File size: 24,099 Bytes
593934b |
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 |
2023-10-18 20:56:03,815 ----------------------------------------------------------------------------------------------------
2023-10-18 20:56:03,816 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=13, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-18 20:56:03,816 ----------------------------------------------------------------------------------------------------
2023-10-18 20:56:03,816 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
2023-10-18 20:56:03,816 ----------------------------------------------------------------------------------------------------
2023-10-18 20:56:03,816 Train: 7936 sentences
2023-10-18 20:56:03,816 (train_with_dev=False, train_with_test=False)
2023-10-18 20:56:03,816 ----------------------------------------------------------------------------------------------------
2023-10-18 20:56:03,816 Training Params:
2023-10-18 20:56:03,816 - learning_rate: "5e-05"
2023-10-18 20:56:03,816 - mini_batch_size: "8"
2023-10-18 20:56:03,816 - max_epochs: "10"
2023-10-18 20:56:03,816 - shuffle: "True"
2023-10-18 20:56:03,816 ----------------------------------------------------------------------------------------------------
2023-10-18 20:56:03,816 Plugins:
2023-10-18 20:56:03,816 - TensorboardLogger
2023-10-18 20:56:03,816 - LinearScheduler | warmup_fraction: '0.1'
2023-10-18 20:56:03,816 ----------------------------------------------------------------------------------------------------
2023-10-18 20:56:03,816 Final evaluation on model from best epoch (best-model.pt)
2023-10-18 20:56:03,816 - metric: "('micro avg', 'f1-score')"
2023-10-18 20:56:03,816 ----------------------------------------------------------------------------------------------------
2023-10-18 20:56:03,816 Computation:
2023-10-18 20:56:03,816 - compute on device: cuda:0
2023-10-18 20:56:03,816 - embedding storage: none
2023-10-18 20:56:03,816 ----------------------------------------------------------------------------------------------------
2023-10-18 20:56:03,816 Model training base path: "hmbench-icdar/fr-dbmdz/bert-tiny-historic-multilingual-cased-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-2"
2023-10-18 20:56:03,816 ----------------------------------------------------------------------------------------------------
2023-10-18 20:56:03,816 ----------------------------------------------------------------------------------------------------
2023-10-18 20:56:03,817 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-18 20:56:06,181 epoch 1 - iter 99/992 - loss 2.37417653 - time (sec): 2.36 - samples/sec: 6970.36 - lr: 0.000005 - momentum: 0.000000
2023-10-18 20:56:08,505 epoch 1 - iter 198/992 - loss 2.04547181 - time (sec): 4.69 - samples/sec: 6978.26 - lr: 0.000010 - momentum: 0.000000
2023-10-18 20:56:10,702 epoch 1 - iter 297/992 - loss 1.63644393 - time (sec): 6.88 - samples/sec: 7234.43 - lr: 0.000015 - momentum: 0.000000
2023-10-18 20:56:12,678 epoch 1 - iter 396/992 - loss 1.35011565 - time (sec): 8.86 - samples/sec: 7467.10 - lr: 0.000020 - momentum: 0.000000
2023-10-18 20:56:14,632 epoch 1 - iter 495/992 - loss 1.18194660 - time (sec): 10.82 - samples/sec: 7693.85 - lr: 0.000025 - momentum: 0.000000
2023-10-18 20:56:16,785 epoch 1 - iter 594/992 - loss 1.05850892 - time (sec): 12.97 - samples/sec: 7664.55 - lr: 0.000030 - momentum: 0.000000
2023-10-18 20:56:18,922 epoch 1 - iter 693/992 - loss 0.96209046 - time (sec): 15.11 - samples/sec: 7664.05 - lr: 0.000035 - momentum: 0.000000
2023-10-18 20:56:21,060 epoch 1 - iter 792/992 - loss 0.88404985 - time (sec): 17.24 - samples/sec: 7634.19 - lr: 0.000040 - momentum: 0.000000
2023-10-18 20:56:23,256 epoch 1 - iter 891/992 - loss 0.82833250 - time (sec): 19.44 - samples/sec: 7576.98 - lr: 0.000045 - momentum: 0.000000
2023-10-18 20:56:25,516 epoch 1 - iter 990/992 - loss 0.77917868 - time (sec): 21.70 - samples/sec: 7543.24 - lr: 0.000050 - momentum: 0.000000
2023-10-18 20:56:25,565 ----------------------------------------------------------------------------------------------------
2023-10-18 20:56:25,565 EPOCH 1 done: loss 0.7778 - lr: 0.000050
2023-10-18 20:56:27,108 DEV : loss 0.2251289188861847 - f1-score (micro avg) 0.2771
2023-10-18 20:56:27,126 saving best model
2023-10-18 20:56:27,158 ----------------------------------------------------------------------------------------------------
2023-10-18 20:56:29,368 epoch 2 - iter 99/992 - loss 0.32167945 - time (sec): 2.21 - samples/sec: 7301.80 - lr: 0.000049 - momentum: 0.000000
2023-10-18 20:56:31,577 epoch 2 - iter 198/992 - loss 0.29255043 - time (sec): 4.42 - samples/sec: 7535.62 - lr: 0.000049 - momentum: 0.000000
2023-10-18 20:56:33,795 epoch 2 - iter 297/992 - loss 0.29067708 - time (sec): 6.64 - samples/sec: 7386.21 - lr: 0.000048 - momentum: 0.000000
2023-10-18 20:56:36,015 epoch 2 - iter 396/992 - loss 0.29025565 - time (sec): 8.86 - samples/sec: 7363.21 - lr: 0.000048 - momentum: 0.000000
2023-10-18 20:56:38,278 epoch 2 - iter 495/992 - loss 0.28143687 - time (sec): 11.12 - samples/sec: 7310.90 - lr: 0.000047 - momentum: 0.000000
2023-10-18 20:56:40,527 epoch 2 - iter 594/992 - loss 0.27945389 - time (sec): 13.37 - samples/sec: 7270.36 - lr: 0.000047 - momentum: 0.000000
2023-10-18 20:56:42,728 epoch 2 - iter 693/992 - loss 0.27975877 - time (sec): 15.57 - samples/sec: 7261.00 - lr: 0.000046 - momentum: 0.000000
2023-10-18 20:56:45,070 epoch 2 - iter 792/992 - loss 0.27735348 - time (sec): 17.91 - samples/sec: 7256.35 - lr: 0.000046 - momentum: 0.000000
2023-10-18 20:56:47,388 epoch 2 - iter 891/992 - loss 0.27681776 - time (sec): 20.23 - samples/sec: 7272.50 - lr: 0.000045 - momentum: 0.000000
2023-10-18 20:56:49,647 epoch 2 - iter 990/992 - loss 0.27381910 - time (sec): 22.49 - samples/sec: 7277.31 - lr: 0.000044 - momentum: 0.000000
2023-10-18 20:56:49,690 ----------------------------------------------------------------------------------------------------
2023-10-18 20:56:49,691 EPOCH 2 done: loss 0.2739 - lr: 0.000044
2023-10-18 20:56:51,906 DEV : loss 0.18277420103549957 - f1-score (micro avg) 0.3656
2023-10-18 20:56:51,924 saving best model
2023-10-18 20:56:51,958 ----------------------------------------------------------------------------------------------------
2023-10-18 20:56:54,200 epoch 3 - iter 99/992 - loss 0.22769296 - time (sec): 2.24 - samples/sec: 7466.59 - lr: 0.000044 - momentum: 0.000000
2023-10-18 20:56:56,425 epoch 3 - iter 198/992 - loss 0.23820508 - time (sec): 4.47 - samples/sec: 7317.55 - lr: 0.000043 - momentum: 0.000000
2023-10-18 20:56:58,657 epoch 3 - iter 297/992 - loss 0.24041108 - time (sec): 6.70 - samples/sec: 7300.94 - lr: 0.000043 - momentum: 0.000000
2023-10-18 20:57:00,926 epoch 3 - iter 396/992 - loss 0.23852730 - time (sec): 8.97 - samples/sec: 7228.49 - lr: 0.000042 - momentum: 0.000000
2023-10-18 20:57:03,244 epoch 3 - iter 495/992 - loss 0.23174392 - time (sec): 11.29 - samples/sec: 7235.37 - lr: 0.000042 - momentum: 0.000000
2023-10-18 20:57:05,436 epoch 3 - iter 594/992 - loss 0.23397501 - time (sec): 13.48 - samples/sec: 7202.93 - lr: 0.000041 - momentum: 0.000000
2023-10-18 20:57:07,666 epoch 3 - iter 693/992 - loss 0.23953902 - time (sec): 15.71 - samples/sec: 7229.79 - lr: 0.000041 - momentum: 0.000000
2023-10-18 20:57:09,905 epoch 3 - iter 792/992 - loss 0.23601306 - time (sec): 17.95 - samples/sec: 7263.88 - lr: 0.000040 - momentum: 0.000000
2023-10-18 20:57:12,148 epoch 3 - iter 891/992 - loss 0.23207443 - time (sec): 20.19 - samples/sec: 7280.65 - lr: 0.000039 - momentum: 0.000000
2023-10-18 20:57:14,478 epoch 3 - iter 990/992 - loss 0.23020989 - time (sec): 22.52 - samples/sec: 7262.19 - lr: 0.000039 - momentum: 0.000000
2023-10-18 20:57:14,524 ----------------------------------------------------------------------------------------------------
2023-10-18 20:57:14,525 EPOCH 3 done: loss 0.2301 - lr: 0.000039
2023-10-18 20:57:16,353 DEV : loss 0.1740075945854187 - f1-score (micro avg) 0.3899
2023-10-18 20:57:16,372 saving best model
2023-10-18 20:57:16,407 ----------------------------------------------------------------------------------------------------
2023-10-18 20:57:18,696 epoch 4 - iter 99/992 - loss 0.21972604 - time (sec): 2.29 - samples/sec: 6880.22 - lr: 0.000038 - momentum: 0.000000
2023-10-18 20:57:20,907 epoch 4 - iter 198/992 - loss 0.21771922 - time (sec): 4.50 - samples/sec: 7255.82 - lr: 0.000038 - momentum: 0.000000
2023-10-18 20:57:23,071 epoch 4 - iter 297/992 - loss 0.21406813 - time (sec): 6.66 - samples/sec: 7184.34 - lr: 0.000037 - momentum: 0.000000
2023-10-18 20:57:25,280 epoch 4 - iter 396/992 - loss 0.21159074 - time (sec): 8.87 - samples/sec: 7161.50 - lr: 0.000037 - momentum: 0.000000
2023-10-18 20:57:27,521 epoch 4 - iter 495/992 - loss 0.21280319 - time (sec): 11.11 - samples/sec: 7252.30 - lr: 0.000036 - momentum: 0.000000
2023-10-18 20:57:29,763 epoch 4 - iter 594/992 - loss 0.21029475 - time (sec): 13.36 - samples/sec: 7261.19 - lr: 0.000036 - momentum: 0.000000
2023-10-18 20:57:31,997 epoch 4 - iter 693/992 - loss 0.20859482 - time (sec): 15.59 - samples/sec: 7267.30 - lr: 0.000035 - momentum: 0.000000
2023-10-18 20:57:34,226 epoch 4 - iter 792/992 - loss 0.20972295 - time (sec): 17.82 - samples/sec: 7290.54 - lr: 0.000034 - momentum: 0.000000
2023-10-18 20:57:36,517 epoch 4 - iter 891/992 - loss 0.20628428 - time (sec): 20.11 - samples/sec: 7298.49 - lr: 0.000034 - momentum: 0.000000
2023-10-18 20:57:38,726 epoch 4 - iter 990/992 - loss 0.20626360 - time (sec): 22.32 - samples/sec: 7330.91 - lr: 0.000033 - momentum: 0.000000
2023-10-18 20:57:38,772 ----------------------------------------------------------------------------------------------------
2023-10-18 20:57:38,772 EPOCH 4 done: loss 0.2061 - lr: 0.000033
2023-10-18 20:57:40,612 DEV : loss 0.16479910910129547 - f1-score (micro avg) 0.4101
2023-10-18 20:57:40,630 saving best model
2023-10-18 20:57:40,666 ----------------------------------------------------------------------------------------------------
2023-10-18 20:57:42,943 epoch 5 - iter 99/992 - loss 0.17177104 - time (sec): 2.28 - samples/sec: 7353.67 - lr: 0.000033 - momentum: 0.000000
2023-10-18 20:57:45,125 epoch 5 - iter 198/992 - loss 0.18004855 - time (sec): 4.46 - samples/sec: 7278.40 - lr: 0.000032 - momentum: 0.000000
2023-10-18 20:57:47,298 epoch 5 - iter 297/992 - loss 0.18559241 - time (sec): 6.63 - samples/sec: 7192.47 - lr: 0.000032 - momentum: 0.000000
2023-10-18 20:57:49,624 epoch 5 - iter 396/992 - loss 0.18522087 - time (sec): 8.96 - samples/sec: 7199.80 - lr: 0.000031 - momentum: 0.000000
2023-10-18 20:57:51,819 epoch 5 - iter 495/992 - loss 0.18653077 - time (sec): 11.15 - samples/sec: 7216.72 - lr: 0.000031 - momentum: 0.000000
2023-10-18 20:57:54,038 epoch 5 - iter 594/992 - loss 0.18784304 - time (sec): 13.37 - samples/sec: 7224.68 - lr: 0.000030 - momentum: 0.000000
2023-10-18 20:57:56,273 epoch 5 - iter 693/992 - loss 0.18868299 - time (sec): 15.61 - samples/sec: 7255.64 - lr: 0.000029 - momentum: 0.000000
2023-10-18 20:57:58,607 epoch 5 - iter 792/992 - loss 0.18764218 - time (sec): 17.94 - samples/sec: 7244.21 - lr: 0.000029 - momentum: 0.000000
2023-10-18 20:58:00,866 epoch 5 - iter 891/992 - loss 0.18960836 - time (sec): 20.20 - samples/sec: 7275.43 - lr: 0.000028 - momentum: 0.000000
2023-10-18 20:58:03,136 epoch 5 - iter 990/992 - loss 0.19225274 - time (sec): 22.47 - samples/sec: 7281.35 - lr: 0.000028 - momentum: 0.000000
2023-10-18 20:58:03,191 ----------------------------------------------------------------------------------------------------
2023-10-18 20:58:03,191 EPOCH 5 done: loss 0.1923 - lr: 0.000028
2023-10-18 20:58:05,000 DEV : loss 0.15571151673793793 - f1-score (micro avg) 0.429
2023-10-18 20:58:05,018 saving best model
2023-10-18 20:58:05,051 ----------------------------------------------------------------------------------------------------
2023-10-18 20:58:07,306 epoch 6 - iter 99/992 - loss 0.17941164 - time (sec): 2.25 - samples/sec: 7240.83 - lr: 0.000027 - momentum: 0.000000
2023-10-18 20:58:09,525 epoch 6 - iter 198/992 - loss 0.17285384 - time (sec): 4.47 - samples/sec: 7150.39 - lr: 0.000027 - momentum: 0.000000
2023-10-18 20:58:11,791 epoch 6 - iter 297/992 - loss 0.17178395 - time (sec): 6.74 - samples/sec: 7272.32 - lr: 0.000026 - momentum: 0.000000
2023-10-18 20:58:14,001 epoch 6 - iter 396/992 - loss 0.17127824 - time (sec): 8.95 - samples/sec: 7269.39 - lr: 0.000026 - momentum: 0.000000
2023-10-18 20:58:16,238 epoch 6 - iter 495/992 - loss 0.17366195 - time (sec): 11.19 - samples/sec: 7241.92 - lr: 0.000025 - momentum: 0.000000
2023-10-18 20:58:18,505 epoch 6 - iter 594/992 - loss 0.17202885 - time (sec): 13.45 - samples/sec: 7246.43 - lr: 0.000024 - momentum: 0.000000
2023-10-18 20:58:20,637 epoch 6 - iter 693/992 - loss 0.17315227 - time (sec): 15.59 - samples/sec: 7274.96 - lr: 0.000024 - momentum: 0.000000
2023-10-18 20:58:22,852 epoch 6 - iter 792/992 - loss 0.17320411 - time (sec): 17.80 - samples/sec: 7271.31 - lr: 0.000023 - momentum: 0.000000
2023-10-18 20:58:25,041 epoch 6 - iter 891/992 - loss 0.17490211 - time (sec): 19.99 - samples/sec: 7285.57 - lr: 0.000023 - momentum: 0.000000
2023-10-18 20:58:27,314 epoch 6 - iter 990/992 - loss 0.17714492 - time (sec): 22.26 - samples/sec: 7351.83 - lr: 0.000022 - momentum: 0.000000
2023-10-18 20:58:27,362 ----------------------------------------------------------------------------------------------------
2023-10-18 20:58:27,363 EPOCH 6 done: loss 0.1770 - lr: 0.000022
2023-10-18 20:58:29,203 DEV : loss 0.15208245813846588 - f1-score (micro avg) 0.4522
2023-10-18 20:58:29,221 saving best model
2023-10-18 20:58:29,259 ----------------------------------------------------------------------------------------------------
2023-10-18 20:58:31,472 epoch 7 - iter 99/992 - loss 0.18307759 - time (sec): 2.21 - samples/sec: 7285.65 - lr: 0.000022 - momentum: 0.000000
2023-10-18 20:58:33,665 epoch 7 - iter 198/992 - loss 0.17513144 - time (sec): 4.41 - samples/sec: 7284.45 - lr: 0.000021 - momentum: 0.000000
2023-10-18 20:58:35,924 epoch 7 - iter 297/992 - loss 0.17363848 - time (sec): 6.66 - samples/sec: 7347.88 - lr: 0.000021 - momentum: 0.000000
2023-10-18 20:58:38,152 epoch 7 - iter 396/992 - loss 0.17849995 - time (sec): 8.89 - samples/sec: 7356.97 - lr: 0.000020 - momentum: 0.000000
2023-10-18 20:58:40,363 epoch 7 - iter 495/992 - loss 0.17302936 - time (sec): 11.10 - samples/sec: 7373.04 - lr: 0.000019 - momentum: 0.000000
2023-10-18 20:58:42,608 epoch 7 - iter 594/992 - loss 0.17173804 - time (sec): 13.35 - samples/sec: 7350.59 - lr: 0.000019 - momentum: 0.000000
2023-10-18 20:58:44,842 epoch 7 - iter 693/992 - loss 0.17176977 - time (sec): 15.58 - samples/sec: 7321.56 - lr: 0.000018 - momentum: 0.000000
2023-10-18 20:58:47,119 epoch 7 - iter 792/992 - loss 0.17139044 - time (sec): 17.86 - samples/sec: 7282.96 - lr: 0.000018 - momentum: 0.000000
2023-10-18 20:58:49,409 epoch 7 - iter 891/992 - loss 0.16779114 - time (sec): 20.15 - samples/sec: 7332.95 - lr: 0.000017 - momentum: 0.000000
2023-10-18 20:58:51,632 epoch 7 - iter 990/992 - loss 0.16865105 - time (sec): 22.37 - samples/sec: 7315.20 - lr: 0.000017 - momentum: 0.000000
2023-10-18 20:58:51,674 ----------------------------------------------------------------------------------------------------
2023-10-18 20:58:51,675 EPOCH 7 done: loss 0.1684 - lr: 0.000017
2023-10-18 20:58:53,875 DEV : loss 0.14835909008979797 - f1-score (micro avg) 0.4563
2023-10-18 20:58:53,893 saving best model
2023-10-18 20:58:53,930 ----------------------------------------------------------------------------------------------------
2023-10-18 20:58:56,137 epoch 8 - iter 99/992 - loss 0.16559554 - time (sec): 2.21 - samples/sec: 7584.76 - lr: 0.000016 - momentum: 0.000000
2023-10-18 20:58:58,402 epoch 8 - iter 198/992 - loss 0.15713025 - time (sec): 4.47 - samples/sec: 7401.59 - lr: 0.000016 - momentum: 0.000000
2023-10-18 20:59:00,698 epoch 8 - iter 297/992 - loss 0.15886284 - time (sec): 6.77 - samples/sec: 7505.02 - lr: 0.000015 - momentum: 0.000000
2023-10-18 20:59:02,941 epoch 8 - iter 396/992 - loss 0.16091614 - time (sec): 9.01 - samples/sec: 7381.27 - lr: 0.000014 - momentum: 0.000000
2023-10-18 20:59:05,171 epoch 8 - iter 495/992 - loss 0.15847892 - time (sec): 11.24 - samples/sec: 7290.06 - lr: 0.000014 - momentum: 0.000000
2023-10-18 20:59:07,413 epoch 8 - iter 594/992 - loss 0.16400434 - time (sec): 13.48 - samples/sec: 7279.96 - lr: 0.000013 - momentum: 0.000000
2023-10-18 20:59:09,638 epoch 8 - iter 693/992 - loss 0.16260905 - time (sec): 15.71 - samples/sec: 7267.37 - lr: 0.000013 - momentum: 0.000000
2023-10-18 20:59:11,837 epoch 8 - iter 792/992 - loss 0.16132712 - time (sec): 17.91 - samples/sec: 7275.69 - lr: 0.000012 - momentum: 0.000000
2023-10-18 20:59:14,069 epoch 8 - iter 891/992 - loss 0.16199039 - time (sec): 20.14 - samples/sec: 7290.01 - lr: 0.000012 - momentum: 0.000000
2023-10-18 20:59:16,290 epoch 8 - iter 990/992 - loss 0.16199479 - time (sec): 22.36 - samples/sec: 7317.37 - lr: 0.000011 - momentum: 0.000000
2023-10-18 20:59:16,333 ----------------------------------------------------------------------------------------------------
2023-10-18 20:59:16,334 EPOCH 8 done: loss 0.1618 - lr: 0.000011
2023-10-18 20:59:18,140 DEV : loss 0.14870117604732513 - f1-score (micro avg) 0.4715
2023-10-18 20:59:18,158 saving best model
2023-10-18 20:59:18,192 ----------------------------------------------------------------------------------------------------
2023-10-18 20:59:20,399 epoch 9 - iter 99/992 - loss 0.16182854 - time (sec): 2.21 - samples/sec: 7607.41 - lr: 0.000011 - momentum: 0.000000
2023-10-18 20:59:22,613 epoch 9 - iter 198/992 - loss 0.16350312 - time (sec): 4.42 - samples/sec: 7814.63 - lr: 0.000010 - momentum: 0.000000
2023-10-18 20:59:24,553 epoch 9 - iter 297/992 - loss 0.15746054 - time (sec): 6.36 - samples/sec: 7922.20 - lr: 0.000009 - momentum: 0.000000
2023-10-18 20:59:26,606 epoch 9 - iter 396/992 - loss 0.15577802 - time (sec): 8.41 - samples/sec: 7832.63 - lr: 0.000009 - momentum: 0.000000
2023-10-18 20:59:28,852 epoch 9 - iter 495/992 - loss 0.15518974 - time (sec): 10.66 - samples/sec: 7730.86 - lr: 0.000008 - momentum: 0.000000
2023-10-18 20:59:31,075 epoch 9 - iter 594/992 - loss 0.16021182 - time (sec): 12.88 - samples/sec: 7660.07 - lr: 0.000008 - momentum: 0.000000
2023-10-18 20:59:33,419 epoch 9 - iter 693/992 - loss 0.16018544 - time (sec): 15.23 - samples/sec: 7495.81 - lr: 0.000007 - momentum: 0.000000
2023-10-18 20:59:35,680 epoch 9 - iter 792/992 - loss 0.15779920 - time (sec): 17.49 - samples/sec: 7456.85 - lr: 0.000007 - momentum: 0.000000
2023-10-18 20:59:37,888 epoch 9 - iter 891/992 - loss 0.15767464 - time (sec): 19.70 - samples/sec: 7472.73 - lr: 0.000006 - momentum: 0.000000
2023-10-18 20:59:40,148 epoch 9 - iter 990/992 - loss 0.15624567 - time (sec): 21.96 - samples/sec: 7453.85 - lr: 0.000006 - momentum: 0.000000
2023-10-18 20:59:40,194 ----------------------------------------------------------------------------------------------------
2023-10-18 20:59:40,194 EPOCH 9 done: loss 0.1560 - lr: 0.000006
2023-10-18 20:59:42,035 DEV : loss 0.14903637766838074 - f1-score (micro avg) 0.4845
2023-10-18 20:59:42,054 saving best model
2023-10-18 20:59:42,087 ----------------------------------------------------------------------------------------------------
2023-10-18 20:59:44,366 epoch 10 - iter 99/992 - loss 0.16504639 - time (sec): 2.28 - samples/sec: 6883.04 - lr: 0.000005 - momentum: 0.000000
2023-10-18 20:59:46,843 epoch 10 - iter 198/992 - loss 0.15098993 - time (sec): 4.76 - samples/sec: 6839.12 - lr: 0.000004 - momentum: 0.000000
2023-10-18 20:59:49,054 epoch 10 - iter 297/992 - loss 0.15199686 - time (sec): 6.97 - samples/sec: 7033.05 - lr: 0.000004 - momentum: 0.000000
2023-10-18 20:59:51,283 epoch 10 - iter 396/992 - loss 0.15271679 - time (sec): 9.19 - samples/sec: 7085.91 - lr: 0.000003 - momentum: 0.000000
2023-10-18 20:59:53,521 epoch 10 - iter 495/992 - loss 0.15018940 - time (sec): 11.43 - samples/sec: 7162.47 - lr: 0.000003 - momentum: 0.000000
2023-10-18 20:59:55,709 epoch 10 - iter 594/992 - loss 0.15324209 - time (sec): 13.62 - samples/sec: 7237.12 - lr: 0.000002 - momentum: 0.000000
2023-10-18 20:59:57,949 epoch 10 - iter 693/992 - loss 0.15234940 - time (sec): 15.86 - samples/sec: 7258.78 - lr: 0.000002 - momentum: 0.000000
2023-10-18 21:00:00,169 epoch 10 - iter 792/992 - loss 0.15171088 - time (sec): 18.08 - samples/sec: 7283.86 - lr: 0.000001 - momentum: 0.000000
2023-10-18 21:00:02,422 epoch 10 - iter 891/992 - loss 0.15226960 - time (sec): 20.33 - samples/sec: 7245.68 - lr: 0.000001 - momentum: 0.000000
2023-10-18 21:00:04,665 epoch 10 - iter 990/992 - loss 0.15307673 - time (sec): 22.58 - samples/sec: 7252.76 - lr: 0.000000 - momentum: 0.000000
2023-10-18 21:00:04,718 ----------------------------------------------------------------------------------------------------
2023-10-18 21:00:04,718 EPOCH 10 done: loss 0.1532 - lr: 0.000000
2023-10-18 21:00:06,578 DEV : loss 0.1486775279045105 - f1-score (micro avg) 0.4917
2023-10-18 21:00:06,597 saving best model
2023-10-18 21:00:06,663 ----------------------------------------------------------------------------------------------------
2023-10-18 21:00:06,664 Loading model from best epoch ...
2023-10-18 21:00:06,735 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
2023-10-18 21:00:08,257
Results:
- F-score (micro) 0.545
- F-score (macro) 0.3587
- Accuracy 0.4189
By class:
precision recall f1-score support
LOC 0.6909 0.6824 0.6866 655
PER 0.2964 0.5157 0.3764 223
ORG 0.0385 0.0079 0.0131 127
micro avg 0.5306 0.5602 0.5450 1005
macro avg 0.3419 0.4020 0.3587 1005
weighted avg 0.5209 0.5602 0.5327 1005
2023-10-18 21:00:08,257 ----------------------------------------------------------------------------------------------------
|