File size: 24,221 Bytes
0a83ca6 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 |
2023-10-17 21:12:15,815 ----------------------------------------------------------------------------------------------------
2023-10-17 21:12:15,816 Model: "SequenceTagger(
(embeddings): TransformerWordEmbeddings(
(model): ElectraModel(
(embeddings): ElectraEmbeddings(
(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): ElectraEncoder(
(layer): ModuleList(
(0-11): 12 x ElectraLayer(
(attention): ElectraAttention(
(self): ElectraSelfAttention(
(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): ElectraSelfOutput(
(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): ElectraIntermediate(
(dense): Linear(in_features=768, out_features=3072, bias=True)
(intermediate_act_fn): GELUActivation()
)
(output): ElectraOutput(
(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)
)
)
)
)
)
)
(locked_dropout): LockedDropout(p=0.5)
(linear): Linear(in_features=768, out_features=21, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-17 21:12:15,816 ----------------------------------------------------------------------------------------------------
2023-10-17 21:12:15,816 MultiCorpus: 5901 train + 1287 dev + 1505 test sentences
- NER_HIPE_2022 Corpus: 5901 train + 1287 dev + 1505 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/hipe2020/fr/with_doc_seperator
2023-10-17 21:12:15,816 ----------------------------------------------------------------------------------------------------
2023-10-17 21:12:15,816 Train: 5901 sentences
2023-10-17 21:12:15,816 (train_with_dev=False, train_with_test=False)
2023-10-17 21:12:15,816 ----------------------------------------------------------------------------------------------------
2023-10-17 21:12:15,816 Training Params:
2023-10-17 21:12:15,816 - learning_rate: "5e-05"
2023-10-17 21:12:15,816 - mini_batch_size: "4"
2023-10-17 21:12:15,816 - max_epochs: "10"
2023-10-17 21:12:15,816 - shuffle: "True"
2023-10-17 21:12:15,816 ----------------------------------------------------------------------------------------------------
2023-10-17 21:12:15,816 Plugins:
2023-10-17 21:12:15,816 - TensorboardLogger
2023-10-17 21:12:15,816 - LinearScheduler | warmup_fraction: '0.1'
2023-10-17 21:12:15,817 ----------------------------------------------------------------------------------------------------
2023-10-17 21:12:15,817 Final evaluation on model from best epoch (best-model.pt)
2023-10-17 21:12:15,817 - metric: "('micro avg', 'f1-score')"
2023-10-17 21:12:15,817 ----------------------------------------------------------------------------------------------------
2023-10-17 21:12:15,817 Computation:
2023-10-17 21:12:15,817 - compute on device: cuda:0
2023-10-17 21:12:15,817 - embedding storage: none
2023-10-17 21:12:15,817 ----------------------------------------------------------------------------------------------------
2023-10-17 21:12:15,817 Model training base path: "hmbench-hipe2020/fr-hmteams/teams-base-historic-multilingual-discriminator-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-2"
2023-10-17 21:12:15,817 ----------------------------------------------------------------------------------------------------
2023-10-17 21:12:15,817 ----------------------------------------------------------------------------------------------------
2023-10-17 21:12:15,817 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-17 21:12:22,757 epoch 1 - iter 147/1476 - loss 2.71106741 - time (sec): 6.94 - samples/sec: 2321.76 - lr: 0.000005 - momentum: 0.000000
2023-10-17 21:12:30,110 epoch 1 - iter 294/1476 - loss 1.53519710 - time (sec): 14.29 - samples/sec: 2468.89 - lr: 0.000010 - momentum: 0.000000
2023-10-17 21:12:37,137 epoch 1 - iter 441/1476 - loss 1.19741914 - time (sec): 21.32 - samples/sec: 2353.25 - lr: 0.000015 - momentum: 0.000000
2023-10-17 21:12:44,328 epoch 1 - iter 588/1476 - loss 0.98125061 - time (sec): 28.51 - samples/sec: 2350.64 - lr: 0.000020 - momentum: 0.000000
2023-10-17 21:12:51,162 epoch 1 - iter 735/1476 - loss 0.84406033 - time (sec): 35.34 - samples/sec: 2352.31 - lr: 0.000025 - momentum: 0.000000
2023-10-17 21:12:58,205 epoch 1 - iter 882/1476 - loss 0.74180713 - time (sec): 42.39 - samples/sec: 2351.66 - lr: 0.000030 - momentum: 0.000000
2023-10-17 21:13:05,428 epoch 1 - iter 1029/1476 - loss 0.66397153 - time (sec): 49.61 - samples/sec: 2348.85 - lr: 0.000035 - momentum: 0.000000
2023-10-17 21:13:12,269 epoch 1 - iter 1176/1476 - loss 0.60399966 - time (sec): 56.45 - samples/sec: 2341.54 - lr: 0.000040 - momentum: 0.000000
2023-10-17 21:13:19,290 epoch 1 - iter 1323/1476 - loss 0.55840279 - time (sec): 63.47 - samples/sec: 2347.10 - lr: 0.000045 - momentum: 0.000000
2023-10-17 21:13:26,560 epoch 1 - iter 1470/1476 - loss 0.52004827 - time (sec): 70.74 - samples/sec: 2342.55 - lr: 0.000050 - momentum: 0.000000
2023-10-17 21:13:26,880 ----------------------------------------------------------------------------------------------------
2023-10-17 21:13:26,880 EPOCH 1 done: loss 0.5189 - lr: 0.000050
2023-10-17 21:13:33,183 DEV : loss 0.15781794488430023 - f1-score (micro avg) 0.7134
2023-10-17 21:13:33,228 saving best model
2023-10-17 21:13:33,643 ----------------------------------------------------------------------------------------------------
2023-10-17 21:13:40,926 epoch 2 - iter 147/1476 - loss 0.13618022 - time (sec): 7.28 - samples/sec: 2059.47 - lr: 0.000049 - momentum: 0.000000
2023-10-17 21:13:48,382 epoch 2 - iter 294/1476 - loss 0.14828668 - time (sec): 14.74 - samples/sec: 2265.73 - lr: 0.000049 - momentum: 0.000000
2023-10-17 21:13:55,425 epoch 2 - iter 441/1476 - loss 0.15814149 - time (sec): 21.78 - samples/sec: 2319.49 - lr: 0.000048 - momentum: 0.000000
2023-10-17 21:14:02,701 epoch 2 - iter 588/1476 - loss 0.15346701 - time (sec): 29.06 - samples/sec: 2360.96 - lr: 0.000048 - momentum: 0.000000
2023-10-17 21:14:09,918 epoch 2 - iter 735/1476 - loss 0.15116044 - time (sec): 36.27 - samples/sec: 2382.74 - lr: 0.000047 - momentum: 0.000000
2023-10-17 21:14:16,991 epoch 2 - iter 882/1476 - loss 0.14674645 - time (sec): 43.35 - samples/sec: 2377.52 - lr: 0.000047 - momentum: 0.000000
2023-10-17 21:14:23,920 epoch 2 - iter 1029/1476 - loss 0.14578389 - time (sec): 50.27 - samples/sec: 2352.17 - lr: 0.000046 - momentum: 0.000000
2023-10-17 21:14:30,697 epoch 2 - iter 1176/1476 - loss 0.14549586 - time (sec): 57.05 - samples/sec: 2339.49 - lr: 0.000046 - momentum: 0.000000
2023-10-17 21:14:37,676 epoch 2 - iter 1323/1476 - loss 0.14491984 - time (sec): 64.03 - samples/sec: 2336.70 - lr: 0.000045 - momentum: 0.000000
2023-10-17 21:14:44,608 epoch 2 - iter 1470/1476 - loss 0.14348353 - time (sec): 70.96 - samples/sec: 2336.48 - lr: 0.000044 - momentum: 0.000000
2023-10-17 21:14:44,869 ----------------------------------------------------------------------------------------------------
2023-10-17 21:14:44,870 EPOCH 2 done: loss 0.1433 - lr: 0.000044
2023-10-17 21:14:56,908 DEV : loss 0.13238579034805298 - f1-score (micro avg) 0.8057
2023-10-17 21:14:56,944 saving best model
2023-10-17 21:14:57,416 ----------------------------------------------------------------------------------------------------
2023-10-17 21:15:04,234 epoch 3 - iter 147/1476 - loss 0.09099243 - time (sec): 6.81 - samples/sec: 2221.06 - lr: 0.000044 - momentum: 0.000000
2023-10-17 21:15:10,879 epoch 3 - iter 294/1476 - loss 0.09767210 - time (sec): 13.46 - samples/sec: 2276.72 - lr: 0.000043 - momentum: 0.000000
2023-10-17 21:15:17,857 epoch 3 - iter 441/1476 - loss 0.09619969 - time (sec): 20.43 - samples/sec: 2337.51 - lr: 0.000043 - momentum: 0.000000
2023-10-17 21:15:25,059 epoch 3 - iter 588/1476 - loss 0.09884334 - time (sec): 27.64 - samples/sec: 2358.50 - lr: 0.000042 - momentum: 0.000000
2023-10-17 21:15:32,052 epoch 3 - iter 735/1476 - loss 0.09130138 - time (sec): 34.63 - samples/sec: 2329.09 - lr: 0.000042 - momentum: 0.000000
2023-10-17 21:15:39,127 epoch 3 - iter 882/1476 - loss 0.09409109 - time (sec): 41.70 - samples/sec: 2324.32 - lr: 0.000041 - momentum: 0.000000
2023-10-17 21:15:46,434 epoch 3 - iter 1029/1476 - loss 0.09210907 - time (sec): 49.01 - samples/sec: 2359.09 - lr: 0.000041 - momentum: 0.000000
2023-10-17 21:15:53,396 epoch 3 - iter 1176/1476 - loss 0.09301213 - time (sec): 55.97 - samples/sec: 2370.49 - lr: 0.000040 - momentum: 0.000000
2023-10-17 21:16:00,199 epoch 3 - iter 1323/1476 - loss 0.09078236 - time (sec): 62.78 - samples/sec: 2380.47 - lr: 0.000039 - momentum: 0.000000
2023-10-17 21:16:07,142 epoch 3 - iter 1470/1476 - loss 0.09325701 - time (sec): 69.72 - samples/sec: 2378.16 - lr: 0.000039 - momentum: 0.000000
2023-10-17 21:16:07,408 ----------------------------------------------------------------------------------------------------
2023-10-17 21:16:07,408 EPOCH 3 done: loss 0.0935 - lr: 0.000039
2023-10-17 21:16:18,751 DEV : loss 0.17908549308776855 - f1-score (micro avg) 0.8022
2023-10-17 21:16:18,780 ----------------------------------------------------------------------------------------------------
2023-10-17 21:16:26,621 epoch 4 - iter 147/1476 - loss 0.05955487 - time (sec): 7.84 - samples/sec: 2174.55 - lr: 0.000038 - momentum: 0.000000
2023-10-17 21:16:34,138 epoch 4 - iter 294/1476 - loss 0.06464650 - time (sec): 15.36 - samples/sec: 2300.33 - lr: 0.000038 - momentum: 0.000000
2023-10-17 21:16:41,169 epoch 4 - iter 441/1476 - loss 0.06441669 - time (sec): 22.39 - samples/sec: 2318.35 - lr: 0.000037 - momentum: 0.000000
2023-10-17 21:16:48,222 epoch 4 - iter 588/1476 - loss 0.06965834 - time (sec): 29.44 - samples/sec: 2329.10 - lr: 0.000037 - momentum: 0.000000
2023-10-17 21:16:55,135 epoch 4 - iter 735/1476 - loss 0.07027131 - time (sec): 36.35 - samples/sec: 2311.10 - lr: 0.000036 - momentum: 0.000000
2023-10-17 21:17:02,168 epoch 4 - iter 882/1476 - loss 0.06940446 - time (sec): 43.39 - samples/sec: 2315.69 - lr: 0.000036 - momentum: 0.000000
2023-10-17 21:17:08,988 epoch 4 - iter 1029/1476 - loss 0.06906629 - time (sec): 50.21 - samples/sec: 2311.07 - lr: 0.000035 - momentum: 0.000000
2023-10-17 21:17:16,069 epoch 4 - iter 1176/1476 - loss 0.06960608 - time (sec): 57.29 - samples/sec: 2318.23 - lr: 0.000034 - momentum: 0.000000
2023-10-17 21:17:23,415 epoch 4 - iter 1323/1476 - loss 0.06848630 - time (sec): 64.63 - samples/sec: 2335.06 - lr: 0.000034 - momentum: 0.000000
2023-10-17 21:17:30,550 epoch 4 - iter 1470/1476 - loss 0.06867505 - time (sec): 71.77 - samples/sec: 2310.89 - lr: 0.000033 - momentum: 0.000000
2023-10-17 21:17:30,826 ----------------------------------------------------------------------------------------------------
2023-10-17 21:17:30,827 EPOCH 4 done: loss 0.0686 - lr: 0.000033
2023-10-17 21:17:42,009 DEV : loss 0.17883314192295074 - f1-score (micro avg) 0.8315
2023-10-17 21:17:42,038 saving best model
2023-10-17 21:17:42,543 ----------------------------------------------------------------------------------------------------
2023-10-17 21:17:49,944 epoch 5 - iter 147/1476 - loss 0.04503414 - time (sec): 7.40 - samples/sec: 2375.63 - lr: 0.000033 - momentum: 0.000000
2023-10-17 21:17:57,061 epoch 5 - iter 294/1476 - loss 0.04077908 - time (sec): 14.51 - samples/sec: 2363.98 - lr: 0.000032 - momentum: 0.000000
2023-10-17 21:18:04,459 epoch 5 - iter 441/1476 - loss 0.04485968 - time (sec): 21.91 - samples/sec: 2394.21 - lr: 0.000032 - momentum: 0.000000
2023-10-17 21:18:11,546 epoch 5 - iter 588/1476 - loss 0.04489000 - time (sec): 29.00 - samples/sec: 2379.88 - lr: 0.000031 - momentum: 0.000000
2023-10-17 21:18:18,929 epoch 5 - iter 735/1476 - loss 0.04378441 - time (sec): 36.38 - samples/sec: 2359.70 - lr: 0.000031 - momentum: 0.000000
2023-10-17 21:18:26,184 epoch 5 - iter 882/1476 - loss 0.04187352 - time (sec): 43.64 - samples/sec: 2361.29 - lr: 0.000030 - momentum: 0.000000
2023-10-17 21:18:33,052 epoch 5 - iter 1029/1476 - loss 0.04478438 - time (sec): 50.50 - samples/sec: 2331.91 - lr: 0.000029 - momentum: 0.000000
2023-10-17 21:18:40,065 epoch 5 - iter 1176/1476 - loss 0.04667311 - time (sec): 57.52 - samples/sec: 2309.59 - lr: 0.000029 - momentum: 0.000000
2023-10-17 21:18:47,060 epoch 5 - iter 1323/1476 - loss 0.04707730 - time (sec): 64.51 - samples/sec: 2314.55 - lr: 0.000028 - momentum: 0.000000
2023-10-17 21:18:54,289 epoch 5 - iter 1470/1476 - loss 0.04804846 - time (sec): 71.74 - samples/sec: 2312.31 - lr: 0.000028 - momentum: 0.000000
2023-10-17 21:18:54,581 ----------------------------------------------------------------------------------------------------
2023-10-17 21:18:54,581 EPOCH 5 done: loss 0.0479 - lr: 0.000028
2023-10-17 21:19:05,740 DEV : loss 0.1932452768087387 - f1-score (micro avg) 0.8216
2023-10-17 21:19:05,770 ----------------------------------------------------------------------------------------------------
2023-10-17 21:19:13,041 epoch 6 - iter 147/1476 - loss 0.02692959 - time (sec): 7.27 - samples/sec: 2251.36 - lr: 0.000027 - momentum: 0.000000
2023-10-17 21:19:20,095 epoch 6 - iter 294/1476 - loss 0.03039685 - time (sec): 14.32 - samples/sec: 2258.57 - lr: 0.000027 - momentum: 0.000000
2023-10-17 21:19:26,924 epoch 6 - iter 441/1476 - loss 0.02969535 - time (sec): 21.15 - samples/sec: 2240.64 - lr: 0.000026 - momentum: 0.000000
2023-10-17 21:19:34,018 epoch 6 - iter 588/1476 - loss 0.02976128 - time (sec): 28.25 - samples/sec: 2269.04 - lr: 0.000026 - momentum: 0.000000
2023-10-17 21:19:41,263 epoch 6 - iter 735/1476 - loss 0.03299645 - time (sec): 35.49 - samples/sec: 2272.63 - lr: 0.000025 - momentum: 0.000000
2023-10-17 21:19:48,189 epoch 6 - iter 882/1476 - loss 0.03080088 - time (sec): 42.42 - samples/sec: 2277.34 - lr: 0.000024 - momentum: 0.000000
2023-10-17 21:19:55,156 epoch 6 - iter 1029/1476 - loss 0.03111720 - time (sec): 49.39 - samples/sec: 2274.80 - lr: 0.000024 - momentum: 0.000000
2023-10-17 21:20:02,179 epoch 6 - iter 1176/1476 - loss 0.03136825 - time (sec): 56.41 - samples/sec: 2267.49 - lr: 0.000023 - momentum: 0.000000
2023-10-17 21:20:09,961 epoch 6 - iter 1323/1476 - loss 0.03283562 - time (sec): 64.19 - samples/sec: 2315.08 - lr: 0.000023 - momentum: 0.000000
2023-10-17 21:20:17,596 epoch 6 - iter 1470/1476 - loss 0.03154629 - time (sec): 71.82 - samples/sec: 2298.45 - lr: 0.000022 - momentum: 0.000000
2023-10-17 21:20:18,023 ----------------------------------------------------------------------------------------------------
2023-10-17 21:20:18,023 EPOCH 6 done: loss 0.0318 - lr: 0.000022
2023-10-17 21:20:29,225 DEV : loss 0.20342403650283813 - f1-score (micro avg) 0.829
2023-10-17 21:20:29,256 ----------------------------------------------------------------------------------------------------
2023-10-17 21:20:36,413 epoch 7 - iter 147/1476 - loss 0.02891200 - time (sec): 7.16 - samples/sec: 2142.34 - lr: 0.000022 - momentum: 0.000000
2023-10-17 21:20:43,330 epoch 7 - iter 294/1476 - loss 0.02394202 - time (sec): 14.07 - samples/sec: 2229.98 - lr: 0.000021 - momentum: 0.000000
2023-10-17 21:20:50,316 epoch 7 - iter 441/1476 - loss 0.02015639 - time (sec): 21.06 - samples/sec: 2184.29 - lr: 0.000021 - momentum: 0.000000
2023-10-17 21:20:57,474 epoch 7 - iter 588/1476 - loss 0.02007236 - time (sec): 28.22 - samples/sec: 2211.82 - lr: 0.000020 - momentum: 0.000000
2023-10-17 21:21:04,503 epoch 7 - iter 735/1476 - loss 0.02231409 - time (sec): 35.25 - samples/sec: 2232.96 - lr: 0.000019 - momentum: 0.000000
2023-10-17 21:21:11,792 epoch 7 - iter 882/1476 - loss 0.02414592 - time (sec): 42.54 - samples/sec: 2270.14 - lr: 0.000019 - momentum: 0.000000
2023-10-17 21:21:19,804 epoch 7 - iter 1029/1476 - loss 0.02619634 - time (sec): 50.55 - samples/sec: 2320.73 - lr: 0.000018 - momentum: 0.000000
2023-10-17 21:21:27,164 epoch 7 - iter 1176/1476 - loss 0.02513161 - time (sec): 57.91 - samples/sec: 2299.35 - lr: 0.000018 - momentum: 0.000000
2023-10-17 21:21:34,416 epoch 7 - iter 1323/1476 - loss 0.02524479 - time (sec): 65.16 - samples/sec: 2297.30 - lr: 0.000017 - momentum: 0.000000
2023-10-17 21:21:41,691 epoch 7 - iter 1470/1476 - loss 0.02525241 - time (sec): 72.43 - samples/sec: 2292.05 - lr: 0.000017 - momentum: 0.000000
2023-10-17 21:21:41,978 ----------------------------------------------------------------------------------------------------
2023-10-17 21:21:41,978 EPOCH 7 done: loss 0.0252 - lr: 0.000017
2023-10-17 21:21:53,349 DEV : loss 0.2042791247367859 - f1-score (micro avg) 0.8458
2023-10-17 21:21:53,382 saving best model
2023-10-17 21:21:53,867 ----------------------------------------------------------------------------------------------------
2023-10-17 21:22:00,762 epoch 8 - iter 147/1476 - loss 0.00937919 - time (sec): 6.89 - samples/sec: 2283.57 - lr: 0.000016 - momentum: 0.000000
2023-10-17 21:22:07,881 epoch 8 - iter 294/1476 - loss 0.00853811 - time (sec): 14.01 - samples/sec: 2258.90 - lr: 0.000016 - momentum: 0.000000
2023-10-17 21:22:14,881 epoch 8 - iter 441/1476 - loss 0.01070567 - time (sec): 21.01 - samples/sec: 2269.68 - lr: 0.000015 - momentum: 0.000000
2023-10-17 21:22:22,036 epoch 8 - iter 588/1476 - loss 0.01179930 - time (sec): 28.17 - samples/sec: 2246.93 - lr: 0.000014 - momentum: 0.000000
2023-10-17 21:22:30,116 epoch 8 - iter 735/1476 - loss 0.01355360 - time (sec): 36.24 - samples/sec: 2315.08 - lr: 0.000014 - momentum: 0.000000
2023-10-17 21:22:37,569 epoch 8 - iter 882/1476 - loss 0.01288325 - time (sec): 43.70 - samples/sec: 2340.72 - lr: 0.000013 - momentum: 0.000000
2023-10-17 21:22:44,608 epoch 8 - iter 1029/1476 - loss 0.01292629 - time (sec): 50.74 - samples/sec: 2330.05 - lr: 0.000013 - momentum: 0.000000
2023-10-17 21:22:51,814 epoch 8 - iter 1176/1476 - loss 0.01385110 - time (sec): 57.94 - samples/sec: 2329.28 - lr: 0.000012 - momentum: 0.000000
2023-10-17 21:22:58,993 epoch 8 - iter 1323/1476 - loss 0.01415473 - time (sec): 65.12 - samples/sec: 2309.57 - lr: 0.000012 - momentum: 0.000000
2023-10-17 21:23:05,874 epoch 8 - iter 1470/1476 - loss 0.01408906 - time (sec): 72.00 - samples/sec: 2299.77 - lr: 0.000011 - momentum: 0.000000
2023-10-17 21:23:06,189 ----------------------------------------------------------------------------------------------------
2023-10-17 21:23:06,190 EPOCH 8 done: loss 0.0140 - lr: 0.000011
2023-10-17 21:23:17,357 DEV : loss 0.23180881142616272 - f1-score (micro avg) 0.8352
2023-10-17 21:23:17,387 ----------------------------------------------------------------------------------------------------
2023-10-17 21:23:24,663 epoch 9 - iter 147/1476 - loss 0.01659959 - time (sec): 7.27 - samples/sec: 2475.68 - lr: 0.000011 - momentum: 0.000000
2023-10-17 21:23:31,742 epoch 9 - iter 294/1476 - loss 0.01094948 - time (sec): 14.35 - samples/sec: 2413.12 - lr: 0.000010 - momentum: 0.000000
2023-10-17 21:23:38,527 epoch 9 - iter 441/1476 - loss 0.00931768 - time (sec): 21.14 - samples/sec: 2376.05 - lr: 0.000009 - momentum: 0.000000
2023-10-17 21:23:45,851 epoch 9 - iter 588/1476 - loss 0.00991294 - time (sec): 28.46 - samples/sec: 2373.51 - lr: 0.000009 - momentum: 0.000000
2023-10-17 21:23:52,800 epoch 9 - iter 735/1476 - loss 0.01019933 - time (sec): 35.41 - samples/sec: 2348.46 - lr: 0.000008 - momentum: 0.000000
2023-10-17 21:24:00,113 epoch 9 - iter 882/1476 - loss 0.01047522 - time (sec): 42.72 - samples/sec: 2354.51 - lr: 0.000008 - momentum: 0.000000
2023-10-17 21:24:07,361 epoch 9 - iter 1029/1476 - loss 0.01030206 - time (sec): 49.97 - samples/sec: 2347.60 - lr: 0.000007 - momentum: 0.000000
2023-10-17 21:24:14,651 epoch 9 - iter 1176/1476 - loss 0.01068040 - time (sec): 57.26 - samples/sec: 2353.18 - lr: 0.000007 - momentum: 0.000000
2023-10-17 21:24:21,700 epoch 9 - iter 1323/1476 - loss 0.01029783 - time (sec): 64.31 - samples/sec: 2344.88 - lr: 0.000006 - momentum: 0.000000
2023-10-17 21:24:28,634 epoch 9 - iter 1470/1476 - loss 0.00979050 - time (sec): 71.25 - samples/sec: 2328.95 - lr: 0.000006 - momentum: 0.000000
2023-10-17 21:24:28,905 ----------------------------------------------------------------------------------------------------
2023-10-17 21:24:28,905 EPOCH 9 done: loss 0.0098 - lr: 0.000006
2023-10-17 21:24:40,143 DEV : loss 0.2325468510389328 - f1-score (micro avg) 0.8472
2023-10-17 21:24:40,174 saving best model
2023-10-17 21:24:40,670 ----------------------------------------------------------------------------------------------------
2023-10-17 21:24:47,921 epoch 10 - iter 147/1476 - loss 0.00447119 - time (sec): 7.25 - samples/sec: 2284.31 - lr: 0.000005 - momentum: 0.000000
2023-10-17 21:24:55,320 epoch 10 - iter 294/1476 - loss 0.00607937 - time (sec): 14.64 - samples/sec: 2399.71 - lr: 0.000004 - momentum: 0.000000
2023-10-17 21:25:02,482 epoch 10 - iter 441/1476 - loss 0.00594049 - time (sec): 21.81 - samples/sec: 2332.13 - lr: 0.000004 - momentum: 0.000000
2023-10-17 21:25:09,895 epoch 10 - iter 588/1476 - loss 0.00571948 - time (sec): 29.22 - samples/sec: 2316.57 - lr: 0.000003 - momentum: 0.000000
2023-10-17 21:25:17,081 epoch 10 - iter 735/1476 - loss 0.00519373 - time (sec): 36.41 - samples/sec: 2296.15 - lr: 0.000003 - momentum: 0.000000
2023-10-17 21:25:23,922 epoch 10 - iter 882/1476 - loss 0.00678593 - time (sec): 43.25 - samples/sec: 2294.83 - lr: 0.000002 - momentum: 0.000000
2023-10-17 21:25:31,170 epoch 10 - iter 1029/1476 - loss 0.00637890 - time (sec): 50.49 - samples/sec: 2285.07 - lr: 0.000002 - momentum: 0.000000
2023-10-17 21:25:38,140 epoch 10 - iter 1176/1476 - loss 0.00583236 - time (sec): 57.47 - samples/sec: 2296.51 - lr: 0.000001 - momentum: 0.000000
2023-10-17 21:25:45,142 epoch 10 - iter 1323/1476 - loss 0.00563599 - time (sec): 64.47 - samples/sec: 2301.07 - lr: 0.000001 - momentum: 0.000000
2023-10-17 21:25:52,319 epoch 10 - iter 1470/1476 - loss 0.00529100 - time (sec): 71.64 - samples/sec: 2315.70 - lr: 0.000000 - momentum: 0.000000
2023-10-17 21:25:52,584 ----------------------------------------------------------------------------------------------------
2023-10-17 21:25:52,584 EPOCH 10 done: loss 0.0053 - lr: 0.000000
2023-10-17 21:26:03,720 DEV : loss 0.2309638112783432 - f1-score (micro avg) 0.8473
2023-10-17 21:26:03,750 saving best model
2023-10-17 21:26:04,626 ----------------------------------------------------------------------------------------------------
2023-10-17 21:26:04,627 Loading model from best epoch ...
2023-10-17 21:26:05,984 SequenceTagger predicts: Dictionary with 21 tags: O, S-loc, B-loc, E-loc, I-loc, S-pers, B-pers, E-pers, I-pers, S-org, B-org, E-org, I-org, S-time, B-time, E-time, I-time, S-prod, B-prod, E-prod, I-prod
2023-10-17 21:26:12,598
Results:
- F-score (micro) 0.805
- F-score (macro) 0.7103
- Accuracy 0.6948
By class:
precision recall f1-score support
loc 0.8445 0.8800 0.8619 858
pers 0.7666 0.8194 0.7921 537
org 0.6532 0.6136 0.6328 132
prod 0.6885 0.6885 0.6885 61
time 0.5312 0.6296 0.5763 54
micro avg 0.7874 0.8234 0.8050 1642
macro avg 0.6968 0.7262 0.7103 1642
weighted avg 0.7875 0.8234 0.8048 1642
2023-10-17 21:26:12,598 ----------------------------------------------------------------------------------------------------
|