File size: 24,213 Bytes
4561ea5 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 |
2023-10-18 15:59:38,444 ----------------------------------------------------------------------------------------------------
2023-10-18 15:59:38,444 Model: "SequenceTagger(
(embeddings): TransformerWordEmbeddings(
(model): BertModel(
(embeddings): BertEmbeddings(
(word_embeddings): Embedding(32001, 128)
(position_embeddings): Embedding(512, 128)
(token_type_embeddings): Embedding(2, 128)
(LayerNorm): LayerNorm((128,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(encoder): BertEncoder(
(layer): ModuleList(
(0-1): 2 x BertLayer(
(attention): BertAttention(
(self): BertSelfAttention(
(query): Linear(in_features=128, out_features=128, bias=True)
(key): Linear(in_features=128, out_features=128, bias=True)
(value): Linear(in_features=128, out_features=128, bias=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(output): BertSelfOutput(
(dense): Linear(in_features=128, out_features=128, bias=True)
(LayerNorm): LayerNorm((128,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
(intermediate): BertIntermediate(
(dense): Linear(in_features=128, out_features=512, bias=True)
(intermediate_act_fn): GELUActivation()
)
(output): BertOutput(
(dense): Linear(in_features=512, out_features=128, bias=True)
(LayerNorm): LayerNorm((128,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
)
(pooler): BertPooler(
(dense): Linear(in_features=128, out_features=128, bias=True)
(activation): Tanh()
)
)
)
(locked_dropout): LockedDropout(p=0.5)
(linear): Linear(in_features=128, out_features=25, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-18 15:59:38,444 ----------------------------------------------------------------------------------------------------
2023-10-18 15:59:38,444 MultiCorpus: 1214 train + 266 dev + 251 test sentences
- NER_HIPE_2022 Corpus: 1214 train + 266 dev + 251 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/ajmc/en/with_doc_seperator
2023-10-18 15:59:38,444 ----------------------------------------------------------------------------------------------------
2023-10-18 15:59:38,444 Train: 1214 sentences
2023-10-18 15:59:38,444 (train_with_dev=False, train_with_test=False)
2023-10-18 15:59:38,444 ----------------------------------------------------------------------------------------------------
2023-10-18 15:59:38,445 Training Params:
2023-10-18 15:59:38,445 - learning_rate: "5e-05"
2023-10-18 15:59:38,445 - mini_batch_size: "4"
2023-10-18 15:59:38,445 - max_epochs: "10"
2023-10-18 15:59:38,445 - shuffle: "True"
2023-10-18 15:59:38,445 ----------------------------------------------------------------------------------------------------
2023-10-18 15:59:38,445 Plugins:
2023-10-18 15:59:38,445 - TensorboardLogger
2023-10-18 15:59:38,445 - LinearScheduler | warmup_fraction: '0.1'
2023-10-18 15:59:38,445 ----------------------------------------------------------------------------------------------------
2023-10-18 15:59:38,445 Final evaluation on model from best epoch (best-model.pt)
2023-10-18 15:59:38,445 - metric: "('micro avg', 'f1-score')"
2023-10-18 15:59:38,445 ----------------------------------------------------------------------------------------------------
2023-10-18 15:59:38,445 Computation:
2023-10-18 15:59:38,445 - compute on device: cuda:0
2023-10-18 15:59:38,445 - embedding storage: none
2023-10-18 15:59:38,445 ----------------------------------------------------------------------------------------------------
2023-10-18 15:59:38,445 Model training base path: "hmbench-ajmc/en-dbmdz/bert-tiny-historic-multilingual-cased-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-1"
2023-10-18 15:59:38,445 ----------------------------------------------------------------------------------------------------
2023-10-18 15:59:38,445 ----------------------------------------------------------------------------------------------------
2023-10-18 15:59:38,445 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-18 15:59:38,893 epoch 1 - iter 30/304 - loss 4.02439859 - time (sec): 0.45 - samples/sec: 6908.01 - lr: 0.000005 - momentum: 0.000000
2023-10-18 15:59:39,332 epoch 1 - iter 60/304 - loss 3.94799824 - time (sec): 0.89 - samples/sec: 6775.15 - lr: 0.000010 - momentum: 0.000000
2023-10-18 15:59:39,798 epoch 1 - iter 90/304 - loss 3.77027124 - time (sec): 1.35 - samples/sec: 6645.20 - lr: 0.000015 - momentum: 0.000000
2023-10-18 15:59:40,240 epoch 1 - iter 120/304 - loss 3.52488959 - time (sec): 1.79 - samples/sec: 6631.52 - lr: 0.000020 - momentum: 0.000000
2023-10-18 15:59:40,680 epoch 1 - iter 150/304 - loss 3.22781102 - time (sec): 2.23 - samples/sec: 6620.98 - lr: 0.000025 - momentum: 0.000000
2023-10-18 15:59:41,127 epoch 1 - iter 180/304 - loss 2.92761720 - time (sec): 2.68 - samples/sec: 6552.60 - lr: 0.000029 - momentum: 0.000000
2023-10-18 15:59:41,584 epoch 1 - iter 210/304 - loss 2.61647495 - time (sec): 3.14 - samples/sec: 6656.08 - lr: 0.000034 - momentum: 0.000000
2023-10-18 15:59:42,044 epoch 1 - iter 240/304 - loss 2.38351652 - time (sec): 3.60 - samples/sec: 6683.60 - lr: 0.000039 - momentum: 0.000000
2023-10-18 15:59:42,493 epoch 1 - iter 270/304 - loss 2.18072309 - time (sec): 4.05 - samples/sec: 6762.41 - lr: 0.000044 - momentum: 0.000000
2023-10-18 15:59:42,947 epoch 1 - iter 300/304 - loss 2.03065121 - time (sec): 4.50 - samples/sec: 6818.46 - lr: 0.000049 - momentum: 0.000000
2023-10-18 15:59:43,004 ----------------------------------------------------------------------------------------------------
2023-10-18 15:59:43,004 EPOCH 1 done: loss 2.0164 - lr: 0.000049
2023-10-18 15:59:43,462 DEV : loss 0.7537270188331604 - f1-score (micro avg) 0.0
2023-10-18 15:59:43,467 ----------------------------------------------------------------------------------------------------
2023-10-18 15:59:43,917 epoch 2 - iter 30/304 - loss 0.67693457 - time (sec): 0.45 - samples/sec: 7263.16 - lr: 0.000049 - momentum: 0.000000
2023-10-18 15:59:44,381 epoch 2 - iter 60/304 - loss 0.68249441 - time (sec): 0.91 - samples/sec: 6828.38 - lr: 0.000049 - momentum: 0.000000
2023-10-18 15:59:44,855 epoch 2 - iter 90/304 - loss 0.68860219 - time (sec): 1.39 - samples/sec: 6743.02 - lr: 0.000048 - momentum: 0.000000
2023-10-18 15:59:45,375 epoch 2 - iter 120/304 - loss 0.68478765 - time (sec): 1.91 - samples/sec: 6636.08 - lr: 0.000048 - momentum: 0.000000
2023-10-18 15:59:45,842 epoch 2 - iter 150/304 - loss 0.64870052 - time (sec): 2.37 - samples/sec: 6488.58 - lr: 0.000047 - momentum: 0.000000
2023-10-18 15:59:46,320 epoch 2 - iter 180/304 - loss 0.64138183 - time (sec): 2.85 - samples/sec: 6412.79 - lr: 0.000047 - momentum: 0.000000
2023-10-18 15:59:46,774 epoch 2 - iter 210/304 - loss 0.60885352 - time (sec): 3.31 - samples/sec: 6538.11 - lr: 0.000046 - momentum: 0.000000
2023-10-18 15:59:47,232 epoch 2 - iter 240/304 - loss 0.59175345 - time (sec): 3.76 - samples/sec: 6515.91 - lr: 0.000046 - momentum: 0.000000
2023-10-18 15:59:47,681 epoch 2 - iter 270/304 - loss 0.58324719 - time (sec): 4.21 - samples/sec: 6578.92 - lr: 0.000045 - momentum: 0.000000
2023-10-18 15:59:48,135 epoch 2 - iter 300/304 - loss 0.58565562 - time (sec): 4.67 - samples/sec: 6569.76 - lr: 0.000045 - momentum: 0.000000
2023-10-18 15:59:48,196 ----------------------------------------------------------------------------------------------------
2023-10-18 15:59:48,196 EPOCH 2 done: loss 0.5838 - lr: 0.000045
2023-10-18 15:59:48,694 DEV : loss 0.39393150806427 - f1-score (micro avg) 0.2408
2023-10-18 15:59:48,699 saving best model
2023-10-18 15:59:48,728 ----------------------------------------------------------------------------------------------------
2023-10-18 15:59:49,181 epoch 3 - iter 30/304 - loss 0.42728775 - time (sec): 0.45 - samples/sec: 6484.68 - lr: 0.000044 - momentum: 0.000000
2023-10-18 15:59:49,667 epoch 3 - iter 60/304 - loss 0.43840468 - time (sec): 0.94 - samples/sec: 6754.01 - lr: 0.000043 - momentum: 0.000000
2023-10-18 15:59:50,138 epoch 3 - iter 90/304 - loss 0.43340952 - time (sec): 1.41 - samples/sec: 6653.01 - lr: 0.000043 - momentum: 0.000000
2023-10-18 15:59:50,622 epoch 3 - iter 120/304 - loss 0.42193644 - time (sec): 1.89 - samples/sec: 6790.83 - lr: 0.000042 - momentum: 0.000000
2023-10-18 15:59:51,123 epoch 3 - iter 150/304 - loss 0.41461292 - time (sec): 2.39 - samples/sec: 6665.34 - lr: 0.000042 - momentum: 0.000000
2023-10-18 15:59:51,586 epoch 3 - iter 180/304 - loss 0.42630692 - time (sec): 2.86 - samples/sec: 6629.85 - lr: 0.000041 - momentum: 0.000000
2023-10-18 15:59:52,043 epoch 3 - iter 210/304 - loss 0.42817892 - time (sec): 3.31 - samples/sec: 6581.34 - lr: 0.000041 - momentum: 0.000000
2023-10-18 15:59:52,481 epoch 3 - iter 240/304 - loss 0.41490933 - time (sec): 3.75 - samples/sec: 6564.30 - lr: 0.000040 - momentum: 0.000000
2023-10-18 15:59:52,929 epoch 3 - iter 270/304 - loss 0.41477427 - time (sec): 4.20 - samples/sec: 6579.62 - lr: 0.000040 - momentum: 0.000000
2023-10-18 15:59:53,383 epoch 3 - iter 300/304 - loss 0.40965081 - time (sec): 4.65 - samples/sec: 6576.98 - lr: 0.000039 - momentum: 0.000000
2023-10-18 15:59:53,440 ----------------------------------------------------------------------------------------------------
2023-10-18 15:59:53,440 EPOCH 3 done: loss 0.4082 - lr: 0.000039
2023-10-18 15:59:53,951 DEV : loss 0.32593443989753723 - f1-score (micro avg) 0.4345
2023-10-18 15:59:53,957 saving best model
2023-10-18 15:59:53,996 ----------------------------------------------------------------------------------------------------
2023-10-18 15:59:54,450 epoch 4 - iter 30/304 - loss 0.34255467 - time (sec): 0.45 - samples/sec: 6790.88 - lr: 0.000038 - momentum: 0.000000
2023-10-18 15:59:54,894 epoch 4 - iter 60/304 - loss 0.39274816 - time (sec): 0.90 - samples/sec: 6877.07 - lr: 0.000038 - momentum: 0.000000
2023-10-18 15:59:55,348 epoch 4 - iter 90/304 - loss 0.37778060 - time (sec): 1.35 - samples/sec: 6903.74 - lr: 0.000037 - momentum: 0.000000
2023-10-18 15:59:55,815 epoch 4 - iter 120/304 - loss 0.37593215 - time (sec): 1.82 - samples/sec: 6776.34 - lr: 0.000037 - momentum: 0.000000
2023-10-18 15:59:56,274 epoch 4 - iter 150/304 - loss 0.36040013 - time (sec): 2.28 - samples/sec: 6707.14 - lr: 0.000036 - momentum: 0.000000
2023-10-18 15:59:56,717 epoch 4 - iter 180/304 - loss 0.35667086 - time (sec): 2.72 - samples/sec: 6684.37 - lr: 0.000036 - momentum: 0.000000
2023-10-18 15:59:57,173 epoch 4 - iter 210/304 - loss 0.35194304 - time (sec): 3.18 - samples/sec: 6724.60 - lr: 0.000035 - momentum: 0.000000
2023-10-18 15:59:57,624 epoch 4 - iter 240/304 - loss 0.34062892 - time (sec): 3.63 - samples/sec: 6771.82 - lr: 0.000035 - momentum: 0.000000
2023-10-18 15:59:58,077 epoch 4 - iter 270/304 - loss 0.33268323 - time (sec): 4.08 - samples/sec: 6742.26 - lr: 0.000034 - momentum: 0.000000
2023-10-18 15:59:58,528 epoch 4 - iter 300/304 - loss 0.33151212 - time (sec): 4.53 - samples/sec: 6759.21 - lr: 0.000033 - momentum: 0.000000
2023-10-18 15:59:58,582 ----------------------------------------------------------------------------------------------------
2023-10-18 15:59:58,582 EPOCH 4 done: loss 0.3309 - lr: 0.000033
2023-10-18 15:59:59,095 DEV : loss 0.29030725359916687 - f1-score (micro avg) 0.4636
2023-10-18 15:59:59,101 saving best model
2023-10-18 15:59:59,136 ----------------------------------------------------------------------------------------------------
2023-10-18 15:59:59,589 epoch 5 - iter 30/304 - loss 0.30069451 - time (sec): 0.45 - samples/sec: 6982.39 - lr: 0.000033 - momentum: 0.000000
2023-10-18 16:00:00,045 epoch 5 - iter 60/304 - loss 0.32261811 - time (sec): 0.91 - samples/sec: 7065.54 - lr: 0.000032 - momentum: 0.000000
2023-10-18 16:00:00,500 epoch 5 - iter 90/304 - loss 0.29691347 - time (sec): 1.36 - samples/sec: 7070.53 - lr: 0.000032 - momentum: 0.000000
2023-10-18 16:00:00,962 epoch 5 - iter 120/304 - loss 0.28228503 - time (sec): 1.83 - samples/sec: 6873.42 - lr: 0.000031 - momentum: 0.000000
2023-10-18 16:00:01,424 epoch 5 - iter 150/304 - loss 0.27506302 - time (sec): 2.29 - samples/sec: 6831.76 - lr: 0.000031 - momentum: 0.000000
2023-10-18 16:00:01,878 epoch 5 - iter 180/304 - loss 0.27966818 - time (sec): 2.74 - samples/sec: 6748.00 - lr: 0.000030 - momentum: 0.000000
2023-10-18 16:00:02,329 epoch 5 - iter 210/304 - loss 0.28859803 - time (sec): 3.19 - samples/sec: 6813.04 - lr: 0.000030 - momentum: 0.000000
2023-10-18 16:00:02,786 epoch 5 - iter 240/304 - loss 0.29244738 - time (sec): 3.65 - samples/sec: 6743.30 - lr: 0.000029 - momentum: 0.000000
2023-10-18 16:00:03,233 epoch 5 - iter 270/304 - loss 0.28822906 - time (sec): 4.10 - samples/sec: 6762.72 - lr: 0.000028 - momentum: 0.000000
2023-10-18 16:00:03,690 epoch 5 - iter 300/304 - loss 0.29090928 - time (sec): 4.55 - samples/sec: 6751.37 - lr: 0.000028 - momentum: 0.000000
2023-10-18 16:00:03,744 ----------------------------------------------------------------------------------------------------
2023-10-18 16:00:03,744 EPOCH 5 done: loss 0.2904 - lr: 0.000028
2023-10-18 16:00:04,278 DEV : loss 0.26164156198501587 - f1-score (micro avg) 0.5045
2023-10-18 16:00:04,283 saving best model
2023-10-18 16:00:04,319 ----------------------------------------------------------------------------------------------------
2023-10-18 16:00:04,772 epoch 6 - iter 30/304 - loss 0.28626393 - time (sec): 0.45 - samples/sec: 6235.37 - lr: 0.000027 - momentum: 0.000000
2023-10-18 16:00:05,211 epoch 6 - iter 60/304 - loss 0.27365340 - time (sec): 0.89 - samples/sec: 6403.24 - lr: 0.000027 - momentum: 0.000000
2023-10-18 16:00:05,666 epoch 6 - iter 90/304 - loss 0.26753734 - time (sec): 1.35 - samples/sec: 6724.20 - lr: 0.000026 - momentum: 0.000000
2023-10-18 16:00:06,122 epoch 6 - iter 120/304 - loss 0.28549914 - time (sec): 1.80 - samples/sec: 6700.80 - lr: 0.000026 - momentum: 0.000000
2023-10-18 16:00:06,578 epoch 6 - iter 150/304 - loss 0.27259411 - time (sec): 2.26 - samples/sec: 6743.60 - lr: 0.000025 - momentum: 0.000000
2023-10-18 16:00:07,035 epoch 6 - iter 180/304 - loss 0.27162046 - time (sec): 2.72 - samples/sec: 6713.65 - lr: 0.000025 - momentum: 0.000000
2023-10-18 16:00:07,499 epoch 6 - iter 210/304 - loss 0.26916890 - time (sec): 3.18 - samples/sec: 6705.00 - lr: 0.000024 - momentum: 0.000000
2023-10-18 16:00:07,951 epoch 6 - iter 240/304 - loss 0.26300420 - time (sec): 3.63 - samples/sec: 6679.51 - lr: 0.000023 - momentum: 0.000000
2023-10-18 16:00:08,408 epoch 6 - iter 270/304 - loss 0.26565702 - time (sec): 4.09 - samples/sec: 6643.81 - lr: 0.000023 - momentum: 0.000000
2023-10-18 16:00:08,886 epoch 6 - iter 300/304 - loss 0.26199835 - time (sec): 4.57 - samples/sec: 6699.47 - lr: 0.000022 - momentum: 0.000000
2023-10-18 16:00:08,945 ----------------------------------------------------------------------------------------------------
2023-10-18 16:00:08,945 EPOCH 6 done: loss 0.2617 - lr: 0.000022
2023-10-18 16:00:09,454 DEV : loss 0.25211378931999207 - f1-score (micro avg) 0.5287
2023-10-18 16:00:09,459 saving best model
2023-10-18 16:00:09,494 ----------------------------------------------------------------------------------------------------
2023-10-18 16:00:09,951 epoch 7 - iter 30/304 - loss 0.24943728 - time (sec): 0.46 - samples/sec: 6462.10 - lr: 0.000022 - momentum: 0.000000
2023-10-18 16:00:10,408 epoch 7 - iter 60/304 - loss 0.25514322 - time (sec): 0.91 - samples/sec: 6606.04 - lr: 0.000021 - momentum: 0.000000
2023-10-18 16:00:10,865 epoch 7 - iter 90/304 - loss 0.25104880 - time (sec): 1.37 - samples/sec: 6578.90 - lr: 0.000021 - momentum: 0.000000
2023-10-18 16:00:11,310 epoch 7 - iter 120/304 - loss 0.25691959 - time (sec): 1.81 - samples/sec: 6662.24 - lr: 0.000020 - momentum: 0.000000
2023-10-18 16:00:11,761 epoch 7 - iter 150/304 - loss 0.25712263 - time (sec): 2.27 - samples/sec: 6781.94 - lr: 0.000020 - momentum: 0.000000
2023-10-18 16:00:12,209 epoch 7 - iter 180/304 - loss 0.25996239 - time (sec): 2.71 - samples/sec: 6747.99 - lr: 0.000019 - momentum: 0.000000
2023-10-18 16:00:12,659 epoch 7 - iter 210/304 - loss 0.25169702 - time (sec): 3.16 - samples/sec: 6721.71 - lr: 0.000018 - momentum: 0.000000
2023-10-18 16:00:13,115 epoch 7 - iter 240/304 - loss 0.25021440 - time (sec): 3.62 - samples/sec: 6715.13 - lr: 0.000018 - momentum: 0.000000
2023-10-18 16:00:13,581 epoch 7 - iter 270/304 - loss 0.24382932 - time (sec): 4.09 - samples/sec: 6722.19 - lr: 0.000017 - momentum: 0.000000
2023-10-18 16:00:14,043 epoch 7 - iter 300/304 - loss 0.24830003 - time (sec): 4.55 - samples/sec: 6742.78 - lr: 0.000017 - momentum: 0.000000
2023-10-18 16:00:14,107 ----------------------------------------------------------------------------------------------------
2023-10-18 16:00:14,107 EPOCH 7 done: loss 0.2474 - lr: 0.000017
2023-10-18 16:00:14,609 DEV : loss 0.2437625378370285 - f1-score (micro avg) 0.5487
2023-10-18 16:00:14,614 saving best model
2023-10-18 16:00:14,649 ----------------------------------------------------------------------------------------------------
2023-10-18 16:00:15,121 epoch 8 - iter 30/304 - loss 0.22448797 - time (sec): 0.47 - samples/sec: 5622.44 - lr: 0.000016 - momentum: 0.000000
2023-10-18 16:00:15,595 epoch 8 - iter 60/304 - loss 0.23152462 - time (sec): 0.95 - samples/sec: 6010.92 - lr: 0.000016 - momentum: 0.000000
2023-10-18 16:00:16,068 epoch 8 - iter 90/304 - loss 0.21243463 - time (sec): 1.42 - samples/sec: 6088.48 - lr: 0.000015 - momentum: 0.000000
2023-10-18 16:00:16,566 epoch 8 - iter 120/304 - loss 0.23068737 - time (sec): 1.92 - samples/sec: 6156.13 - lr: 0.000015 - momentum: 0.000000
2023-10-18 16:00:17,042 epoch 8 - iter 150/304 - loss 0.23114576 - time (sec): 2.39 - samples/sec: 6336.38 - lr: 0.000014 - momentum: 0.000000
2023-10-18 16:00:17,499 epoch 8 - iter 180/304 - loss 0.23233799 - time (sec): 2.85 - samples/sec: 6309.95 - lr: 0.000013 - momentum: 0.000000
2023-10-18 16:00:17,942 epoch 8 - iter 210/304 - loss 0.23090640 - time (sec): 3.29 - samples/sec: 6464.13 - lr: 0.000013 - momentum: 0.000000
2023-10-18 16:00:18,388 epoch 8 - iter 240/304 - loss 0.23271789 - time (sec): 3.74 - samples/sec: 6553.42 - lr: 0.000012 - momentum: 0.000000
2023-10-18 16:00:18,864 epoch 8 - iter 270/304 - loss 0.23726179 - time (sec): 4.21 - samples/sec: 6553.74 - lr: 0.000012 - momentum: 0.000000
2023-10-18 16:00:19,319 epoch 8 - iter 300/304 - loss 0.23601616 - time (sec): 4.67 - samples/sec: 6567.33 - lr: 0.000011 - momentum: 0.000000
2023-10-18 16:00:19,375 ----------------------------------------------------------------------------------------------------
2023-10-18 16:00:19,376 EPOCH 8 done: loss 0.2358 - lr: 0.000011
2023-10-18 16:00:19,894 DEV : loss 0.23939570784568787 - f1-score (micro avg) 0.5522
2023-10-18 16:00:19,899 saving best model
2023-10-18 16:00:19,933 ----------------------------------------------------------------------------------------------------
2023-10-18 16:00:20,398 epoch 9 - iter 30/304 - loss 0.21473866 - time (sec): 0.46 - samples/sec: 6421.88 - lr: 0.000011 - momentum: 0.000000
2023-10-18 16:00:20,878 epoch 9 - iter 60/304 - loss 0.24345455 - time (sec): 0.94 - samples/sec: 6481.83 - lr: 0.000010 - momentum: 0.000000
2023-10-18 16:00:21,342 epoch 9 - iter 90/304 - loss 0.25235638 - time (sec): 1.41 - samples/sec: 6575.74 - lr: 0.000010 - momentum: 0.000000
2023-10-18 16:00:21,791 epoch 9 - iter 120/304 - loss 0.23860452 - time (sec): 1.86 - samples/sec: 6701.57 - lr: 0.000009 - momentum: 0.000000
2023-10-18 16:00:22,252 epoch 9 - iter 150/304 - loss 0.23208699 - time (sec): 2.32 - samples/sec: 6708.59 - lr: 0.000008 - momentum: 0.000000
2023-10-18 16:00:22,695 epoch 9 - iter 180/304 - loss 0.23166946 - time (sec): 2.76 - samples/sec: 6639.55 - lr: 0.000008 - momentum: 0.000000
2023-10-18 16:00:23,155 epoch 9 - iter 210/304 - loss 0.23227634 - time (sec): 3.22 - samples/sec: 6721.93 - lr: 0.000007 - momentum: 0.000000
2023-10-18 16:00:23,602 epoch 9 - iter 240/304 - loss 0.23524665 - time (sec): 3.67 - samples/sec: 6681.01 - lr: 0.000007 - momentum: 0.000000
2023-10-18 16:00:24,048 epoch 9 - iter 270/304 - loss 0.23400306 - time (sec): 4.11 - samples/sec: 6646.96 - lr: 0.000006 - momentum: 0.000000
2023-10-18 16:00:24,505 epoch 9 - iter 300/304 - loss 0.23000825 - time (sec): 4.57 - samples/sec: 6696.02 - lr: 0.000006 - momentum: 0.000000
2023-10-18 16:00:24,561 ----------------------------------------------------------------------------------------------------
2023-10-18 16:00:24,561 EPOCH 9 done: loss 0.2303 - lr: 0.000006
2023-10-18 16:00:25,084 DEV : loss 0.2380855232477188 - f1-score (micro avg) 0.5534
2023-10-18 16:00:25,090 saving best model
2023-10-18 16:00:25,123 ----------------------------------------------------------------------------------------------------
2023-10-18 16:00:25,583 epoch 10 - iter 30/304 - loss 0.19168195 - time (sec): 0.46 - samples/sec: 6711.13 - lr: 0.000005 - momentum: 0.000000
2023-10-18 16:00:26,055 epoch 10 - iter 60/304 - loss 0.17965363 - time (sec): 0.93 - samples/sec: 6704.56 - lr: 0.000005 - momentum: 0.000000
2023-10-18 16:00:26,495 epoch 10 - iter 90/304 - loss 0.20505631 - time (sec): 1.37 - samples/sec: 6766.85 - lr: 0.000004 - momentum: 0.000000
2023-10-18 16:00:26,934 epoch 10 - iter 120/304 - loss 0.20380845 - time (sec): 1.81 - samples/sec: 6660.94 - lr: 0.000003 - momentum: 0.000000
2023-10-18 16:00:27,377 epoch 10 - iter 150/304 - loss 0.20047948 - time (sec): 2.25 - samples/sec: 6623.80 - lr: 0.000003 - momentum: 0.000000
2023-10-18 16:00:27,828 epoch 10 - iter 180/304 - loss 0.21028851 - time (sec): 2.70 - samples/sec: 6690.30 - lr: 0.000002 - momentum: 0.000000
2023-10-18 16:00:28,295 epoch 10 - iter 210/304 - loss 0.21399489 - time (sec): 3.17 - samples/sec: 6675.00 - lr: 0.000002 - momentum: 0.000000
2023-10-18 16:00:28,759 epoch 10 - iter 240/304 - loss 0.22201720 - time (sec): 3.64 - samples/sec: 6667.29 - lr: 0.000001 - momentum: 0.000000
2023-10-18 16:00:29,232 epoch 10 - iter 270/304 - loss 0.22425873 - time (sec): 4.11 - samples/sec: 6634.53 - lr: 0.000001 - momentum: 0.000000
2023-10-18 16:00:29,693 epoch 10 - iter 300/304 - loss 0.22507873 - time (sec): 4.57 - samples/sec: 6694.32 - lr: 0.000000 - momentum: 0.000000
2023-10-18 16:00:29,753 ----------------------------------------------------------------------------------------------------
2023-10-18 16:00:29,753 EPOCH 10 done: loss 0.2235 - lr: 0.000000
2023-10-18 16:00:30,275 DEV : loss 0.23677615821361542 - f1-score (micro avg) 0.5579
2023-10-18 16:00:30,281 saving best model
2023-10-18 16:00:30,340 ----------------------------------------------------------------------------------------------------
2023-10-18 16:00:30,340 Loading model from best epoch ...
2023-10-18 16:00:30,417 SequenceTagger predicts: Dictionary with 25 tags: O, S-scope, B-scope, E-scope, I-scope, S-pers, B-pers, E-pers, I-pers, S-work, B-work, E-work, I-work, S-loc, B-loc, E-loc, I-loc, S-date, B-date, E-date, I-date, S-object, B-object, E-object, I-object
2023-10-18 16:00:30,907
Results:
- F-score (micro) 0.5992
- F-score (macro) 0.3691
- Accuracy 0.4524
By class:
precision recall f1-score support
scope 0.5082 0.6159 0.5569 151
work 0.5175 0.7789 0.6218 95
pers 0.7011 0.6354 0.6667 96
loc 0.0000 0.0000 0.0000 3
date 0.0000 0.0000 0.0000 3
micro avg 0.5521 0.6552 0.5992 348
macro avg 0.3454 0.4061 0.3691 348
weighted avg 0.5552 0.6552 0.5953 348
2023-10-18 16:00:30,907 ----------------------------------------------------------------------------------------------------
|