File size: 23,826 Bytes
a903ec3 |
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 |
2023-10-17 11:21:40,050 ----------------------------------------------------------------------------------------------------
2023-10-17 11:21:40,051 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=13, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-17 11:21:40,051 ----------------------------------------------------------------------------------------------------
2023-10-17 11:21:40,051 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
2023-10-17 11:21:40,051 ----------------------------------------------------------------------------------------------------
2023-10-17 11:21:40,052 Train: 7936 sentences
2023-10-17 11:21:40,052 (train_with_dev=False, train_with_test=False)
2023-10-17 11:21:40,052 ----------------------------------------------------------------------------------------------------
2023-10-17 11:21:40,052 Training Params:
2023-10-17 11:21:40,052 - learning_rate: "5e-05"
2023-10-17 11:21:40,052 - mini_batch_size: "8"
2023-10-17 11:21:40,052 - max_epochs: "10"
2023-10-17 11:21:40,052 - shuffle: "True"
2023-10-17 11:21:40,052 ----------------------------------------------------------------------------------------------------
2023-10-17 11:21:40,052 Plugins:
2023-10-17 11:21:40,052 - TensorboardLogger
2023-10-17 11:21:40,052 - LinearScheduler | warmup_fraction: '0.1'
2023-10-17 11:21:40,052 ----------------------------------------------------------------------------------------------------
2023-10-17 11:21:40,052 Final evaluation on model from best epoch (best-model.pt)
2023-10-17 11:21:40,052 - metric: "('micro avg', 'f1-score')"
2023-10-17 11:21:40,052 ----------------------------------------------------------------------------------------------------
2023-10-17 11:21:40,052 Computation:
2023-10-17 11:21:40,052 - compute on device: cuda:0
2023-10-17 11:21:40,052 - embedding storage: none
2023-10-17 11:21:40,052 ----------------------------------------------------------------------------------------------------
2023-10-17 11:21:40,052 Model training base path: "hmbench-icdar/fr-hmteams/teams-base-historic-multilingual-discriminator-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-1"
2023-10-17 11:21:40,052 ----------------------------------------------------------------------------------------------------
2023-10-17 11:21:40,052 ----------------------------------------------------------------------------------------------------
2023-10-17 11:21:40,052 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-17 11:21:45,743 epoch 1 - iter 99/992 - loss 2.00752470 - time (sec): 5.69 - samples/sec: 2843.79 - lr: 0.000005 - momentum: 0.000000
2023-10-17 11:21:51,606 epoch 1 - iter 198/992 - loss 1.16110239 - time (sec): 11.55 - samples/sec: 2877.04 - lr: 0.000010 - momentum: 0.000000
2023-10-17 11:21:57,730 epoch 1 - iter 297/992 - loss 0.86009292 - time (sec): 17.68 - samples/sec: 2807.08 - lr: 0.000015 - momentum: 0.000000
2023-10-17 11:22:03,344 epoch 1 - iter 396/992 - loss 0.70056867 - time (sec): 23.29 - samples/sec: 2807.69 - lr: 0.000020 - momentum: 0.000000
2023-10-17 11:22:09,204 epoch 1 - iter 495/992 - loss 0.59290102 - time (sec): 29.15 - samples/sec: 2807.40 - lr: 0.000025 - momentum: 0.000000
2023-10-17 11:22:15,182 epoch 1 - iter 594/992 - loss 0.51678549 - time (sec): 35.13 - samples/sec: 2808.16 - lr: 0.000030 - momentum: 0.000000
2023-10-17 11:22:20,959 epoch 1 - iter 693/992 - loss 0.46411643 - time (sec): 40.91 - samples/sec: 2806.30 - lr: 0.000035 - momentum: 0.000000
2023-10-17 11:22:26,973 epoch 1 - iter 792/992 - loss 0.42507693 - time (sec): 46.92 - samples/sec: 2798.32 - lr: 0.000040 - momentum: 0.000000
2023-10-17 11:22:32,888 epoch 1 - iter 891/992 - loss 0.39307524 - time (sec): 52.83 - samples/sec: 2786.92 - lr: 0.000045 - momentum: 0.000000
2023-10-17 11:22:38,704 epoch 1 - iter 990/992 - loss 0.36782772 - time (sec): 58.65 - samples/sec: 2790.32 - lr: 0.000050 - momentum: 0.000000
2023-10-17 11:22:38,823 ----------------------------------------------------------------------------------------------------
2023-10-17 11:22:38,823 EPOCH 1 done: loss 0.3674 - lr: 0.000050
2023-10-17 11:22:41,920 DEV : loss 0.09267440438270569 - f1-score (micro avg) 0.7287
2023-10-17 11:22:41,943 saving best model
2023-10-17 11:22:42,310 ----------------------------------------------------------------------------------------------------
2023-10-17 11:22:48,213 epoch 2 - iter 99/992 - loss 0.11596098 - time (sec): 5.90 - samples/sec: 2864.40 - lr: 0.000049 - momentum: 0.000000
2023-10-17 11:22:54,188 epoch 2 - iter 198/992 - loss 0.11376424 - time (sec): 11.88 - samples/sec: 2788.50 - lr: 0.000049 - momentum: 0.000000
2023-10-17 11:23:00,295 epoch 2 - iter 297/992 - loss 0.11209663 - time (sec): 17.98 - samples/sec: 2759.41 - lr: 0.000048 - momentum: 0.000000
2023-10-17 11:23:06,316 epoch 2 - iter 396/992 - loss 0.11161254 - time (sec): 24.00 - samples/sec: 2767.95 - lr: 0.000048 - momentum: 0.000000
2023-10-17 11:23:12,200 epoch 2 - iter 495/992 - loss 0.10962236 - time (sec): 29.89 - samples/sec: 2762.55 - lr: 0.000047 - momentum: 0.000000
2023-10-17 11:23:18,445 epoch 2 - iter 594/992 - loss 0.10658082 - time (sec): 36.13 - samples/sec: 2749.28 - lr: 0.000047 - momentum: 0.000000
2023-10-17 11:23:24,452 epoch 2 - iter 693/992 - loss 0.10590489 - time (sec): 42.14 - samples/sec: 2752.00 - lr: 0.000046 - momentum: 0.000000
2023-10-17 11:23:30,157 epoch 2 - iter 792/992 - loss 0.10554854 - time (sec): 47.85 - samples/sec: 2752.16 - lr: 0.000046 - momentum: 0.000000
2023-10-17 11:23:35,744 epoch 2 - iter 891/992 - loss 0.10485385 - time (sec): 53.43 - samples/sec: 2756.15 - lr: 0.000045 - momentum: 0.000000
2023-10-17 11:23:41,651 epoch 2 - iter 990/992 - loss 0.10541228 - time (sec): 59.34 - samples/sec: 2757.92 - lr: 0.000044 - momentum: 0.000000
2023-10-17 11:23:41,762 ----------------------------------------------------------------------------------------------------
2023-10-17 11:23:41,763 EPOCH 2 done: loss 0.1053 - lr: 0.000044
2023-10-17 11:23:45,546 DEV : loss 0.07957779616117477 - f1-score (micro avg) 0.7488
2023-10-17 11:23:45,567 saving best model
2023-10-17 11:23:46,062 ----------------------------------------------------------------------------------------------------
2023-10-17 11:23:51,998 epoch 3 - iter 99/992 - loss 0.07399883 - time (sec): 5.93 - samples/sec: 2792.44 - lr: 0.000044 - momentum: 0.000000
2023-10-17 11:23:57,894 epoch 3 - iter 198/992 - loss 0.07782338 - time (sec): 11.83 - samples/sec: 2777.78 - lr: 0.000043 - momentum: 0.000000
2023-10-17 11:24:03,807 epoch 3 - iter 297/992 - loss 0.07965927 - time (sec): 17.74 - samples/sec: 2795.99 - lr: 0.000043 - momentum: 0.000000
2023-10-17 11:24:09,711 epoch 3 - iter 396/992 - loss 0.07595699 - time (sec): 23.65 - samples/sec: 2808.81 - lr: 0.000042 - momentum: 0.000000
2023-10-17 11:24:15,396 epoch 3 - iter 495/992 - loss 0.07484677 - time (sec): 29.33 - samples/sec: 2819.60 - lr: 0.000042 - momentum: 0.000000
2023-10-17 11:24:21,231 epoch 3 - iter 594/992 - loss 0.07449912 - time (sec): 35.17 - samples/sec: 2800.02 - lr: 0.000041 - momentum: 0.000000
2023-10-17 11:24:27,126 epoch 3 - iter 693/992 - loss 0.07526423 - time (sec): 41.06 - samples/sec: 2795.76 - lr: 0.000041 - momentum: 0.000000
2023-10-17 11:24:33,067 epoch 3 - iter 792/992 - loss 0.07578093 - time (sec): 47.00 - samples/sec: 2790.69 - lr: 0.000040 - momentum: 0.000000
2023-10-17 11:24:39,214 epoch 3 - iter 891/992 - loss 0.07554286 - time (sec): 53.15 - samples/sec: 2782.60 - lr: 0.000039 - momentum: 0.000000
2023-10-17 11:24:45,213 epoch 3 - iter 990/992 - loss 0.07580919 - time (sec): 59.15 - samples/sec: 2766.04 - lr: 0.000039 - momentum: 0.000000
2023-10-17 11:24:45,340 ----------------------------------------------------------------------------------------------------
2023-10-17 11:24:45,340 EPOCH 3 done: loss 0.0761 - lr: 0.000039
2023-10-17 11:24:48,746 DEV : loss 0.09182097762823105 - f1-score (micro avg) 0.7489
2023-10-17 11:24:48,769 saving best model
2023-10-17 11:24:49,263 ----------------------------------------------------------------------------------------------------
2023-10-17 11:24:55,170 epoch 4 - iter 99/992 - loss 0.05468852 - time (sec): 5.90 - samples/sec: 2780.10 - lr: 0.000038 - momentum: 0.000000
2023-10-17 11:25:01,487 epoch 4 - iter 198/992 - loss 0.05293064 - time (sec): 12.22 - samples/sec: 2794.12 - lr: 0.000038 - momentum: 0.000000
2023-10-17 11:25:07,607 epoch 4 - iter 297/992 - loss 0.05364317 - time (sec): 18.34 - samples/sec: 2776.94 - lr: 0.000037 - momentum: 0.000000
2023-10-17 11:25:13,550 epoch 4 - iter 396/992 - loss 0.05402008 - time (sec): 24.28 - samples/sec: 2777.04 - lr: 0.000037 - momentum: 0.000000
2023-10-17 11:25:19,296 epoch 4 - iter 495/992 - loss 0.05392577 - time (sec): 30.03 - samples/sec: 2786.92 - lr: 0.000036 - momentum: 0.000000
2023-10-17 11:25:24,951 epoch 4 - iter 594/992 - loss 0.05443890 - time (sec): 35.69 - samples/sec: 2782.91 - lr: 0.000036 - momentum: 0.000000
2023-10-17 11:25:30,557 epoch 4 - iter 693/992 - loss 0.05350202 - time (sec): 41.29 - samples/sec: 2783.23 - lr: 0.000035 - momentum: 0.000000
2023-10-17 11:25:36,431 epoch 4 - iter 792/992 - loss 0.05357310 - time (sec): 47.17 - samples/sec: 2777.82 - lr: 0.000034 - momentum: 0.000000
2023-10-17 11:25:42,388 epoch 4 - iter 891/992 - loss 0.05445889 - time (sec): 53.12 - samples/sec: 2780.60 - lr: 0.000034 - momentum: 0.000000
2023-10-17 11:25:48,076 epoch 4 - iter 990/992 - loss 0.05431231 - time (sec): 58.81 - samples/sec: 2784.57 - lr: 0.000033 - momentum: 0.000000
2023-10-17 11:25:48,189 ----------------------------------------------------------------------------------------------------
2023-10-17 11:25:48,190 EPOCH 4 done: loss 0.0544 - lr: 0.000033
2023-10-17 11:25:51,644 DEV : loss 0.1407020390033722 - f1-score (micro avg) 0.7598
2023-10-17 11:25:51,666 saving best model
2023-10-17 11:25:52,143 ----------------------------------------------------------------------------------------------------
2023-10-17 11:25:58,088 epoch 5 - iter 99/992 - loss 0.04345102 - time (sec): 5.94 - samples/sec: 2737.20 - lr: 0.000033 - momentum: 0.000000
2023-10-17 11:26:04,229 epoch 5 - iter 198/992 - loss 0.04013191 - time (sec): 12.08 - samples/sec: 2773.50 - lr: 0.000032 - momentum: 0.000000
2023-10-17 11:26:10,153 epoch 5 - iter 297/992 - loss 0.04013654 - time (sec): 18.01 - samples/sec: 2788.59 - lr: 0.000032 - momentum: 0.000000
2023-10-17 11:26:16,359 epoch 5 - iter 396/992 - loss 0.04153776 - time (sec): 24.21 - samples/sec: 2797.16 - lr: 0.000031 - momentum: 0.000000
2023-10-17 11:26:22,273 epoch 5 - iter 495/992 - loss 0.04178549 - time (sec): 30.13 - samples/sec: 2793.12 - lr: 0.000031 - momentum: 0.000000
2023-10-17 11:26:27,904 epoch 5 - iter 594/992 - loss 0.04313324 - time (sec): 35.76 - samples/sec: 2796.33 - lr: 0.000030 - momentum: 0.000000
2023-10-17 11:26:34,115 epoch 5 - iter 693/992 - loss 0.04394145 - time (sec): 41.97 - samples/sec: 2773.35 - lr: 0.000029 - momentum: 0.000000
2023-10-17 11:26:40,034 epoch 5 - iter 792/992 - loss 0.04428695 - time (sec): 47.89 - samples/sec: 2761.16 - lr: 0.000029 - momentum: 0.000000
2023-10-17 11:26:45,880 epoch 5 - iter 891/992 - loss 0.04347387 - time (sec): 53.73 - samples/sec: 2758.23 - lr: 0.000028 - momentum: 0.000000
2023-10-17 11:26:51,571 epoch 5 - iter 990/992 - loss 0.04302101 - time (sec): 59.42 - samples/sec: 2753.59 - lr: 0.000028 - momentum: 0.000000
2023-10-17 11:26:51,696 ----------------------------------------------------------------------------------------------------
2023-10-17 11:26:51,696 EPOCH 5 done: loss 0.0429 - lr: 0.000028
2023-10-17 11:26:55,092 DEV : loss 0.1663055419921875 - f1-score (micro avg) 0.7778
2023-10-17 11:26:55,113 saving best model
2023-10-17 11:26:55,582 ----------------------------------------------------------------------------------------------------
2023-10-17 11:27:01,664 epoch 6 - iter 99/992 - loss 0.03345702 - time (sec): 6.08 - samples/sec: 2705.79 - lr: 0.000027 - momentum: 0.000000
2023-10-17 11:27:07,592 epoch 6 - iter 198/992 - loss 0.03258281 - time (sec): 12.01 - samples/sec: 2720.35 - lr: 0.000027 - momentum: 0.000000
2023-10-17 11:27:13,856 epoch 6 - iter 297/992 - loss 0.03031919 - time (sec): 18.27 - samples/sec: 2750.49 - lr: 0.000026 - momentum: 0.000000
2023-10-17 11:27:19,774 epoch 6 - iter 396/992 - loss 0.03089649 - time (sec): 24.19 - samples/sec: 2754.91 - lr: 0.000026 - momentum: 0.000000
2023-10-17 11:27:25,776 epoch 6 - iter 495/992 - loss 0.02999409 - time (sec): 30.19 - samples/sec: 2762.61 - lr: 0.000025 - momentum: 0.000000
2023-10-17 11:27:31,638 epoch 6 - iter 594/992 - loss 0.03070964 - time (sec): 36.05 - samples/sec: 2765.56 - lr: 0.000024 - momentum: 0.000000
2023-10-17 11:27:37,231 epoch 6 - iter 693/992 - loss 0.03109293 - time (sec): 41.65 - samples/sec: 2769.11 - lr: 0.000024 - momentum: 0.000000
2023-10-17 11:27:42,876 epoch 6 - iter 792/992 - loss 0.03056361 - time (sec): 47.29 - samples/sec: 2773.64 - lr: 0.000023 - momentum: 0.000000
2023-10-17 11:27:48,608 epoch 6 - iter 891/992 - loss 0.03117794 - time (sec): 53.02 - samples/sec: 2778.13 - lr: 0.000023 - momentum: 0.000000
2023-10-17 11:27:54,391 epoch 6 - iter 990/992 - loss 0.03126223 - time (sec): 58.80 - samples/sec: 2781.63 - lr: 0.000022 - momentum: 0.000000
2023-10-17 11:27:54,511 ----------------------------------------------------------------------------------------------------
2023-10-17 11:27:54,511 EPOCH 6 done: loss 0.0312 - lr: 0.000022
2023-10-17 11:27:59,227 DEV : loss 0.1633480340242386 - f1-score (micro avg) 0.7584
2023-10-17 11:27:59,263 ----------------------------------------------------------------------------------------------------
2023-10-17 11:28:05,215 epoch 7 - iter 99/992 - loss 0.01789778 - time (sec): 5.95 - samples/sec: 2739.15 - lr: 0.000022 - momentum: 0.000000
2023-10-17 11:28:11,218 epoch 7 - iter 198/992 - loss 0.01724841 - time (sec): 11.95 - samples/sec: 2745.60 - lr: 0.000021 - momentum: 0.000000
2023-10-17 11:28:17,621 epoch 7 - iter 297/992 - loss 0.01961745 - time (sec): 18.36 - samples/sec: 2700.02 - lr: 0.000021 - momentum: 0.000000
2023-10-17 11:28:23,473 epoch 7 - iter 396/992 - loss 0.01992616 - time (sec): 24.21 - samples/sec: 2712.23 - lr: 0.000020 - momentum: 0.000000
2023-10-17 11:28:29,388 epoch 7 - iter 495/992 - loss 0.02094923 - time (sec): 30.12 - samples/sec: 2720.79 - lr: 0.000019 - momentum: 0.000000
2023-10-17 11:28:35,340 epoch 7 - iter 594/992 - loss 0.01985550 - time (sec): 36.08 - samples/sec: 2728.57 - lr: 0.000019 - momentum: 0.000000
2023-10-17 11:28:41,393 epoch 7 - iter 693/992 - loss 0.02067558 - time (sec): 42.13 - samples/sec: 2724.08 - lr: 0.000018 - momentum: 0.000000
2023-10-17 11:28:47,206 epoch 7 - iter 792/992 - loss 0.02093216 - time (sec): 47.94 - samples/sec: 2718.22 - lr: 0.000018 - momentum: 0.000000
2023-10-17 11:28:53,073 epoch 7 - iter 891/992 - loss 0.02197206 - time (sec): 53.81 - samples/sec: 2739.75 - lr: 0.000017 - momentum: 0.000000
2023-10-17 11:28:58,840 epoch 7 - iter 990/992 - loss 0.02197388 - time (sec): 59.58 - samples/sec: 2747.71 - lr: 0.000017 - momentum: 0.000000
2023-10-17 11:28:58,961 ----------------------------------------------------------------------------------------------------
2023-10-17 11:28:58,961 EPOCH 7 done: loss 0.0219 - lr: 0.000017
2023-10-17 11:29:02,566 DEV : loss 0.1898750215768814 - f1-score (micro avg) 0.7665
2023-10-17 11:29:02,594 ----------------------------------------------------------------------------------------------------
2023-10-17 11:29:08,347 epoch 8 - iter 99/992 - loss 0.00858606 - time (sec): 5.75 - samples/sec: 2846.11 - lr: 0.000016 - momentum: 0.000000
2023-10-17 11:29:14,064 epoch 8 - iter 198/992 - loss 0.01171469 - time (sec): 11.47 - samples/sec: 2824.80 - lr: 0.000016 - momentum: 0.000000
2023-10-17 11:29:20,281 epoch 8 - iter 297/992 - loss 0.01211733 - time (sec): 17.69 - samples/sec: 2828.38 - lr: 0.000015 - momentum: 0.000000
2023-10-17 11:29:26,152 epoch 8 - iter 396/992 - loss 0.01169790 - time (sec): 23.56 - samples/sec: 2805.84 - lr: 0.000014 - momentum: 0.000000
2023-10-17 11:29:32,084 epoch 8 - iter 495/992 - loss 0.01211186 - time (sec): 29.49 - samples/sec: 2821.28 - lr: 0.000014 - momentum: 0.000000
2023-10-17 11:29:38,180 epoch 8 - iter 594/992 - loss 0.01413340 - time (sec): 35.58 - samples/sec: 2807.68 - lr: 0.000013 - momentum: 0.000000
2023-10-17 11:29:43,898 epoch 8 - iter 693/992 - loss 0.01436134 - time (sec): 41.30 - samples/sec: 2798.96 - lr: 0.000013 - momentum: 0.000000
2023-10-17 11:29:49,496 epoch 8 - iter 792/992 - loss 0.01496934 - time (sec): 46.90 - samples/sec: 2784.54 - lr: 0.000012 - momentum: 0.000000
2023-10-17 11:29:55,299 epoch 8 - iter 891/992 - loss 0.01507510 - time (sec): 52.70 - samples/sec: 2793.17 - lr: 0.000012 - momentum: 0.000000
2023-10-17 11:30:01,345 epoch 8 - iter 990/992 - loss 0.01559016 - time (sec): 58.75 - samples/sec: 2785.33 - lr: 0.000011 - momentum: 0.000000
2023-10-17 11:30:01,478 ----------------------------------------------------------------------------------------------------
2023-10-17 11:30:01,478 EPOCH 8 done: loss 0.0156 - lr: 0.000011
2023-10-17 11:30:05,096 DEV : loss 0.21736501157283783 - f1-score (micro avg) 0.7736
2023-10-17 11:30:05,127 ----------------------------------------------------------------------------------------------------
2023-10-17 11:30:11,217 epoch 9 - iter 99/992 - loss 0.01189736 - time (sec): 6.09 - samples/sec: 2601.72 - lr: 0.000011 - momentum: 0.000000
2023-10-17 11:30:17,689 epoch 9 - iter 198/992 - loss 0.01120784 - time (sec): 12.56 - samples/sec: 2643.77 - lr: 0.000010 - momentum: 0.000000
2023-10-17 11:30:23,899 epoch 9 - iter 297/992 - loss 0.01112974 - time (sec): 18.77 - samples/sec: 2685.89 - lr: 0.000009 - momentum: 0.000000
2023-10-17 11:30:29,897 epoch 9 - iter 396/992 - loss 0.00969053 - time (sec): 24.77 - samples/sec: 2692.32 - lr: 0.000009 - momentum: 0.000000
2023-10-17 11:30:35,679 epoch 9 - iter 495/992 - loss 0.00960030 - time (sec): 30.55 - samples/sec: 2702.51 - lr: 0.000008 - momentum: 0.000000
2023-10-17 11:30:41,692 epoch 9 - iter 594/992 - loss 0.00957094 - time (sec): 36.56 - samples/sec: 2700.06 - lr: 0.000008 - momentum: 0.000000
2023-10-17 11:30:47,567 epoch 9 - iter 693/992 - loss 0.00941102 - time (sec): 42.44 - samples/sec: 2704.48 - lr: 0.000007 - momentum: 0.000000
2023-10-17 11:30:53,670 epoch 9 - iter 792/992 - loss 0.01010825 - time (sec): 48.54 - samples/sec: 2709.90 - lr: 0.000007 - momentum: 0.000000
2023-10-17 11:30:59,631 epoch 9 - iter 891/992 - loss 0.01035941 - time (sec): 54.50 - samples/sec: 2710.26 - lr: 0.000006 - momentum: 0.000000
2023-10-17 11:31:05,562 epoch 9 - iter 990/992 - loss 0.01067263 - time (sec): 60.43 - samples/sec: 2708.56 - lr: 0.000006 - momentum: 0.000000
2023-10-17 11:31:05,685 ----------------------------------------------------------------------------------------------------
2023-10-17 11:31:05,685 EPOCH 9 done: loss 0.0107 - lr: 0.000006
2023-10-17 11:31:09,296 DEV : loss 0.2289990335702896 - f1-score (micro avg) 0.7598
2023-10-17 11:31:09,320 ----------------------------------------------------------------------------------------------------
2023-10-17 11:31:15,610 epoch 10 - iter 99/992 - loss 0.00433310 - time (sec): 6.29 - samples/sec: 2625.70 - lr: 0.000005 - momentum: 0.000000
2023-10-17 11:31:21,873 epoch 10 - iter 198/992 - loss 0.00533994 - time (sec): 12.55 - samples/sec: 2561.60 - lr: 0.000004 - momentum: 0.000000
2023-10-17 11:31:28,202 epoch 10 - iter 297/992 - loss 0.00637622 - time (sec): 18.88 - samples/sec: 2575.30 - lr: 0.000004 - momentum: 0.000000
2023-10-17 11:31:34,426 epoch 10 - iter 396/992 - loss 0.00717457 - time (sec): 25.10 - samples/sec: 2581.66 - lr: 0.000003 - momentum: 0.000000
2023-10-17 11:31:40,384 epoch 10 - iter 495/992 - loss 0.00716627 - time (sec): 31.06 - samples/sec: 2593.97 - lr: 0.000003 - momentum: 0.000000
2023-10-17 11:31:46,392 epoch 10 - iter 594/992 - loss 0.00678007 - time (sec): 37.07 - samples/sec: 2628.76 - lr: 0.000002 - momentum: 0.000000
2023-10-17 11:31:52,420 epoch 10 - iter 693/992 - loss 0.00705269 - time (sec): 43.10 - samples/sec: 2659.76 - lr: 0.000002 - momentum: 0.000000
2023-10-17 11:31:58,173 epoch 10 - iter 792/992 - loss 0.00695957 - time (sec): 48.85 - samples/sec: 2683.39 - lr: 0.000001 - momentum: 0.000000
2023-10-17 11:32:04,222 epoch 10 - iter 891/992 - loss 0.00746367 - time (sec): 54.90 - samples/sec: 2676.79 - lr: 0.000001 - momentum: 0.000000
2023-10-17 11:32:10,346 epoch 10 - iter 990/992 - loss 0.00797105 - time (sec): 61.02 - samples/sec: 2682.68 - lr: 0.000000 - momentum: 0.000000
2023-10-17 11:32:10,450 ----------------------------------------------------------------------------------------------------
2023-10-17 11:32:10,450 EPOCH 10 done: loss 0.0080 - lr: 0.000000
2023-10-17 11:32:15,038 DEV : loss 0.23297961056232452 - f1-score (micro avg) 0.7648
2023-10-17 11:32:15,550 ----------------------------------------------------------------------------------------------------
2023-10-17 11:32:15,552 Loading model from best epoch ...
2023-10-17 11:32:17,148 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
2023-10-17 11:32:21,019
Results:
- F-score (micro) 0.7587
- F-score (macro) 0.6925
- Accuracy 0.6447
By class:
precision recall f1-score support
LOC 0.8560 0.7893 0.8213 655
PER 0.6335 0.7982 0.7063 223
ORG 0.5565 0.5433 0.5498 127
micro avg 0.7572 0.7602 0.7587 1005
macro avg 0.6820 0.7103 0.6925 1005
weighted avg 0.7687 0.7602 0.7615 1005
2023-10-17 11:32:21,020 ----------------------------------------------------------------------------------------------------
|