 - iter 396/992 - loss 0.67694574 - time (sec): 23.47 - samples/sec: 2781.96 - lr: 0.000020 - momentum: 0.000000
2023-10-13 21:37:53,157 epoch 1 - iter 495/992 - loss 0.57510064 - time (sec): 29.19 - samples/sec: 2789.19 - lr: 0.000025 - momentum: 0.000000
2023-10-13 21:37:58,914 epoch 1 - iter 594/992 - loss 0.50275111 - time (sec): 34.95 - samples/sec: 2791.76 - lr: 0.000030 - momentum: 0.000000
2023-10-13 21:38:04,772 epoch 1 - iter 693/992 - loss 0.44880423 - time (sec): 40.80 - samples/sec: 2810.61 - lr: 0.000035 - momentum: 0.000000
2023-10-13 21:38:11,079 epoch 1 - iter 792/992 - loss 0.40684356 - time (sec): 47.11 - samples/sec: 2805.49 - lr: 0.000040 - momentum: 0.000000
2023-10-13 21:38:16,874 epoch 1 - iter 891/992 - loss 0.37820026 - time (sec): 52.90 - samples/sec: 2799.25 - lr: 0.000045 - momentum: 0.000000
2023-10-13 21:38:22,677 epoch 1 - iter 990/992 - loss 0.35457753 - time (sec): 58.71 - samples/sec: 2789.75 - lr: 0.000050 - momentum: 0.000000
2023-10-13 21:38:22,786 ----------------------------------------------------------------------------------------------------
2023-10-13 21:38:22,786 EPOCH 1 done: loss 0.3542 - lr: 0.000050
2023-10-13 21:38:26,277 DEV : loss 0.09475447982549667 - f1-score (micro avg) 0.6958
2023-10-13 21:38:26,298 saving best model
2023-10-13 21:38:26,721 ----------------------------------------------------------------------------------------------------
2023-10-13 21:38:32,814 epoch 2 - iter 99/992 - loss 0.12012769 - time (sec): 6.09 - samples/sec: 2833.35 - lr: 0.000049 - momentum: 0.000000
2023-10-13 21:38:38,526 epoch 2 - iter 198/992 - loss 0.11631000 - time (sec): 11.80 - samples/sec: 2746.93 - lr: 0.000049 - momentum: 0.000000
2023-10-13 21:38:44,938 epoch 2 - iter 297/992 - loss 0.11145763 - time (sec): 18.22 - samples/sec: 2735.81 - lr: 0.000048 - momentum: 0.000000
2023-10-13 21:38:50,662 epoch 2 - iter 396/992 - loss 0.10881005 - time (sec): 23.94 - samples/sec: 2704.74 - lr: 0.000048 - momentum: 0.000000
2023-10-13 21:38:56,694 epoch 2 - iter 495/992 - loss 0.10846778 - time (sec): 29.97 - samples/sec: 2740.68 - lr: 0.000047 - momentum: 0.000000
2023-10-13 21:39:02,441 epoch 2 - iter 594/992 - loss 0.10567866 - time (sec): 35.72 - samples/sec: 2746.65 - lr: 0.000047 - momentum: 0.000000
2023-10-13 21:39:08,610 epoch 2 - iter 693/992 - loss 0.10559153 - time (sec): 41.89 - samples/sec: 2735.81 - lr: 0.000046 - momentum: 0.000000
2023-10-13 21:39:14,672 epoch 2 - iter 792/992 - loss 0.10488835 - time (sec): 47.95 - samples/sec: 2722.53 - lr: 0.000046 - momentum: 0.000000
2023-10-13 21:39:20,740 epoch 2 - iter 891/992 - loss 0.10442316 - time (sec): 54.02 - samples/sec: 2727.03 - lr: 0.000045 - momentum: 0.000000
2023-10-13 21:39:26,803 epoch 2 - iter 990/992 - loss 0.10329774 - time (sec): 60.08 - samples/sec: 2725.62 - lr: 0.000044 - momentum: 0.000000
2023-10-13 21:39:26,916 ----------------------------------------------------------------------------------------------------
2023-10-13 21:39:26,916 EPOCH 2 done: loss 0.1033 - lr: 0.000044
2023-10-13 21:39:30,341 DEV : loss 0.08205121755599976 - f1-score (micro avg) 0.7129
2023-10-13 21:39:30,361 saving best model
2023-10-13 21:39:30,861 ----------------------------------------------------------------------------------------------------
2023-10-13 21:39:36,742 epoch 3 - iter 99/992 - loss 0.07082610 - time (sec): 5.87 - samples/sec: 2789.48 - lr: 0.000044 - momentum: 0.000000
2023-10-13 21:39:42,620 epoch 3 - iter 198/992 - loss 0.06839682 - time (sec): 11.75 - samples/sec: 2713.38 - lr: 0.000043 - momentum: 0.000000
2023-10-13 21:39:48,758 epoch 3 - iter 297/992 - loss 0.06531227 - time (sec): 17.89 - samples/sec: 2763.77 - lr: 0.000043 - momentum: 0.000000
2023-10-13 21:39:54,864 epoch 3 - iter 396/992 - loss 0.06974993 - time (sec): 23.99 - samples/sec: 2777.66 - lr: 0.000042 - momentum: 0.000000
2023-10-13 21:40:00,891 epoch 3 - iter 495/992 - loss 0.07037521 - time (sec): 30.02 - samples/sec: 2759.70 - lr: 0.000042 - momentum: 0.000000
2023-10-13 21:40:06,636 epoch 3 - iter 594/992 - loss 0.07072352 - time (sec): 35.77 - samples/sec: 2768.83 - lr: 0.000041 - momentum: 0.000000
2023-10-13 21:40:12,505 epoch 3 - iter 693/992 - loss 0.07026764 - time (sec): 41.64 - samples/sec: 2761.90 - lr: 0.000041 - momentum: 0.000000
2023-10-13 21:40:18,247 epoch 3 - iter 792/992 - loss 0.07260697 - time (sec): 47.38 - samples/sec: 2775.98 - lr: 0.000040 - momentum: 0.000000
2023-10-13 21:40:24,041 epoch 3 - iter 891/992 - loss 0.07256251 - time (sec): 53.17 - samples/sec: 2778.82 - lr: 0.000039 - momentum: 0.000000
2023-10-13 21:40:29,699 epoch 3 - iter 990/992 - loss 0.07225091 - time (sec): 58.83 - samples/sec: 2780.17 - lr: 0.000039 - momentum: 0.000000
2023-10-13 21:40:29,815 ----------------------------------------------------------------------------------------------------
2023-10-13 21:40:29,815 EPOCH 3 done: loss 0.0722 - lr: 0.000039
2023-10-13 21:40:33,902 DEV : loss 0.10864556580781937 - f1-score (micro avg) 0.758
2023-10-13 21:40:33,939 saving best model
2023-10-13 21:40:34,466 ----------------------------------------------------------------------------------------------------
2023-10-13 21:40:40,380 epoch 4 - iter 99/992 - loss 0.05150709 - time (sec): 5.91 - samples/sec: 2796.73 - lr: 0.000038 - momentum: 0.000000
2023-10-13 21:40:46,332 epoch 4 - iter 198/992 - loss 0.04958120 - time (sec): 11.86 - samples/sec: 2802.33 - lr: 0.000038 - momentum: 0.000000
2023-10-13 21:40:51,999 epoch 4 - iter 297/992 - loss 0.05125257 - time (sec): 17.53 - samples/sec: 2798.55 - lr: 0.000037 - momentum: 0.000000
2023-10-13 21:40:57,854 epoch 4 - iter 396/992 - loss 0.04922105 - time (sec): 23.39 - samples/sec: 2782.53 - lr: 0.000037 - momentum: 0.000000
2023-10-13 21:41:03,725 epoch 4 - iter 495/992 - loss 0.04834639 - time (sec): 29.26 - samples/sec: 2783.19 - lr: 0.000036 - momentum: 0.000000
2023-10-13 21:41:09,534 epoch 4 - iter 594/992 - loss 0.04924934 - time (sec): 35.07 - samples/sec: 2786.12 - lr: 0.000036 - momentum: 0.000000
2023-10-13 21:41:15,489 epoch 4 - iter 693/992 - loss 0.04948734 - time (sec): 41.02 - samples/sec: 2778.64 - lr: 0.000035 - momentum: 0.000000
2023-10-13 21:41:21,333 epoch 4 - iter 792/992 - loss 0.04908108 - time (sec): 46.87 - samples/sec: 2773.09 - lr: 0.000034 - momentum: 0.000000
2023-10-13 21:41:27,452 epoch 4 - iter 891/992 - loss 0.04886483 - time (sec): 52.98 - samples/sec: 2765.98 - lr: 0.000034 - momentum: 0.000000
2023-10-13 21:41:33,681 epoch 4 - iter 990/992 - loss 0.04998705 - time (sec): 59.21 - samples/sec: 2765.34 - lr: 0.000033 - momentum: 0.000000
2023-10-13 21:41:33,794 ----------------------------------------------------------------------------------------------------
2023-10-13 21:41:33,794 EPOCH 4 done: loss 0.0499 - lr: 0.000033
2023-10-13 21:41:37,201 DEV : loss 0.11597025394439697 - f1-score (micro avg) 0.7675
2023-10-13 21:41:37,222 saving best model
2023-10-13 21:41:37,774 ----------------------------------------------------------------------------------------------------
2023-10-13 21:41:43,640 epoch 5 - iter 99/992 - loss 0.04977178 - time (sec): 5.86 - samples/sec: 2838.67 - lr: 0.000033 - momentum: 0.000000
2023-10-13 21:41:49,509 epoch 5 - iter 198/992 - loss 0.04238345 - time (sec): 11.73 - samples/sec: 2809.15 - lr: 0.000032 - momentum: 0.000000
2023-10-13 21:41:55,313 epoch 5 - iter 297/992 - loss 0.04044960 - time (sec): 17.53 - samples/sec: 2809.68 - lr: 0.000032 - momentum: 0.000000
2023-10-13 21:42:01,758 epoch 5 - iter 396/992 - loss 0.03908575 - time (sec): 23.98 - samples/sec: 2780.41 - lr: 0.000031 - momentum: 0.000000
2023-10-13 21:42:07,657 epoch 5 - iter 495/992 - loss 0.03930276 - time (sec): 29.88 - samples/sec: 2784.79 - lr: 0.000031 - momentum: 0.000000
2023-10-13 21:42:13,458 epoch 5 - iter 594/992 - loss 0.03815672 - time (sec): 35.68 - samples/sec: 2781.01 - lr: 0.000030 - momentum: 0.000000
2023-10-13 21:42:19,292 epoch 5 - iter 693/992 - loss 0.03945281 - time (sec): 41.51 - samples/sec: 2772.88 - lr: 0.000029 - momentum: 0.000000
2023-10-13 21:42:24,988 epoch 5 - iter 792/992 - loss 0.03857818 - time (sec): 47.21 - samples/sec: 2787.22 - lr: 0.000029 - momentum: 0.000000
2023-10-13 21:42:31,007 epoch 5 - iter 891/992 - loss 0.03988100 - time (sec): 53.23 - samples/sec: 2781.52 - lr: 0.000028 - momentum: 0.000000
2023-10-13 21:42:36,766 epoch 5 - iter 990/992 - loss 0.04081192 - time (sec): 58.99 - samples/sec: 2773.33 - lr: 0.000028 - momentum: 0.000000
2023-10-13 21:42:36,903 ----------------------------------------------------------------------------------------------------
2023-10-13 21:42:36,903 EPOCH 5 done: loss 0.0408 - lr: 0.000028
2023-10-13 21:42:41,542 DEV : loss 0.13763324916362762 - f1-score (micro avg) 0.7589
2023-10-13 21:42:41,580 ----------------------------------------------------------------------------------------------------
2023-10-13 21:42:47,750 epoch 6 - iter 99/992 - loss 0.02422172 - time (sec): 6.17 - samples/sec: 2688.98 - lr: 0.000027 - momentum: 0.000000
2023-10-13 21:42:53,863 epoch 6 - iter 198/992 - loss 0.02863550 - time (sec): 12.28 - samples/sec: 2753.55 - lr: 0.000027 - momentum: 0.000000
2023-10-13 21:42:59,675 epoch 6 - iter 297/992 - loss 0.02988685 - time (sec): 18.09 - samples/sec: 2746.20 - lr: 0.000026 - momentum: 0.000000
2023-10-13 21:43:05,663 epoch 6 - iter 396/992 - loss 0.02833903 - time (sec): 24.08 - samples/sec: 2722.44 - lr: 0.000026 - momentum: 0.000000
2023-10-13 21:43:11,508 epoch 6 - iter 495/992 - loss 0.02786454 - time (sec): 29.93 - samples/sec: 2737.83 - lr: 0.000025 - momentum: 0.000000
2023-10-13 21:43:17,288 epoch 6 - iter 594/992 - loss 0.02866839 - time (sec): 35.71 - samples/sec: 2755.15 - lr: 0.000024 - momentum: 0.000000
2023-10-13 21:43:23,408 epoch 6 - iter 693/992 - loss 0.02806587 - time (sec): 41.83 - samples/sec: 2753.01 - lr: 0.000024 - momentum: 0.000000
2023-10-13 21:43:29,273 epoch 6 - iter 792/992 - loss 0.02884132 - time (sec): 47.69 - samples/sec: 2755.88 - lr: 0.000023 - momentum: 0.000000
2023-10-13 21:43:35,118 epoch 6 - iter 891/992 - loss 0.02883868 - time (sec): 53.54 - samples/sec: 2753.63 - lr: 0.000023 - momentum: 0.000000
2023-10-13 21:43:41,138 epoch 6 - iter 990/992 - loss 0.02863755 - time (sec): 59.56 - samples/sec: 2748.21 - lr: 0.000022 - momentum: 0.000000
2023-10-13 21:43:41,250 ----------------------------------------------------------------------------------------------------
2023-10-13 21:43:41,250 EPOCH 6 done: loss 0.0288 - lr: 0.000022
2023-10-13 21:43:44,796 DEV : loss 0.17916934192180634 - f1-score (micro avg) 0.7563
2023-10-13 21:43:44,821 ----------------------------------------------------------------------------------------------------
2023-10-13 21:43:50,755 epoch 7 - iter 99/992 - loss 0.01996214 - time (sec): 5.93 - samples/sec: 2590.21 - lr: 0.000022 - momentum: 0.000000
2023-10-13 21:43:57,458 epoch 7 - iter 198/992 - loss 0.02277291 - time (sec): 12.63 - samples/sec: 2548.91 - lr: 0.000021 - momentum: 0.000000
2023-10-13 21:44:03,311 epoch 7 - iter 297/992 - loss 0.02180867 - time (sec): 18.49 - samples/sec: 2584.45 - lr: 0.000021 - momentum: 0.000000
2023-10-13 21:44:09,225 epoch 7 - iter 396/992 - loss 0.02147002 - time (sec): 24.40 - samples/sec: 2621.14 - lr: 0.000020 - momentum: 0.000000
2023-10-13 21:44:15,345 epoch 7 - iter 495/992 - loss 0.02147281 - time (sec): 30.52 - samples/sec: 2658.70 - lr: 0.000019 - momentum: 0.000000
2023-10-13 21:44:21,676 epoch 7 - iter 594/992 - loss 0.02278340 - time (sec): 36.85 - samples/sec: 2672.02 - lr: 0.000019 - momentum: 0.000000
2023-10-13 21:44:27,314 epoch 7 - iter 693/992 - loss 0.02302071 - time (sec): 42.49 - samples/sec: 2675.36 - lr: 0.000018 - momentum: 0.000000
2023-10-13 21:44:33,289 epoch 7 - iter 792/992 - loss 0.02238297 - time (sec): 48.47 - samples/sec: 2696.36 - lr: 0.000018 - momentum: 0.000000
2023-10-13 21:44:38,932 epoch 7 - iter 891/992 - loss 0.02194504 - time (sec): 54.11 - samples/sec: 2717.35 - lr: 0.000017 - momentum: 0.000000
2023-10-13 21:44:44,806 epoch 7 - iter 990/992 - loss 0.02178032 - time (sec): 59.98 - samples/sec: 2727.36 - lr: 0.000017 - momentum: 0.000000
2023-10-13 21:44:44,942 ----------------------------------------------------------------------------------------------------
2023-10-13 21:44:44,942 EPOCH 7 done: loss 0.0217 - lr: 0.000017
2023-10-13 21:44:48,386 DEV : loss 0.19963695108890533 - f1-score (micro avg) 0.7511
2023-10-13 21:44:48,411 ----------------------------------------------------------------------------------------------------
2023-10-13 21:44:54,512 epoch 8 - iter 99/992 - loss 0.00802566 - time (sec): 6.10 - samples/sec: 2678.43 - lr: 0.000016 - momentum: 0.000000
2023-10-13 21:45:00,449 epoch 8 - iter 198/992 - loss 0.01380324 - time (sec): 12.04 - samples/sec: 2738.01 - lr: 0.000016 - momentum: 0.000000
2023-10-13 21:45:06,122 epoch 8 - iter 297/992 - loss 0.01227669 - time (sec): 17.71 - samples/sec: 2757.67 - lr: 0.000015 - momentum: 0.000000
2023-10-13 21:45:12,518 epoch 8 - iter 396/992 - loss 0.01400552 - time (sec): 24.11 - samples/sec: 2729.93 - lr: 0.000014 - momentum: 0.000000
2023-10-13 21:45:18,333 epoch 8 - iter 495/992 - loss 0.01475903 - time (sec): 29.92 - samples/sec: 2736.32 - lr: 0.000014 - momentum: 0.000000
2023-10-13 21:45:23,984 epoch 8 - iter 594/992 - loss 0.01530820 - time (sec): 35.57 - samples/sec: 2735.01 - lr: 0.000013 - momentum: 0.000000
2023-10-13 21:45:30,069 epoch 8 - iter 693/992 - loss 0.01589983 - time (sec): 41.66 - samples/sec: 2736.91 - lr: 0.000013 - momentum: 0.000000
2023-10-13 21:45:36,359 epoch 8 - iter 792/992 - loss 0.01603725 - time (sec): 47.95 - samples/sec: 2744.42 - lr: 0.000012 - momentum: 0.000000
2023-10-13 21:45:42,042 epoch 8 - iter 891/992 - loss 0.01533557 - time (sec): 53.63 - samples/sec: 2749.62 - lr: 0.000012 - momentum: 0.000000
2023-10-13 21:45:47,922 epoch 8 - iter 990/992 - loss 0.01568211 - time (sec): 59.51 - samples/sec: 2751.39 - lr: 0.000011 - momentum: 0.000000
2023-10-13 21:45:48,031 ----------------------------------------------------------------------------------------------------
2023-10-13 21:45:48,031 EPOCH 8 done: loss 0.0157 - lr: 0.000011
2023-10-13 21:45:51,538 DEV : loss 0.2106373906135559 - f1-score (micro avg) 0.7511
2023-10-13 21:45:51,560 ----------------------------------------------------------------------------------------------------
2023-10-13 21:45:57,529 epoch 9 - iter 99/992 - loss 0.00887012 - time (sec): 5.97 - samples/sec: 2806.18 - lr: 0.000011 - momentum: 0.000000
2023-10-13 21:46:03,489 epoch 9 - iter 198/992 - loss 0.00899000 - time (sec): 11.93 - samples/sec: 2748.69 - lr: 0.000010 - momentum: 0.000000
2023-10-13 21:46:09,616 epoch 9 - iter 297/992 - loss 0.00742884 - time (sec): 18.05 - samples/sec: 2694.16 - lr: 0.000009 - momentum: 0.000000
2023-10-13 21:46:15,833 epoch 9 - iter 396/992 - loss 0.00794530 - time (sec): 24.27 - samples/sec: 2710.07 - lr: 0.000009 - momentum: 0.000000
2023-10-13 21:46:21,629 epoch 9 - iter 495/992 - loss 0.00848285 - time (sec): 30.07 - samples/sec: 2744.72 - lr: 0.000008 - momentum: 0.000000
2023-10-13 21:46:27,363 epoch 9 - iter 594/992 - loss 0.00917051 - time (sec): 35.80 - samples/sec: 2739.94 - lr: 0.000008 - momentum: 0.000000
2023-10-13 21:46:32,985 epoch 9 - iter 693/992 - loss 0.00939071 - time (sec): 41.42 - samples/sec: 2755.95 - lr: 0.000007 - momentum: 0.000000
2023-10-13 21:46:38,976 epoch 9 - iter 792/992 - loss 0.01010125 - time (sec): 47.41 - samples/sec: 2762.85 - lr: 0.000007 - momentum: 0.000000
2023-10-13 21:46:45,090 epoch 9 - iter 891/992 - loss 0.01075447 - time (sec): 53.53 - samples/sec: 2761.88 - lr: 0.000006 - momentum: 0.000000
2023-10-13 21:46:50,883 epoch 9 - iter 990/992 - loss 0.01068750 - time (sec): 59.32 - samples/sec: 2757.49 - lr: 0.000006 - momentum: 0.000000
2023-10-13 21:46:51,018 ----------------------------------------------------------------------------------------------------
2023-10-13 21:46:51,018 EPOCH 9 done: loss 0.0107 - lr: 0.000006
2023-10-13 21:46:54,559 DEV : loss 0.22137963771820068 - f1-score (micro avg) 0.7493
2023-10-13 21:46:54,581 ----------------------------------------------------------------------------------------------------
2023-10-13 21:47:00,669 epoch 10 - iter 99/992 - loss 0.00885115 - time (sec): 6.09 - samples/sec: 2800.45 - lr: 0.000005 - momentum: 0.000000
2023-10-13 21:47:06,567 epoch 10 - iter 198/992 - loss 0.00701261 - time (sec): 11.98 - samples/sec: 2734.37 - lr: 0.000004 - momentum: 0.000000
2023-10-13 21:47:12,490 epoch 10 - iter 297/992 - loss 0.00709994 - time (sec): 17.91 - samples/sec: 2725.21 - lr: 0.000004 - momentum: 0.000000
2023-10-13 21:47:19,212 epoch 10 - iter 396/992 - loss 0.00713180 - time (sec): 24.63 - samples/sec: 2657.26 - lr: 0.000003 - momentum: 0.000000
2023-10-13 21:47:24,800 epoch 10 - iter 495/992 - loss 0.00712523 - time (sec): 30.22 - samples/sec: 2696.43 - lr: 0.000003 - momentum: 0.000000
2023-10-13 21:47:30,657 epoch 10 - iter 594/992 - loss 0.00752130 - time (sec): 36.07 - samples/sec: 2715.70 - lr: 0.000002 - momentum: 0.000000
2023-10-13 21:47:36,606 epoch 10 - iter 693/992 - loss 0.00756143 - time (sec): 42.02 - samples/sec: 2720.45 - lr: 0.000002 - momentum: 0.000000
2023-10-13 21:47:42,502 epoch 10 - iter 792/992 - loss 0.00745144 - time (sec): 47.92 - samples/sec: 2732.37 - lr: 0.000001 - momentum: 0.000000
2023-10-13 21:47:48,499 epoch 10 - iter 891/992 - loss 0.00734368 - time (sec): 53.92 - samples/sec: 2744.99 - lr: 0.000001 - momentum: 0.000000
2023-10-13 21:47:54,258 epoch 10 - iter 990/992 - loss 0.00723729 - time (sec): 59.68 - samples/sec: 2742.25 - lr: 0.000000 - momentum: 0.000000
2023-10-13 21:47:54,381 ----------------------------------------------------------------------------------------------------
2023-10-13 21:47:54,381 EPOCH 10 done: loss 0.0072 - lr: 0.000000
2023-10-13 21:47:57,909 DEV : loss 0.22665657103061676 - f1-score (micro avg) 0.7513
2023-10-13 21:47:58,383 ----------------------------------------------------------------------------------------------------
2023-10-13 21:47:58,384 Loading model from best epoch ...
2023-10-13 21:47:59,826 SequenceTagger predicts: Dictionary with 13 tags: O, S-PER, B-PER, E-PER, I-PER, S-LOC, B-LOC, E-LOC, I-LOC, S-ORG, B-ORG, E-ORG, I-ORG
2023-10-13 21:48:03,320
Results:
- F-score (micro) 0.7736
- F-score (macro) 0.6613
- Accuracy 0.6522
By class:
precision recall f1-score support
LOC 0.7826 0.8794 0.8282 655
PER 0.8556 0.6906 0.7643 223
ORG 0.5968 0.2913 0.3915 127
micro avg 0.7843 0.7632 0.7736 1005
macro avg 0.7450 0.6204 0.6613 1005
weighted avg 0.7753 0.7632 0.7588 1005
2023-10-13 21:48:03,320 ----------------------------------------------------------------------------------------------------
|