File size: 24,070 Bytes
da1a668 |
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 |
2023-10-18 16:47:47,857 ----------------------------------------------------------------------------------------------------
2023-10-18 16:47:47,857 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 16:47:47,858 ----------------------------------------------------------------------------------------------------
2023-10-18 16:47:47,858 MultiCorpus: 966 train + 219 dev + 204 test sentences
- NER_HIPE_2022 Corpus: 966 train + 219 dev + 204 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/ajmc/fr/with_doc_seperator
2023-10-18 16:47:47,858 ----------------------------------------------------------------------------------------------------
2023-10-18 16:47:47,858 Train: 966 sentences
2023-10-18 16:47:47,858 (train_with_dev=False, train_with_test=False)
2023-10-18 16:47:47,858 ----------------------------------------------------------------------------------------------------
2023-10-18 16:47:47,858 Training Params:
2023-10-18 16:47:47,858 - learning_rate: "5e-05"
2023-10-18 16:47:47,858 - mini_batch_size: "4"
2023-10-18 16:47:47,858 - max_epochs: "10"
2023-10-18 16:47:47,858 - shuffle: "True"
2023-10-18 16:47:47,858 ----------------------------------------------------------------------------------------------------
2023-10-18 16:47:47,858 Plugins:
2023-10-18 16:47:47,858 - TensorboardLogger
2023-10-18 16:47:47,858 - LinearScheduler | warmup_fraction: '0.1'
2023-10-18 16:47:47,858 ----------------------------------------------------------------------------------------------------
2023-10-18 16:47:47,858 Final evaluation on model from best epoch (best-model.pt)
2023-10-18 16:47:47,858 - metric: "('micro avg', 'f1-score')"
2023-10-18 16:47:47,858 ----------------------------------------------------------------------------------------------------
2023-10-18 16:47:47,858 Computation:
2023-10-18 16:47:47,858 - compute on device: cuda:0
2023-10-18 16:47:47,858 - embedding storage: none
2023-10-18 16:47:47,858 ----------------------------------------------------------------------------------------------------
2023-10-18 16:47:47,858 Model training base path: "hmbench-ajmc/fr-dbmdz/bert-tiny-historic-multilingual-cased-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-5"
2023-10-18 16:47:47,858 ----------------------------------------------------------------------------------------------------
2023-10-18 16:47:47,858 ----------------------------------------------------------------------------------------------------
2023-10-18 16:47:47,859 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-18 16:47:48,254 epoch 1 - iter 24/242 - loss 3.75850769 - time (sec): 0.40 - samples/sec: 5890.04 - lr: 0.000005 - momentum: 0.000000
2023-10-18 16:47:48,614 epoch 1 - iter 48/242 - loss 3.69682205 - time (sec): 0.76 - samples/sec: 5693.84 - lr: 0.000010 - momentum: 0.000000
2023-10-18 16:47:49,004 epoch 1 - iter 72/242 - loss 3.55368346 - time (sec): 1.14 - samples/sec: 6244.70 - lr: 0.000015 - momentum: 0.000000
2023-10-18 16:47:49,375 epoch 1 - iter 96/242 - loss 3.39273430 - time (sec): 1.52 - samples/sec: 6261.71 - lr: 0.000020 - momentum: 0.000000
2023-10-18 16:47:49,743 epoch 1 - iter 120/242 - loss 3.19611354 - time (sec): 1.88 - samples/sec: 6331.42 - lr: 0.000025 - momentum: 0.000000
2023-10-18 16:47:50,103 epoch 1 - iter 144/242 - loss 2.97659823 - time (sec): 2.24 - samples/sec: 6258.32 - lr: 0.000030 - momentum: 0.000000
2023-10-18 16:47:50,496 epoch 1 - iter 168/242 - loss 2.68820008 - time (sec): 2.64 - samples/sec: 6368.43 - lr: 0.000035 - momentum: 0.000000
2023-10-18 16:47:50,881 epoch 1 - iter 192/242 - loss 2.43041565 - time (sec): 3.02 - samples/sec: 6522.56 - lr: 0.000039 - momentum: 0.000000
2023-10-18 16:47:51,261 epoch 1 - iter 216/242 - loss 2.23894411 - time (sec): 3.40 - samples/sec: 6554.26 - lr: 0.000044 - momentum: 0.000000
2023-10-18 16:47:51,619 epoch 1 - iter 240/242 - loss 2.10798640 - time (sec): 3.76 - samples/sec: 6556.78 - lr: 0.000049 - momentum: 0.000000
2023-10-18 16:47:51,644 ----------------------------------------------------------------------------------------------------
2023-10-18 16:47:51,644 EPOCH 1 done: loss 2.1061 - lr: 0.000049
2023-10-18 16:47:52,153 DEV : loss 0.6494866609573364 - f1-score (micro avg) 0.0
2023-10-18 16:47:52,158 ----------------------------------------------------------------------------------------------------
2023-10-18 16:47:52,526 epoch 2 - iter 24/242 - loss 0.70753707 - time (sec): 0.37 - samples/sec: 6256.75 - lr: 0.000049 - momentum: 0.000000
2023-10-18 16:47:52,897 epoch 2 - iter 48/242 - loss 0.69061854 - time (sec): 0.74 - samples/sec: 6476.07 - lr: 0.000049 - momentum: 0.000000
2023-10-18 16:47:53,260 epoch 2 - iter 72/242 - loss 0.66939334 - time (sec): 1.10 - samples/sec: 6544.56 - lr: 0.000048 - momentum: 0.000000
2023-10-18 16:47:53,643 epoch 2 - iter 96/242 - loss 0.67917229 - time (sec): 1.48 - samples/sec: 6655.00 - lr: 0.000048 - momentum: 0.000000
2023-10-18 16:47:54,016 epoch 2 - iter 120/242 - loss 0.66943186 - time (sec): 1.86 - samples/sec: 6506.69 - lr: 0.000047 - momentum: 0.000000
2023-10-18 16:47:54,394 epoch 2 - iter 144/242 - loss 0.65133492 - time (sec): 2.24 - samples/sec: 6580.86 - lr: 0.000047 - momentum: 0.000000
2023-10-18 16:47:54,794 epoch 2 - iter 168/242 - loss 0.61596553 - time (sec): 2.64 - samples/sec: 6568.16 - lr: 0.000046 - momentum: 0.000000
2023-10-18 16:47:55,162 epoch 2 - iter 192/242 - loss 0.59965228 - time (sec): 3.00 - samples/sec: 6595.00 - lr: 0.000046 - momentum: 0.000000
2023-10-18 16:47:55,521 epoch 2 - iter 216/242 - loss 0.60012603 - time (sec): 3.36 - samples/sec: 6573.83 - lr: 0.000045 - momentum: 0.000000
2023-10-18 16:47:55,887 epoch 2 - iter 240/242 - loss 0.59113460 - time (sec): 3.73 - samples/sec: 6573.62 - lr: 0.000045 - momentum: 0.000000
2023-10-18 16:47:55,915 ----------------------------------------------------------------------------------------------------
2023-10-18 16:47:55,915 EPOCH 2 done: loss 0.5933 - lr: 0.000045
2023-10-18 16:47:56,345 DEV : loss 0.3973432183265686 - f1-score (micro avg) 0.2292
2023-10-18 16:47:56,350 saving best model
2023-10-18 16:47:56,377 ----------------------------------------------------------------------------------------------------
2023-10-18 16:47:56,758 epoch 3 - iter 24/242 - loss 0.47094547 - time (sec): 0.38 - samples/sec: 6558.17 - lr: 0.000044 - momentum: 0.000000
2023-10-18 16:47:57,124 epoch 3 - iter 48/242 - loss 0.49377345 - time (sec): 0.75 - samples/sec: 6428.73 - lr: 0.000043 - momentum: 0.000000
2023-10-18 16:47:57,481 epoch 3 - iter 72/242 - loss 0.47276259 - time (sec): 1.10 - samples/sec: 6393.37 - lr: 0.000043 - momentum: 0.000000
2023-10-18 16:47:57,859 epoch 3 - iter 96/242 - loss 0.47065066 - time (sec): 1.48 - samples/sec: 6450.24 - lr: 0.000042 - momentum: 0.000000
2023-10-18 16:47:58,250 epoch 3 - iter 120/242 - loss 0.46440215 - time (sec): 1.87 - samples/sec: 6423.01 - lr: 0.000042 - momentum: 0.000000
2023-10-18 16:47:58,622 epoch 3 - iter 144/242 - loss 0.44989391 - time (sec): 2.24 - samples/sec: 6365.98 - lr: 0.000041 - momentum: 0.000000
2023-10-18 16:47:58,996 epoch 3 - iter 168/242 - loss 0.44144086 - time (sec): 2.62 - samples/sec: 6442.83 - lr: 0.000041 - momentum: 0.000000
2023-10-18 16:47:59,377 epoch 3 - iter 192/242 - loss 0.43386345 - time (sec): 3.00 - samples/sec: 6568.59 - lr: 0.000040 - momentum: 0.000000
2023-10-18 16:47:59,760 epoch 3 - iter 216/242 - loss 0.42346837 - time (sec): 3.38 - samples/sec: 6565.04 - lr: 0.000040 - momentum: 0.000000
2023-10-18 16:48:00,133 epoch 3 - iter 240/242 - loss 0.42387037 - time (sec): 3.76 - samples/sec: 6544.86 - lr: 0.000039 - momentum: 0.000000
2023-10-18 16:48:00,159 ----------------------------------------------------------------------------------------------------
2023-10-18 16:48:00,160 EPOCH 3 done: loss 0.4236 - lr: 0.000039
2023-10-18 16:48:00,579 DEV : loss 0.315167635679245 - f1-score (micro avg) 0.5118
2023-10-18 16:48:00,583 saving best model
2023-10-18 16:48:00,617 ----------------------------------------------------------------------------------------------------
2023-10-18 16:48:00,976 epoch 4 - iter 24/242 - loss 0.37876134 - time (sec): 0.36 - samples/sec: 5829.16 - lr: 0.000038 - momentum: 0.000000
2023-10-18 16:48:01,352 epoch 4 - iter 48/242 - loss 0.39303874 - time (sec): 0.73 - samples/sec: 6092.55 - lr: 0.000038 - momentum: 0.000000
2023-10-18 16:48:01,733 epoch 4 - iter 72/242 - loss 0.38107267 - time (sec): 1.11 - samples/sec: 6523.26 - lr: 0.000037 - momentum: 0.000000
2023-10-18 16:48:02,105 epoch 4 - iter 96/242 - loss 0.38620105 - time (sec): 1.49 - samples/sec: 6472.65 - lr: 0.000037 - momentum: 0.000000
2023-10-18 16:48:02,472 epoch 4 - iter 120/242 - loss 0.38815951 - time (sec): 1.85 - samples/sec: 6515.11 - lr: 0.000036 - momentum: 0.000000
2023-10-18 16:48:02,836 epoch 4 - iter 144/242 - loss 0.37887595 - time (sec): 2.22 - samples/sec: 6579.66 - lr: 0.000036 - momentum: 0.000000
2023-10-18 16:48:03,210 epoch 4 - iter 168/242 - loss 0.37809209 - time (sec): 2.59 - samples/sec: 6674.66 - lr: 0.000035 - momentum: 0.000000
2023-10-18 16:48:03,586 epoch 4 - iter 192/242 - loss 0.36934478 - time (sec): 2.97 - samples/sec: 6664.24 - lr: 0.000035 - momentum: 0.000000
2023-10-18 16:48:03,939 epoch 4 - iter 216/242 - loss 0.37593866 - time (sec): 3.32 - samples/sec: 6695.54 - lr: 0.000034 - momentum: 0.000000
2023-10-18 16:48:04,276 epoch 4 - iter 240/242 - loss 0.36981793 - time (sec): 3.66 - samples/sec: 6742.05 - lr: 0.000033 - momentum: 0.000000
2023-10-18 16:48:04,300 ----------------------------------------------------------------------------------------------------
2023-10-18 16:48:04,300 EPOCH 4 done: loss 0.3695 - lr: 0.000033
2023-10-18 16:48:04,730 DEV : loss 0.2958347797393799 - f1-score (micro avg) 0.512
2023-10-18 16:48:04,734 saving best model
2023-10-18 16:48:04,769 ----------------------------------------------------------------------------------------------------
2023-10-18 16:48:05,110 epoch 5 - iter 24/242 - loss 0.35480541 - time (sec): 0.34 - samples/sec: 7224.38 - lr: 0.000033 - momentum: 0.000000
2023-10-18 16:48:05,450 epoch 5 - iter 48/242 - loss 0.35459295 - time (sec): 0.68 - samples/sec: 7398.74 - lr: 0.000032 - momentum: 0.000000
2023-10-18 16:48:05,784 epoch 5 - iter 72/242 - loss 0.34181991 - time (sec): 1.01 - samples/sec: 7421.59 - lr: 0.000032 - momentum: 0.000000
2023-10-18 16:48:06,124 epoch 5 - iter 96/242 - loss 0.35420576 - time (sec): 1.35 - samples/sec: 7517.23 - lr: 0.000031 - momentum: 0.000000
2023-10-18 16:48:06,457 epoch 5 - iter 120/242 - loss 0.35768803 - time (sec): 1.69 - samples/sec: 7460.18 - lr: 0.000031 - momentum: 0.000000
2023-10-18 16:48:06,792 epoch 5 - iter 144/242 - loss 0.35518931 - time (sec): 2.02 - samples/sec: 7420.94 - lr: 0.000030 - momentum: 0.000000
2023-10-18 16:48:07,137 epoch 5 - iter 168/242 - loss 0.34918538 - time (sec): 2.37 - samples/sec: 7362.92 - lr: 0.000030 - momentum: 0.000000
2023-10-18 16:48:07,471 epoch 5 - iter 192/242 - loss 0.34897337 - time (sec): 2.70 - samples/sec: 7380.41 - lr: 0.000029 - momentum: 0.000000
2023-10-18 16:48:07,809 epoch 5 - iter 216/242 - loss 0.33635374 - time (sec): 3.04 - samples/sec: 7341.39 - lr: 0.000028 - momentum: 0.000000
2023-10-18 16:48:08,132 epoch 5 - iter 240/242 - loss 0.33487180 - time (sec): 3.36 - samples/sec: 7325.66 - lr: 0.000028 - momentum: 0.000000
2023-10-18 16:48:08,157 ----------------------------------------------------------------------------------------------------
2023-10-18 16:48:08,157 EPOCH 5 done: loss 0.3352 - lr: 0.000028
2023-10-18 16:48:08,597 DEV : loss 0.2639392018318176 - f1-score (micro avg) 0.5037
2023-10-18 16:48:08,602 ----------------------------------------------------------------------------------------------------
2023-10-18 16:48:08,962 epoch 6 - iter 24/242 - loss 0.31375370 - time (sec): 0.36 - samples/sec: 7149.00 - lr: 0.000027 - momentum: 0.000000
2023-10-18 16:48:09,335 epoch 6 - iter 48/242 - loss 0.31377803 - time (sec): 0.73 - samples/sec: 6842.22 - lr: 0.000027 - momentum: 0.000000
2023-10-18 16:48:09,710 epoch 6 - iter 72/242 - loss 0.32611832 - time (sec): 1.11 - samples/sec: 6809.96 - lr: 0.000026 - momentum: 0.000000
2023-10-18 16:48:10,072 epoch 6 - iter 96/242 - loss 0.29529436 - time (sec): 1.47 - samples/sec: 6638.72 - lr: 0.000026 - momentum: 0.000000
2023-10-18 16:48:10,448 epoch 6 - iter 120/242 - loss 0.29275130 - time (sec): 1.85 - samples/sec: 6661.09 - lr: 0.000025 - momentum: 0.000000
2023-10-18 16:48:10,841 epoch 6 - iter 144/242 - loss 0.29614971 - time (sec): 2.24 - samples/sec: 6602.79 - lr: 0.000025 - momentum: 0.000000
2023-10-18 16:48:11,207 epoch 6 - iter 168/242 - loss 0.30099987 - time (sec): 2.60 - samples/sec: 6576.18 - lr: 0.000024 - momentum: 0.000000
2023-10-18 16:48:11,598 epoch 6 - iter 192/242 - loss 0.30400838 - time (sec): 3.00 - samples/sec: 6568.08 - lr: 0.000023 - momentum: 0.000000
2023-10-18 16:48:11,968 epoch 6 - iter 216/242 - loss 0.30647937 - time (sec): 3.37 - samples/sec: 6519.76 - lr: 0.000023 - momentum: 0.000000
2023-10-18 16:48:12,357 epoch 6 - iter 240/242 - loss 0.30712937 - time (sec): 3.75 - samples/sec: 6543.49 - lr: 0.000022 - momentum: 0.000000
2023-10-18 16:48:12,383 ----------------------------------------------------------------------------------------------------
2023-10-18 16:48:12,384 EPOCH 6 done: loss 0.3079 - lr: 0.000022
2023-10-18 16:48:12,821 DEV : loss 0.2603849470615387 - f1-score (micro avg) 0.5104
2023-10-18 16:48:12,826 ----------------------------------------------------------------------------------------------------
2023-10-18 16:48:13,215 epoch 7 - iter 24/242 - loss 0.34851167 - time (sec): 0.39 - samples/sec: 6616.61 - lr: 0.000022 - momentum: 0.000000
2023-10-18 16:48:13,577 epoch 7 - iter 48/242 - loss 0.34722378 - time (sec): 0.75 - samples/sec: 6247.39 - lr: 0.000021 - momentum: 0.000000
2023-10-18 16:48:13,944 epoch 7 - iter 72/242 - loss 0.32206806 - time (sec): 1.12 - samples/sec: 6327.44 - lr: 0.000021 - momentum: 0.000000
2023-10-18 16:48:14,324 epoch 7 - iter 96/242 - loss 0.30779121 - time (sec): 1.50 - samples/sec: 6279.87 - lr: 0.000020 - momentum: 0.000000
2023-10-18 16:48:14,695 epoch 7 - iter 120/242 - loss 0.29721250 - time (sec): 1.87 - samples/sec: 6374.73 - lr: 0.000020 - momentum: 0.000000
2023-10-18 16:48:15,019 epoch 7 - iter 144/242 - loss 0.29074408 - time (sec): 2.19 - samples/sec: 6609.94 - lr: 0.000019 - momentum: 0.000000
2023-10-18 16:48:15,322 epoch 7 - iter 168/242 - loss 0.28770394 - time (sec): 2.50 - samples/sec: 6789.27 - lr: 0.000018 - momentum: 0.000000
2023-10-18 16:48:15,738 epoch 7 - iter 192/242 - loss 0.28551813 - time (sec): 2.91 - samples/sec: 6828.57 - lr: 0.000018 - momentum: 0.000000
2023-10-18 16:48:16,136 epoch 7 - iter 216/242 - loss 0.28346652 - time (sec): 3.31 - samples/sec: 6699.72 - lr: 0.000017 - momentum: 0.000000
2023-10-18 16:48:16,525 epoch 7 - iter 240/242 - loss 0.28920613 - time (sec): 3.70 - samples/sec: 6634.99 - lr: 0.000017 - momentum: 0.000000
2023-10-18 16:48:16,553 ----------------------------------------------------------------------------------------------------
2023-10-18 16:48:16,553 EPOCH 7 done: loss 0.2910 - lr: 0.000017
2023-10-18 16:48:16,998 DEV : loss 0.24460992217063904 - f1-score (micro avg) 0.5291
2023-10-18 16:48:17,002 saving best model
2023-10-18 16:48:17,034 ----------------------------------------------------------------------------------------------------
2023-10-18 16:48:17,417 epoch 8 - iter 24/242 - loss 0.26034674 - time (sec): 0.38 - samples/sec: 6988.76 - lr: 0.000016 - momentum: 0.000000
2023-10-18 16:48:17,790 epoch 8 - iter 48/242 - loss 0.27344716 - time (sec): 0.76 - samples/sec: 6928.02 - lr: 0.000016 - momentum: 0.000000
2023-10-18 16:48:18,146 epoch 8 - iter 72/242 - loss 0.29604730 - time (sec): 1.11 - samples/sec: 6753.89 - lr: 0.000015 - momentum: 0.000000
2023-10-18 16:48:18,516 epoch 8 - iter 96/242 - loss 0.28714663 - time (sec): 1.48 - samples/sec: 6654.86 - lr: 0.000015 - momentum: 0.000000
2023-10-18 16:48:18,918 epoch 8 - iter 120/242 - loss 0.28223276 - time (sec): 1.88 - samples/sec: 6585.68 - lr: 0.000014 - momentum: 0.000000
2023-10-18 16:48:19,302 epoch 8 - iter 144/242 - loss 0.29541982 - time (sec): 2.27 - samples/sec: 6662.43 - lr: 0.000013 - momentum: 0.000000
2023-10-18 16:48:19,670 epoch 8 - iter 168/242 - loss 0.29484539 - time (sec): 2.63 - samples/sec: 6590.53 - lr: 0.000013 - momentum: 0.000000
2023-10-18 16:48:20,039 epoch 8 - iter 192/242 - loss 0.28665546 - time (sec): 3.00 - samples/sec: 6571.13 - lr: 0.000012 - momentum: 0.000000
2023-10-18 16:48:20,419 epoch 8 - iter 216/242 - loss 0.28212190 - time (sec): 3.38 - samples/sec: 6572.82 - lr: 0.000012 - momentum: 0.000000
2023-10-18 16:48:20,800 epoch 8 - iter 240/242 - loss 0.28301868 - time (sec): 3.76 - samples/sec: 6532.36 - lr: 0.000011 - momentum: 0.000000
2023-10-18 16:48:20,830 ----------------------------------------------------------------------------------------------------
2023-10-18 16:48:20,830 EPOCH 8 done: loss 0.2845 - lr: 0.000011
2023-10-18 16:48:21,262 DEV : loss 0.24663908779621124 - f1-score (micro avg) 0.547
2023-10-18 16:48:21,266 saving best model
2023-10-18 16:48:21,306 ----------------------------------------------------------------------------------------------------
2023-10-18 16:48:21,698 epoch 9 - iter 24/242 - loss 0.25854915 - time (sec): 0.39 - samples/sec: 5206.54 - lr: 0.000011 - momentum: 0.000000
2023-10-18 16:48:22,074 epoch 9 - iter 48/242 - loss 0.26339176 - time (sec): 0.77 - samples/sec: 5874.13 - lr: 0.000010 - momentum: 0.000000
2023-10-18 16:48:22,487 epoch 9 - iter 72/242 - loss 0.27772691 - time (sec): 1.18 - samples/sec: 6048.06 - lr: 0.000010 - momentum: 0.000000
2023-10-18 16:48:22,904 epoch 9 - iter 96/242 - loss 0.29699368 - time (sec): 1.60 - samples/sec: 6003.44 - lr: 0.000009 - momentum: 0.000000
2023-10-18 16:48:23,309 epoch 9 - iter 120/242 - loss 0.29822567 - time (sec): 2.00 - samples/sec: 6020.52 - lr: 0.000008 - momentum: 0.000000
2023-10-18 16:48:23,675 epoch 9 - iter 144/242 - loss 0.28198629 - time (sec): 2.37 - samples/sec: 6033.25 - lr: 0.000008 - momentum: 0.000000
2023-10-18 16:48:24,029 epoch 9 - iter 168/242 - loss 0.28545819 - time (sec): 2.72 - samples/sec: 6058.61 - lr: 0.000007 - momentum: 0.000000
2023-10-18 16:48:24,407 epoch 9 - iter 192/242 - loss 0.28498541 - time (sec): 3.10 - samples/sec: 6204.10 - lr: 0.000007 - momentum: 0.000000
2023-10-18 16:48:24,789 epoch 9 - iter 216/242 - loss 0.27924178 - time (sec): 3.48 - samples/sec: 6316.12 - lr: 0.000006 - momentum: 0.000000
2023-10-18 16:48:25,167 epoch 9 - iter 240/242 - loss 0.27642574 - time (sec): 3.86 - samples/sec: 6399.42 - lr: 0.000006 - momentum: 0.000000
2023-10-18 16:48:25,195 ----------------------------------------------------------------------------------------------------
2023-10-18 16:48:25,195 EPOCH 9 done: loss 0.2766 - lr: 0.000006
2023-10-18 16:48:25,629 DEV : loss 0.2363634556531906 - f1-score (micro avg) 0.5531
2023-10-18 16:48:25,633 saving best model
2023-10-18 16:48:25,666 ----------------------------------------------------------------------------------------------------
2023-10-18 16:48:26,029 epoch 10 - iter 24/242 - loss 0.25436219 - time (sec): 0.36 - samples/sec: 6207.54 - lr: 0.000005 - momentum: 0.000000
2023-10-18 16:48:26,399 epoch 10 - iter 48/242 - loss 0.25221311 - time (sec): 0.73 - samples/sec: 6402.65 - lr: 0.000005 - momentum: 0.000000
2023-10-18 16:48:26,783 epoch 10 - iter 72/242 - loss 0.25841165 - time (sec): 1.12 - samples/sec: 6376.23 - lr: 0.000004 - momentum: 0.000000
2023-10-18 16:48:27,177 epoch 10 - iter 96/242 - loss 0.25213135 - time (sec): 1.51 - samples/sec: 6420.68 - lr: 0.000003 - momentum: 0.000000
2023-10-18 16:48:27,546 epoch 10 - iter 120/242 - loss 0.26644826 - time (sec): 1.88 - samples/sec: 6397.10 - lr: 0.000003 - momentum: 0.000000
2023-10-18 16:48:27,924 epoch 10 - iter 144/242 - loss 0.27616057 - time (sec): 2.26 - samples/sec: 6541.61 - lr: 0.000002 - momentum: 0.000000
2023-10-18 16:48:28,299 epoch 10 - iter 168/242 - loss 0.26407937 - time (sec): 2.63 - samples/sec: 6555.33 - lr: 0.000002 - momentum: 0.000000
2023-10-18 16:48:28,661 epoch 10 - iter 192/242 - loss 0.26815631 - time (sec): 2.99 - samples/sec: 6561.04 - lr: 0.000001 - momentum: 0.000000
2023-10-18 16:48:29,039 epoch 10 - iter 216/242 - loss 0.26617524 - time (sec): 3.37 - samples/sec: 6565.71 - lr: 0.000001 - momentum: 0.000000
2023-10-18 16:48:29,406 epoch 10 - iter 240/242 - loss 0.27271415 - time (sec): 3.74 - samples/sec: 6568.71 - lr: 0.000000 - momentum: 0.000000
2023-10-18 16:48:29,435 ----------------------------------------------------------------------------------------------------
2023-10-18 16:48:29,435 EPOCH 10 done: loss 0.2712 - lr: 0.000000
2023-10-18 16:48:29,878 DEV : loss 0.23566073179244995 - f1-score (micro avg) 0.5531
2023-10-18 16:48:29,912 ----------------------------------------------------------------------------------------------------
2023-10-18 16:48:29,913 Loading model from best epoch ...
2023-10-18 16:48:29,981 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-object, B-object, E-object, I-object, S-date, B-date, E-date, I-date
2023-10-18 16:48:30,398
Results:
- F-score (micro) 0.4879
- F-score (macro) 0.2774
- Accuracy 0.3387
By class:
precision recall f1-score support
scope 0.3192 0.5271 0.3977 129
pers 0.6154 0.7482 0.6753 139
work 0.4634 0.2375 0.3140 80
loc 0.0000 0.0000 0.0000 9
date 0.0000 0.0000 0.0000 3
micro avg 0.4515 0.5306 0.4879 360
macro avg 0.2796 0.3026 0.2774 360
weighted avg 0.4550 0.5306 0.4730 360
2023-10-18 16:48:30,398 ----------------------------------------------------------------------------------------------------
|