File size: 23,740 Bytes
9546cce |
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 |
2023-10-17 17:28:02,197 ----------------------------------------------------------------------------------------------------
2023-10-17 17:28:02,198 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 17:28:02,198 ----------------------------------------------------------------------------------------------------
2023-10-17 17:28:02,199 MultiCorpus: 5777 train + 722 dev + 723 test sentences
- NER_ICDAR_EUROPEANA Corpus: 5777 train + 722 dev + 723 test sentences - /root/.flair/datasets/ner_icdar_europeana/nl
2023-10-17 17:28:02,199 ----------------------------------------------------------------------------------------------------
2023-10-17 17:28:02,199 Train: 5777 sentences
2023-10-17 17:28:02,199 (train_with_dev=False, train_with_test=False)
2023-10-17 17:28:02,199 ----------------------------------------------------------------------------------------------------
2023-10-17 17:28:02,199 Training Params:
2023-10-17 17:28:02,199 - learning_rate: "5e-05"
2023-10-17 17:28:02,199 - mini_batch_size: "8"
2023-10-17 17:28:02,199 - max_epochs: "10"
2023-10-17 17:28:02,199 - shuffle: "True"
2023-10-17 17:28:02,199 ----------------------------------------------------------------------------------------------------
2023-10-17 17:28:02,199 Plugins:
2023-10-17 17:28:02,199 - TensorboardLogger
2023-10-17 17:28:02,199 - LinearScheduler | warmup_fraction: '0.1'
2023-10-17 17:28:02,199 ----------------------------------------------------------------------------------------------------
2023-10-17 17:28:02,199 Final evaluation on model from best epoch (best-model.pt)
2023-10-17 17:28:02,199 - metric: "('micro avg', 'f1-score')"
2023-10-17 17:28:02,199 ----------------------------------------------------------------------------------------------------
2023-10-17 17:28:02,199 Computation:
2023-10-17 17:28:02,199 - compute on device: cuda:0
2023-10-17 17:28:02,199 - embedding storage: none
2023-10-17 17:28:02,199 ----------------------------------------------------------------------------------------------------
2023-10-17 17:28:02,199 Model training base path: "hmbench-icdar/nl-hmteams/teams-base-historic-multilingual-discriminator-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-3"
2023-10-17 17:28:02,199 ----------------------------------------------------------------------------------------------------
2023-10-17 17:28:02,199 ----------------------------------------------------------------------------------------------------
2023-10-17 17:28:02,199 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-17 17:28:07,437 epoch 1 - iter 72/723 - loss 2.62229683 - time (sec): 5.24 - samples/sec: 3280.94 - lr: 0.000005 - momentum: 0.000000
2023-10-17 17:28:12,712 epoch 1 - iter 144/723 - loss 1.53564932 - time (sec): 10.51 - samples/sec: 3238.36 - lr: 0.000010 - momentum: 0.000000
2023-10-17 17:28:17,693 epoch 1 - iter 216/723 - loss 1.07363928 - time (sec): 15.49 - samples/sec: 3338.32 - lr: 0.000015 - momentum: 0.000000
2023-10-17 17:28:23,105 epoch 1 - iter 288/723 - loss 0.84654223 - time (sec): 20.90 - samples/sec: 3324.64 - lr: 0.000020 - momentum: 0.000000
2023-10-17 17:28:28,670 epoch 1 - iter 360/723 - loss 0.69584770 - time (sec): 26.47 - samples/sec: 3332.68 - lr: 0.000025 - momentum: 0.000000
2023-10-17 17:28:33,963 epoch 1 - iter 432/723 - loss 0.59690308 - time (sec): 31.76 - samples/sec: 3353.05 - lr: 0.000030 - momentum: 0.000000
2023-10-17 17:28:39,179 epoch 1 - iter 504/723 - loss 0.52741417 - time (sec): 36.98 - samples/sec: 3347.54 - lr: 0.000035 - momentum: 0.000000
2023-10-17 17:28:44,472 epoch 1 - iter 576/723 - loss 0.47603842 - time (sec): 42.27 - samples/sec: 3345.25 - lr: 0.000040 - momentum: 0.000000
2023-10-17 17:28:49,270 epoch 1 - iter 648/723 - loss 0.44069521 - time (sec): 47.07 - samples/sec: 3335.95 - lr: 0.000045 - momentum: 0.000000
2023-10-17 17:28:54,524 epoch 1 - iter 720/723 - loss 0.40554441 - time (sec): 52.32 - samples/sec: 3352.88 - lr: 0.000050 - momentum: 0.000000
2023-10-17 17:28:54,815 ----------------------------------------------------------------------------------------------------
2023-10-17 17:28:54,815 EPOCH 1 done: loss 0.4040 - lr: 0.000050
2023-10-17 17:28:58,191 DEV : loss 0.08499432355165482 - f1-score (micro avg) 0.8175
2023-10-17 17:28:58,207 saving best model
2023-10-17 17:28:58,558 ----------------------------------------------------------------------------------------------------
2023-10-17 17:29:03,714 epoch 2 - iter 72/723 - loss 0.08166951 - time (sec): 5.16 - samples/sec: 3366.35 - lr: 0.000049 - momentum: 0.000000
2023-10-17 17:29:09,042 epoch 2 - iter 144/723 - loss 0.09238710 - time (sec): 10.48 - samples/sec: 3318.53 - lr: 0.000049 - momentum: 0.000000
2023-10-17 17:29:14,060 epoch 2 - iter 216/723 - loss 0.09656663 - time (sec): 15.50 - samples/sec: 3340.45 - lr: 0.000048 - momentum: 0.000000
2023-10-17 17:29:19,239 epoch 2 - iter 288/723 - loss 0.09515758 - time (sec): 20.68 - samples/sec: 3340.74 - lr: 0.000048 - momentum: 0.000000
2023-10-17 17:29:24,949 epoch 2 - iter 360/723 - loss 0.09351682 - time (sec): 26.39 - samples/sec: 3330.41 - lr: 0.000047 - momentum: 0.000000
2023-10-17 17:29:31,143 epoch 2 - iter 432/723 - loss 0.09171543 - time (sec): 32.58 - samples/sec: 3302.14 - lr: 0.000047 - momentum: 0.000000
2023-10-17 17:29:36,313 epoch 2 - iter 504/723 - loss 0.08991783 - time (sec): 37.75 - samples/sec: 3317.34 - lr: 0.000046 - momentum: 0.000000
2023-10-17 17:29:41,332 epoch 2 - iter 576/723 - loss 0.08743025 - time (sec): 42.77 - samples/sec: 3322.67 - lr: 0.000046 - momentum: 0.000000
2023-10-17 17:29:46,509 epoch 2 - iter 648/723 - loss 0.08684948 - time (sec): 47.95 - samples/sec: 3308.97 - lr: 0.000045 - momentum: 0.000000
2023-10-17 17:29:52,156 epoch 2 - iter 720/723 - loss 0.08531715 - time (sec): 53.60 - samples/sec: 3275.07 - lr: 0.000044 - momentum: 0.000000
2023-10-17 17:29:52,352 ----------------------------------------------------------------------------------------------------
2023-10-17 17:29:52,352 EPOCH 2 done: loss 0.0854 - lr: 0.000044
2023-10-17 17:29:56,717 DEV : loss 0.12996011972427368 - f1-score (micro avg) 0.6985
2023-10-17 17:29:56,733 ----------------------------------------------------------------------------------------------------
2023-10-17 17:30:02,368 epoch 3 - iter 72/723 - loss 0.07263265 - time (sec): 5.63 - samples/sec: 3086.04 - lr: 0.000044 - momentum: 0.000000
2023-10-17 17:30:08,080 epoch 3 - iter 144/723 - loss 0.06576995 - time (sec): 11.34 - samples/sec: 3162.51 - lr: 0.000043 - momentum: 0.000000
2023-10-17 17:30:13,506 epoch 3 - iter 216/723 - loss 0.06486849 - time (sec): 16.77 - samples/sec: 3233.55 - lr: 0.000043 - momentum: 0.000000
2023-10-17 17:30:18,957 epoch 3 - iter 288/723 - loss 0.05999593 - time (sec): 22.22 - samples/sec: 3239.41 - lr: 0.000042 - momentum: 0.000000
2023-10-17 17:30:24,068 epoch 3 - iter 360/723 - loss 0.05991573 - time (sec): 27.33 - samples/sec: 3235.14 - lr: 0.000042 - momentum: 0.000000
2023-10-17 17:30:29,685 epoch 3 - iter 432/723 - loss 0.06066152 - time (sec): 32.95 - samples/sec: 3234.13 - lr: 0.000041 - momentum: 0.000000
2023-10-17 17:30:35,555 epoch 3 - iter 504/723 - loss 0.06270855 - time (sec): 38.82 - samples/sec: 3207.12 - lr: 0.000041 - momentum: 0.000000
2023-10-17 17:30:40,841 epoch 3 - iter 576/723 - loss 0.06209895 - time (sec): 44.11 - samples/sec: 3200.17 - lr: 0.000040 - momentum: 0.000000
2023-10-17 17:30:46,401 epoch 3 - iter 648/723 - loss 0.06130730 - time (sec): 49.67 - samples/sec: 3186.94 - lr: 0.000039 - momentum: 0.000000
2023-10-17 17:30:52,184 epoch 3 - iter 720/723 - loss 0.06221705 - time (sec): 55.45 - samples/sec: 3172.81 - lr: 0.000039 - momentum: 0.000000
2023-10-17 17:30:52,371 ----------------------------------------------------------------------------------------------------
2023-10-17 17:30:52,371 EPOCH 3 done: loss 0.0622 - lr: 0.000039
2023-10-17 17:30:56,033 DEV : loss 0.06001274287700653 - f1-score (micro avg) 0.8684
2023-10-17 17:30:56,053 saving best model
2023-10-17 17:30:56,585 ----------------------------------------------------------------------------------------------------
2023-10-17 17:31:01,784 epoch 4 - iter 72/723 - loss 0.03813632 - time (sec): 5.20 - samples/sec: 3213.96 - lr: 0.000038 - momentum: 0.000000
2023-10-17 17:31:07,908 epoch 4 - iter 144/723 - loss 0.03747693 - time (sec): 11.32 - samples/sec: 3060.36 - lr: 0.000038 - momentum: 0.000000
2023-10-17 17:31:13,233 epoch 4 - iter 216/723 - loss 0.03936375 - time (sec): 16.64 - samples/sec: 3109.39 - lr: 0.000037 - momentum: 0.000000
2023-10-17 17:31:18,623 epoch 4 - iter 288/723 - loss 0.04307423 - time (sec): 22.03 - samples/sec: 3142.16 - lr: 0.000037 - momentum: 0.000000
2023-10-17 17:31:23,678 epoch 4 - iter 360/723 - loss 0.04236315 - time (sec): 27.09 - samples/sec: 3185.58 - lr: 0.000036 - momentum: 0.000000
2023-10-17 17:31:29,478 epoch 4 - iter 432/723 - loss 0.04209825 - time (sec): 32.89 - samples/sec: 3158.73 - lr: 0.000036 - momentum: 0.000000
2023-10-17 17:31:34,915 epoch 4 - iter 504/723 - loss 0.04200339 - time (sec): 38.33 - samples/sec: 3176.76 - lr: 0.000035 - momentum: 0.000000
2023-10-17 17:31:40,837 epoch 4 - iter 576/723 - loss 0.04275515 - time (sec): 44.25 - samples/sec: 3185.87 - lr: 0.000034 - momentum: 0.000000
2023-10-17 17:31:46,159 epoch 4 - iter 648/723 - loss 0.04212727 - time (sec): 49.57 - samples/sec: 3186.36 - lr: 0.000034 - momentum: 0.000000
2023-10-17 17:31:51,490 epoch 4 - iter 720/723 - loss 0.04197729 - time (sec): 54.90 - samples/sec: 3201.14 - lr: 0.000033 - momentum: 0.000000
2023-10-17 17:31:51,670 ----------------------------------------------------------------------------------------------------
2023-10-17 17:31:51,671 EPOCH 4 done: loss 0.0420 - lr: 0.000033
2023-10-17 17:31:55,071 DEV : loss 0.0824296697974205 - f1-score (micro avg) 0.8726
2023-10-17 17:31:55,090 saving best model
2023-10-17 17:31:55,766 ----------------------------------------------------------------------------------------------------
2023-10-17 17:32:01,345 epoch 5 - iter 72/723 - loss 0.02976008 - time (sec): 5.58 - samples/sec: 3219.59 - lr: 0.000033 - momentum: 0.000000
2023-10-17 17:32:06,577 epoch 5 - iter 144/723 - loss 0.02756584 - time (sec): 10.81 - samples/sec: 3239.64 - lr: 0.000032 - momentum: 0.000000
2023-10-17 17:32:12,023 epoch 5 - iter 216/723 - loss 0.03103086 - time (sec): 16.26 - samples/sec: 3261.33 - lr: 0.000032 - momentum: 0.000000
2023-10-17 17:32:17,738 epoch 5 - iter 288/723 - loss 0.02922852 - time (sec): 21.97 - samples/sec: 3237.78 - lr: 0.000031 - momentum: 0.000000
2023-10-17 17:32:23,061 epoch 5 - iter 360/723 - loss 0.02855856 - time (sec): 27.29 - samples/sec: 3222.87 - lr: 0.000031 - momentum: 0.000000
2023-10-17 17:32:28,710 epoch 5 - iter 432/723 - loss 0.03052075 - time (sec): 32.94 - samples/sec: 3218.79 - lr: 0.000030 - momentum: 0.000000
2023-10-17 17:32:34,177 epoch 5 - iter 504/723 - loss 0.03129817 - time (sec): 38.41 - samples/sec: 3228.96 - lr: 0.000029 - momentum: 0.000000
2023-10-17 17:32:39,137 epoch 5 - iter 576/723 - loss 0.03242960 - time (sec): 43.37 - samples/sec: 3246.68 - lr: 0.000029 - momentum: 0.000000
2023-10-17 17:32:44,212 epoch 5 - iter 648/723 - loss 0.03218350 - time (sec): 48.44 - samples/sec: 3259.18 - lr: 0.000028 - momentum: 0.000000
2023-10-17 17:32:49,585 epoch 5 - iter 720/723 - loss 0.03177889 - time (sec): 53.82 - samples/sec: 3264.47 - lr: 0.000028 - momentum: 0.000000
2023-10-17 17:32:49,775 ----------------------------------------------------------------------------------------------------
2023-10-17 17:32:49,775 EPOCH 5 done: loss 0.0318 - lr: 0.000028
2023-10-17 17:32:53,581 DEV : loss 0.10424862802028656 - f1-score (micro avg) 0.8471
2023-10-17 17:32:53,601 ----------------------------------------------------------------------------------------------------
2023-10-17 17:32:58,963 epoch 6 - iter 72/723 - loss 0.02973653 - time (sec): 5.36 - samples/sec: 3496.63 - lr: 0.000027 - momentum: 0.000000
2023-10-17 17:33:04,263 epoch 6 - iter 144/723 - loss 0.02285847 - time (sec): 10.66 - samples/sec: 3353.33 - lr: 0.000027 - momentum: 0.000000
2023-10-17 17:33:09,929 epoch 6 - iter 216/723 - loss 0.02429447 - time (sec): 16.33 - samples/sec: 3326.30 - lr: 0.000026 - momentum: 0.000000
2023-10-17 17:33:15,711 epoch 6 - iter 288/723 - loss 0.02499962 - time (sec): 22.11 - samples/sec: 3282.73 - lr: 0.000026 - momentum: 0.000000
2023-10-17 17:33:20,942 epoch 6 - iter 360/723 - loss 0.02571431 - time (sec): 27.34 - samples/sec: 3278.43 - lr: 0.000025 - momentum: 0.000000
2023-10-17 17:33:26,294 epoch 6 - iter 432/723 - loss 0.02517103 - time (sec): 32.69 - samples/sec: 3294.69 - lr: 0.000024 - momentum: 0.000000
2023-10-17 17:33:31,447 epoch 6 - iter 504/723 - loss 0.02605168 - time (sec): 37.84 - samples/sec: 3287.99 - lr: 0.000024 - momentum: 0.000000
2023-10-17 17:33:36,580 epoch 6 - iter 576/723 - loss 0.02631389 - time (sec): 42.98 - samples/sec: 3304.30 - lr: 0.000023 - momentum: 0.000000
2023-10-17 17:33:41,733 epoch 6 - iter 648/723 - loss 0.02543096 - time (sec): 48.13 - samples/sec: 3295.75 - lr: 0.000023 - momentum: 0.000000
2023-10-17 17:33:46,823 epoch 6 - iter 720/723 - loss 0.02521898 - time (sec): 53.22 - samples/sec: 3302.40 - lr: 0.000022 - momentum: 0.000000
2023-10-17 17:33:47,018 ----------------------------------------------------------------------------------------------------
2023-10-17 17:33:47,018 EPOCH 6 done: loss 0.0252 - lr: 0.000022
2023-10-17 17:33:50,373 DEV : loss 0.11773881316184998 - f1-score (micro avg) 0.8452
2023-10-17 17:33:50,391 ----------------------------------------------------------------------------------------------------
2023-10-17 17:33:55,498 epoch 7 - iter 72/723 - loss 0.01211954 - time (sec): 5.11 - samples/sec: 3363.93 - lr: 0.000022 - momentum: 0.000000
2023-10-17 17:34:00,867 epoch 7 - iter 144/723 - loss 0.01732599 - time (sec): 10.47 - samples/sec: 3294.18 - lr: 0.000021 - momentum: 0.000000
2023-10-17 17:34:06,234 epoch 7 - iter 216/723 - loss 0.01866389 - time (sec): 15.84 - samples/sec: 3299.82 - lr: 0.000021 - momentum: 0.000000
2023-10-17 17:34:11,575 epoch 7 - iter 288/723 - loss 0.02100187 - time (sec): 21.18 - samples/sec: 3323.79 - lr: 0.000020 - momentum: 0.000000
2023-10-17 17:34:17,437 epoch 7 - iter 360/723 - loss 0.01933756 - time (sec): 27.04 - samples/sec: 3241.66 - lr: 0.000019 - momentum: 0.000000
2023-10-17 17:34:22,923 epoch 7 - iter 432/723 - loss 0.01996565 - time (sec): 32.53 - samples/sec: 3262.22 - lr: 0.000019 - momentum: 0.000000
2023-10-17 17:34:28,273 epoch 7 - iter 504/723 - loss 0.01973901 - time (sec): 37.88 - samples/sec: 3284.32 - lr: 0.000018 - momentum: 0.000000
2023-10-17 17:34:33,625 epoch 7 - iter 576/723 - loss 0.01903934 - time (sec): 43.23 - samples/sec: 3274.98 - lr: 0.000018 - momentum: 0.000000
2023-10-17 17:34:38,898 epoch 7 - iter 648/723 - loss 0.01808703 - time (sec): 48.51 - samples/sec: 3262.65 - lr: 0.000017 - momentum: 0.000000
2023-10-17 17:34:44,431 epoch 7 - iter 720/723 - loss 0.01730273 - time (sec): 54.04 - samples/sec: 3245.34 - lr: 0.000017 - momentum: 0.000000
2023-10-17 17:34:44,838 ----------------------------------------------------------------------------------------------------
2023-10-17 17:34:44,838 EPOCH 7 done: loss 0.0174 - lr: 0.000017
2023-10-17 17:34:48,262 DEV : loss 0.1411609798669815 - f1-score (micro avg) 0.8591
2023-10-17 17:34:48,280 ----------------------------------------------------------------------------------------------------
2023-10-17 17:34:53,518 epoch 8 - iter 72/723 - loss 0.01428987 - time (sec): 5.24 - samples/sec: 3189.97 - lr: 0.000016 - momentum: 0.000000
2023-10-17 17:34:59,013 epoch 8 - iter 144/723 - loss 0.00959426 - time (sec): 10.73 - samples/sec: 3163.77 - lr: 0.000016 - momentum: 0.000000
2023-10-17 17:35:04,311 epoch 8 - iter 216/723 - loss 0.01074675 - time (sec): 16.03 - samples/sec: 3194.32 - lr: 0.000015 - momentum: 0.000000
2023-10-17 17:35:09,779 epoch 8 - iter 288/723 - loss 0.01082205 - time (sec): 21.50 - samples/sec: 3207.07 - lr: 0.000014 - momentum: 0.000000
2023-10-17 17:35:15,456 epoch 8 - iter 360/723 - loss 0.01174131 - time (sec): 27.17 - samples/sec: 3185.76 - lr: 0.000014 - momentum: 0.000000
2023-10-17 17:35:20,762 epoch 8 - iter 432/723 - loss 0.01179153 - time (sec): 32.48 - samples/sec: 3199.23 - lr: 0.000013 - momentum: 0.000000
2023-10-17 17:35:26,144 epoch 8 - iter 504/723 - loss 0.01159015 - time (sec): 37.86 - samples/sec: 3218.57 - lr: 0.000013 - momentum: 0.000000
2023-10-17 17:35:31,465 epoch 8 - iter 576/723 - loss 0.01238696 - time (sec): 43.18 - samples/sec: 3231.32 - lr: 0.000012 - momentum: 0.000000
2023-10-17 17:35:37,327 epoch 8 - iter 648/723 - loss 0.01238150 - time (sec): 49.05 - samples/sec: 3242.41 - lr: 0.000012 - momentum: 0.000000
2023-10-17 17:35:42,459 epoch 8 - iter 720/723 - loss 0.01243442 - time (sec): 54.18 - samples/sec: 3241.49 - lr: 0.000011 - momentum: 0.000000
2023-10-17 17:35:42,643 ----------------------------------------------------------------------------------------------------
2023-10-17 17:35:42,643 EPOCH 8 done: loss 0.0125 - lr: 0.000011
2023-10-17 17:35:46,562 DEV : loss 0.13958391547203064 - f1-score (micro avg) 0.865
2023-10-17 17:35:46,583 ----------------------------------------------------------------------------------------------------
2023-10-17 17:35:51,843 epoch 9 - iter 72/723 - loss 0.00536784 - time (sec): 5.26 - samples/sec: 3259.05 - lr: 0.000011 - momentum: 0.000000
2023-10-17 17:35:57,193 epoch 9 - iter 144/723 - loss 0.00538410 - time (sec): 10.61 - samples/sec: 3301.20 - lr: 0.000010 - momentum: 0.000000
2023-10-17 17:36:02,650 epoch 9 - iter 216/723 - loss 0.00537525 - time (sec): 16.07 - samples/sec: 3268.19 - lr: 0.000009 - momentum: 0.000000
2023-10-17 17:36:08,285 epoch 9 - iter 288/723 - loss 0.00698419 - time (sec): 21.70 - samples/sec: 3267.87 - lr: 0.000009 - momentum: 0.000000
2023-10-17 17:36:14,022 epoch 9 - iter 360/723 - loss 0.00741743 - time (sec): 27.44 - samples/sec: 3253.34 - lr: 0.000008 - momentum: 0.000000
2023-10-17 17:36:19,237 epoch 9 - iter 432/723 - loss 0.00850564 - time (sec): 32.65 - samples/sec: 3272.18 - lr: 0.000008 - momentum: 0.000000
2023-10-17 17:36:24,435 epoch 9 - iter 504/723 - loss 0.00834734 - time (sec): 37.85 - samples/sec: 3261.14 - lr: 0.000007 - momentum: 0.000000
2023-10-17 17:36:29,366 epoch 9 - iter 576/723 - loss 0.00779422 - time (sec): 42.78 - samples/sec: 3263.01 - lr: 0.000007 - momentum: 0.000000
2023-10-17 17:36:34,772 epoch 9 - iter 648/723 - loss 0.00761217 - time (sec): 48.19 - samples/sec: 3279.14 - lr: 0.000006 - momentum: 0.000000
2023-10-17 17:36:40,289 epoch 9 - iter 720/723 - loss 0.00782569 - time (sec): 53.70 - samples/sec: 3271.12 - lr: 0.000006 - momentum: 0.000000
2023-10-17 17:36:40,464 ----------------------------------------------------------------------------------------------------
2023-10-17 17:36:40,464 EPOCH 9 done: loss 0.0078 - lr: 0.000006
2023-10-17 17:36:43,871 DEV : loss 0.1421152502298355 - f1-score (micro avg) 0.8627
2023-10-17 17:36:43,891 ----------------------------------------------------------------------------------------------------
2023-10-17 17:36:49,280 epoch 10 - iter 72/723 - loss 0.01028871 - time (sec): 5.39 - samples/sec: 3344.29 - lr: 0.000005 - momentum: 0.000000
2023-10-17 17:36:54,336 epoch 10 - iter 144/723 - loss 0.00650322 - time (sec): 10.44 - samples/sec: 3325.81 - lr: 0.000004 - momentum: 0.000000
2023-10-17 17:37:00,082 epoch 10 - iter 216/723 - loss 0.00664087 - time (sec): 16.19 - samples/sec: 3281.14 - lr: 0.000004 - momentum: 0.000000
2023-10-17 17:37:05,727 epoch 10 - iter 288/723 - loss 0.00546743 - time (sec): 21.83 - samples/sec: 3259.82 - lr: 0.000003 - momentum: 0.000000
2023-10-17 17:37:11,027 epoch 10 - iter 360/723 - loss 0.00555269 - time (sec): 27.13 - samples/sec: 3271.40 - lr: 0.000003 - momentum: 0.000000
2023-10-17 17:37:16,744 epoch 10 - iter 432/723 - loss 0.00516482 - time (sec): 32.85 - samples/sec: 3237.38 - lr: 0.000002 - momentum: 0.000000
2023-10-17 17:37:21,902 epoch 10 - iter 504/723 - loss 0.00532677 - time (sec): 38.01 - samples/sec: 3248.19 - lr: 0.000002 - momentum: 0.000000
2023-10-17 17:37:27,390 epoch 10 - iter 576/723 - loss 0.00486427 - time (sec): 43.50 - samples/sec: 3241.67 - lr: 0.000001 - momentum: 0.000000
2023-10-17 17:37:32,975 epoch 10 - iter 648/723 - loss 0.00551894 - time (sec): 49.08 - samples/sec: 3237.10 - lr: 0.000001 - momentum: 0.000000
2023-10-17 17:37:38,248 epoch 10 - iter 720/723 - loss 0.00553918 - time (sec): 54.36 - samples/sec: 3235.06 - lr: 0.000000 - momentum: 0.000000
2023-10-17 17:37:38,400 ----------------------------------------------------------------------------------------------------
2023-10-17 17:37:38,401 EPOCH 10 done: loss 0.0055 - lr: 0.000000
2023-10-17 17:37:41,831 DEV : loss 0.15293623507022858 - f1-score (micro avg) 0.8606
2023-10-17 17:37:42,249 ----------------------------------------------------------------------------------------------------
2023-10-17 17:37:42,251 Loading model from best epoch ...
2023-10-17 17:37:44,011 SequenceTagger predicts: Dictionary with 13 tags: O, S-LOC, B-LOC, E-LOC, I-LOC, S-PER, B-PER, E-PER, I-PER, S-ORG, B-ORG, E-ORG, I-ORG
2023-10-17 17:37:47,497
Results:
- F-score (micro) 0.86
- F-score (macro) 0.7669
- Accuracy 0.765
By class:
precision recall f1-score support
PER 0.8685 0.8361 0.8520 482
LOC 0.9177 0.9258 0.9217 458
ORG 0.4937 0.5652 0.5270 69
micro avg 0.8617 0.8583 0.8600 1009
macro avg 0.7600 0.7757 0.7669 1009
weighted avg 0.8652 0.8583 0.8614 1009
2023-10-17 17:37:47,497 ----------------------------------------------------------------------------------------------------
|