File size: 23,925 Bytes
114076f |
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 17:47:58,832 ----------------------------------------------------------------------------------------------------
2023-10-17 17:47:58,833 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=17, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-17 17:47:58,833 ----------------------------------------------------------------------------------------------------
2023-10-17 17:47:58,833 MultiCorpus: 1166 train + 165 dev + 415 test sentences
- NER_HIPE_2022 Corpus: 1166 train + 165 dev + 415 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/newseye/fi/with_doc_seperator
2023-10-17 17:47:58,833 ----------------------------------------------------------------------------------------------------
2023-10-17 17:47:58,833 Train: 1166 sentences
2023-10-17 17:47:58,833 (train_with_dev=False, train_with_test=False)
2023-10-17 17:47:58,833 ----------------------------------------------------------------------------------------------------
2023-10-17 17:47:58,833 Training Params:
2023-10-17 17:47:58,833 - learning_rate: "3e-05"
2023-10-17 17:47:58,833 - mini_batch_size: "8"
2023-10-17 17:47:58,833 - max_epochs: "10"
2023-10-17 17:47:58,833 - shuffle: "True"
2023-10-17 17:47:58,834 ----------------------------------------------------------------------------------------------------
2023-10-17 17:47:58,834 Plugins:
2023-10-17 17:47:58,834 - TensorboardLogger
2023-10-17 17:47:58,834 - LinearScheduler | warmup_fraction: '0.1'
2023-10-17 17:47:58,834 ----------------------------------------------------------------------------------------------------
2023-10-17 17:47:58,834 Final evaluation on model from best epoch (best-model.pt)
2023-10-17 17:47:58,834 - metric: "('micro avg', 'f1-score')"
2023-10-17 17:47:58,834 ----------------------------------------------------------------------------------------------------
2023-10-17 17:47:58,834 Computation:
2023-10-17 17:47:58,834 - compute on device: cuda:0
2023-10-17 17:47:58,834 - embedding storage: none
2023-10-17 17:47:58,834 ----------------------------------------------------------------------------------------------------
2023-10-17 17:47:58,834 Model training base path: "hmbench-newseye/fi-hmteams/teams-base-historic-multilingual-discriminator-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-2"
2023-10-17 17:47:58,834 ----------------------------------------------------------------------------------------------------
2023-10-17 17:47:58,834 ----------------------------------------------------------------------------------------------------
2023-10-17 17:47:58,834 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-17 17:48:00,326 epoch 1 - iter 14/146 - loss 3.70831643 - time (sec): 1.49 - samples/sec: 2870.09 - lr: 0.000003 - momentum: 0.000000
2023-10-17 17:48:01,701 epoch 1 - iter 28/146 - loss 3.48850072 - time (sec): 2.87 - samples/sec: 3072.96 - lr: 0.000006 - momentum: 0.000000
2023-10-17 17:48:03,431 epoch 1 - iter 42/146 - loss 3.02057847 - time (sec): 4.60 - samples/sec: 2846.31 - lr: 0.000008 - momentum: 0.000000
2023-10-17 17:48:04,717 epoch 1 - iter 56/146 - loss 2.50123340 - time (sec): 5.88 - samples/sec: 2868.63 - lr: 0.000011 - momentum: 0.000000
2023-10-17 17:48:05,892 epoch 1 - iter 70/146 - loss 2.20026054 - time (sec): 7.06 - samples/sec: 2894.66 - lr: 0.000014 - momentum: 0.000000
2023-10-17 17:48:07,176 epoch 1 - iter 84/146 - loss 1.93860951 - time (sec): 8.34 - samples/sec: 2910.26 - lr: 0.000017 - momentum: 0.000000
2023-10-17 17:48:08,604 epoch 1 - iter 98/146 - loss 1.71290433 - time (sec): 9.77 - samples/sec: 2947.37 - lr: 0.000020 - momentum: 0.000000
2023-10-17 17:48:09,829 epoch 1 - iter 112/146 - loss 1.56730272 - time (sec): 10.99 - samples/sec: 2961.94 - lr: 0.000023 - momentum: 0.000000
2023-10-17 17:48:11,184 epoch 1 - iter 126/146 - loss 1.43497338 - time (sec): 12.35 - samples/sec: 2966.86 - lr: 0.000026 - momentum: 0.000000
2023-10-17 17:48:13,063 epoch 1 - iter 140/146 - loss 1.30380866 - time (sec): 14.23 - samples/sec: 2968.73 - lr: 0.000029 - momentum: 0.000000
2023-10-17 17:48:13,871 ----------------------------------------------------------------------------------------------------
2023-10-17 17:48:13,871 EPOCH 1 done: loss 1.2702 - lr: 0.000029
2023-10-17 17:48:14,887 DEV : loss 0.20914216339588165 - f1-score (micro avg) 0.4434
2023-10-17 17:48:14,891 saving best model
2023-10-17 17:48:15,219 ----------------------------------------------------------------------------------------------------
2023-10-17 17:48:16,888 epoch 2 - iter 14/146 - loss 0.34822885 - time (sec): 1.67 - samples/sec: 2862.44 - lr: 0.000030 - momentum: 0.000000
2023-10-17 17:48:18,130 epoch 2 - iter 28/146 - loss 0.31488826 - time (sec): 2.91 - samples/sec: 2912.01 - lr: 0.000029 - momentum: 0.000000
2023-10-17 17:48:19,353 epoch 2 - iter 42/146 - loss 0.28349960 - time (sec): 4.13 - samples/sec: 2950.95 - lr: 0.000029 - momentum: 0.000000
2023-10-17 17:48:20,504 epoch 2 - iter 56/146 - loss 0.27893653 - time (sec): 5.28 - samples/sec: 2993.09 - lr: 0.000029 - momentum: 0.000000
2023-10-17 17:48:22,128 epoch 2 - iter 70/146 - loss 0.28179222 - time (sec): 6.91 - samples/sec: 2975.41 - lr: 0.000028 - momentum: 0.000000
2023-10-17 17:48:23,714 epoch 2 - iter 84/146 - loss 0.26094264 - time (sec): 8.49 - samples/sec: 2918.48 - lr: 0.000028 - momentum: 0.000000
2023-10-17 17:48:25,054 epoch 2 - iter 98/146 - loss 0.24468379 - time (sec): 9.83 - samples/sec: 2907.34 - lr: 0.000028 - momentum: 0.000000
2023-10-17 17:48:26,279 epoch 2 - iter 112/146 - loss 0.23942792 - time (sec): 11.06 - samples/sec: 2929.29 - lr: 0.000027 - momentum: 0.000000
2023-10-17 17:48:27,639 epoch 2 - iter 126/146 - loss 0.22920784 - time (sec): 12.42 - samples/sec: 2958.30 - lr: 0.000027 - momentum: 0.000000
2023-10-17 17:48:29,567 epoch 2 - iter 140/146 - loss 0.22136768 - time (sec): 14.35 - samples/sec: 2968.22 - lr: 0.000027 - momentum: 0.000000
2023-10-17 17:48:30,113 ----------------------------------------------------------------------------------------------------
2023-10-17 17:48:30,113 EPOCH 2 done: loss 0.2179 - lr: 0.000027
2023-10-17 17:48:31,360 DEV : loss 0.1328418254852295 - f1-score (micro avg) 0.6194
2023-10-17 17:48:31,365 saving best model
2023-10-17 17:48:31,804 ----------------------------------------------------------------------------------------------------
2023-10-17 17:48:32,991 epoch 3 - iter 14/146 - loss 0.16652025 - time (sec): 1.19 - samples/sec: 2928.21 - lr: 0.000026 - momentum: 0.000000
2023-10-17 17:48:34,759 epoch 3 - iter 28/146 - loss 0.13123510 - time (sec): 2.95 - samples/sec: 2859.92 - lr: 0.000026 - momentum: 0.000000
2023-10-17 17:48:36,129 epoch 3 - iter 42/146 - loss 0.12284675 - time (sec): 4.32 - samples/sec: 2937.26 - lr: 0.000026 - momentum: 0.000000
2023-10-17 17:48:37,838 epoch 3 - iter 56/146 - loss 0.11703501 - time (sec): 6.03 - samples/sec: 2887.88 - lr: 0.000025 - momentum: 0.000000
2023-10-17 17:48:39,065 epoch 3 - iter 70/146 - loss 0.12806525 - time (sec): 7.26 - samples/sec: 2910.71 - lr: 0.000025 - momentum: 0.000000
2023-10-17 17:48:40,354 epoch 3 - iter 84/146 - loss 0.12466717 - time (sec): 8.55 - samples/sec: 2959.90 - lr: 0.000025 - momentum: 0.000000
2023-10-17 17:48:41,475 epoch 3 - iter 98/146 - loss 0.12411716 - time (sec): 9.67 - samples/sec: 2953.35 - lr: 0.000024 - momentum: 0.000000
2023-10-17 17:48:43,083 epoch 3 - iter 112/146 - loss 0.12366808 - time (sec): 11.28 - samples/sec: 2980.33 - lr: 0.000024 - momentum: 0.000000
2023-10-17 17:48:44,797 epoch 3 - iter 126/146 - loss 0.12260040 - time (sec): 12.99 - samples/sec: 2924.89 - lr: 0.000024 - momentum: 0.000000
2023-10-17 17:48:46,433 epoch 3 - iter 140/146 - loss 0.12422938 - time (sec): 14.63 - samples/sec: 2936.40 - lr: 0.000024 - momentum: 0.000000
2023-10-17 17:48:46,872 ----------------------------------------------------------------------------------------------------
2023-10-17 17:48:46,872 EPOCH 3 done: loss 0.1216 - lr: 0.000024
2023-10-17 17:48:48,107 DEV : loss 0.09678074717521667 - f1-score (micro avg) 0.7391
2023-10-17 17:48:48,113 saving best model
2023-10-17 17:48:48,546 ----------------------------------------------------------------------------------------------------
2023-10-17 17:48:50,020 epoch 4 - iter 14/146 - loss 0.07210812 - time (sec): 1.47 - samples/sec: 3219.07 - lr: 0.000023 - momentum: 0.000000
2023-10-17 17:48:51,407 epoch 4 - iter 28/146 - loss 0.08676056 - time (sec): 2.86 - samples/sec: 3182.49 - lr: 0.000023 - momentum: 0.000000
2023-10-17 17:48:52,963 epoch 4 - iter 42/146 - loss 0.09667761 - time (sec): 4.42 - samples/sec: 3032.38 - lr: 0.000022 - momentum: 0.000000
2023-10-17 17:48:54,228 epoch 4 - iter 56/146 - loss 0.09045365 - time (sec): 5.68 - samples/sec: 2967.39 - lr: 0.000022 - momentum: 0.000000
2023-10-17 17:48:55,704 epoch 4 - iter 70/146 - loss 0.08679327 - time (sec): 7.16 - samples/sec: 2956.52 - lr: 0.000022 - momentum: 0.000000
2023-10-17 17:48:57,298 epoch 4 - iter 84/146 - loss 0.08850398 - time (sec): 8.75 - samples/sec: 2964.11 - lr: 0.000021 - momentum: 0.000000
2023-10-17 17:48:58,514 epoch 4 - iter 98/146 - loss 0.08454841 - time (sec): 9.97 - samples/sec: 2943.86 - lr: 0.000021 - momentum: 0.000000
2023-10-17 17:48:59,963 epoch 4 - iter 112/146 - loss 0.08390306 - time (sec): 11.42 - samples/sec: 2920.25 - lr: 0.000021 - momentum: 0.000000
2023-10-17 17:49:01,517 epoch 4 - iter 126/146 - loss 0.08184749 - time (sec): 12.97 - samples/sec: 2928.00 - lr: 0.000021 - momentum: 0.000000
2023-10-17 17:49:03,073 epoch 4 - iter 140/146 - loss 0.07829550 - time (sec): 14.53 - samples/sec: 2931.91 - lr: 0.000020 - momentum: 0.000000
2023-10-17 17:49:03,702 ----------------------------------------------------------------------------------------------------
2023-10-17 17:49:03,703 EPOCH 4 done: loss 0.0787 - lr: 0.000020
2023-10-17 17:49:04,965 DEV : loss 0.10548200458288193 - f1-score (micro avg) 0.7217
2023-10-17 17:49:04,970 ----------------------------------------------------------------------------------------------------
2023-10-17 17:49:06,352 epoch 5 - iter 14/146 - loss 0.07117211 - time (sec): 1.38 - samples/sec: 2781.96 - lr: 0.000020 - momentum: 0.000000
2023-10-17 17:49:07,894 epoch 5 - iter 28/146 - loss 0.06763706 - time (sec): 2.92 - samples/sec: 2897.40 - lr: 0.000019 - momentum: 0.000000
2023-10-17 17:49:09,304 epoch 5 - iter 42/146 - loss 0.05999953 - time (sec): 4.33 - samples/sec: 3006.81 - lr: 0.000019 - momentum: 0.000000
2023-10-17 17:49:11,007 epoch 5 - iter 56/146 - loss 0.06257175 - time (sec): 6.03 - samples/sec: 2940.26 - lr: 0.000019 - momentum: 0.000000
2023-10-17 17:49:12,341 epoch 5 - iter 70/146 - loss 0.05800479 - time (sec): 7.37 - samples/sec: 2936.86 - lr: 0.000018 - momentum: 0.000000
2023-10-17 17:49:13,615 epoch 5 - iter 84/146 - loss 0.05788663 - time (sec): 8.64 - samples/sec: 2906.02 - lr: 0.000018 - momentum: 0.000000
2023-10-17 17:49:15,184 epoch 5 - iter 98/146 - loss 0.05396484 - time (sec): 10.21 - samples/sec: 2930.36 - lr: 0.000018 - momentum: 0.000000
2023-10-17 17:49:16,441 epoch 5 - iter 112/146 - loss 0.05252850 - time (sec): 11.47 - samples/sec: 2935.02 - lr: 0.000018 - momentum: 0.000000
2023-10-17 17:49:18,320 epoch 5 - iter 126/146 - loss 0.05460215 - time (sec): 13.35 - samples/sec: 2899.42 - lr: 0.000017 - momentum: 0.000000
2023-10-17 17:49:19,809 epoch 5 - iter 140/146 - loss 0.05860187 - time (sec): 14.84 - samples/sec: 2878.39 - lr: 0.000017 - momentum: 0.000000
2023-10-17 17:49:20,338 ----------------------------------------------------------------------------------------------------
2023-10-17 17:49:20,338 EPOCH 5 done: loss 0.0577 - lr: 0.000017
2023-10-17 17:49:21,645 DEV : loss 0.11666107177734375 - f1-score (micro avg) 0.7357
2023-10-17 17:49:21,653 ----------------------------------------------------------------------------------------------------
2023-10-17 17:49:23,019 epoch 6 - iter 14/146 - loss 0.03065155 - time (sec): 1.36 - samples/sec: 2882.79 - lr: 0.000016 - momentum: 0.000000
2023-10-17 17:49:24,773 epoch 6 - iter 28/146 - loss 0.03493187 - time (sec): 3.12 - samples/sec: 2845.14 - lr: 0.000016 - momentum: 0.000000
2023-10-17 17:49:26,054 epoch 6 - iter 42/146 - loss 0.03689635 - time (sec): 4.40 - samples/sec: 2766.74 - lr: 0.000016 - momentum: 0.000000
2023-10-17 17:49:27,639 epoch 6 - iter 56/146 - loss 0.03772627 - time (sec): 5.98 - samples/sec: 2815.18 - lr: 0.000015 - momentum: 0.000000
2023-10-17 17:49:29,447 epoch 6 - iter 70/146 - loss 0.04035301 - time (sec): 7.79 - samples/sec: 2768.42 - lr: 0.000015 - momentum: 0.000000
2023-10-17 17:49:30,920 epoch 6 - iter 84/146 - loss 0.04015679 - time (sec): 9.27 - samples/sec: 2812.19 - lr: 0.000015 - momentum: 0.000000
2023-10-17 17:49:32,425 epoch 6 - iter 98/146 - loss 0.04222793 - time (sec): 10.77 - samples/sec: 2813.85 - lr: 0.000015 - momentum: 0.000000
2023-10-17 17:49:33,802 epoch 6 - iter 112/146 - loss 0.04043294 - time (sec): 12.15 - samples/sec: 2809.62 - lr: 0.000014 - momentum: 0.000000
2023-10-17 17:49:35,285 epoch 6 - iter 126/146 - loss 0.04103862 - time (sec): 13.63 - samples/sec: 2815.75 - lr: 0.000014 - momentum: 0.000000
2023-10-17 17:49:36,664 epoch 6 - iter 140/146 - loss 0.04074709 - time (sec): 15.01 - samples/sec: 2855.51 - lr: 0.000014 - momentum: 0.000000
2023-10-17 17:49:37,163 ----------------------------------------------------------------------------------------------------
2023-10-17 17:49:37,163 EPOCH 6 done: loss 0.0412 - lr: 0.000014
2023-10-17 17:49:38,675 DEV : loss 0.117428258061409 - f1-score (micro avg) 0.7669
2023-10-17 17:49:38,681 saving best model
2023-10-17 17:49:39,120 ----------------------------------------------------------------------------------------------------
2023-10-17 17:49:40,477 epoch 7 - iter 14/146 - loss 0.01790611 - time (sec): 1.35 - samples/sec: 2742.33 - lr: 0.000013 - momentum: 0.000000
2023-10-17 17:49:41,979 epoch 7 - iter 28/146 - loss 0.02367119 - time (sec): 2.86 - samples/sec: 2953.13 - lr: 0.000013 - momentum: 0.000000
2023-10-17 17:49:43,472 epoch 7 - iter 42/146 - loss 0.03528884 - time (sec): 4.35 - samples/sec: 2978.62 - lr: 0.000012 - momentum: 0.000000
2023-10-17 17:49:45,039 epoch 7 - iter 56/146 - loss 0.03487856 - time (sec): 5.92 - samples/sec: 2939.16 - lr: 0.000012 - momentum: 0.000000
2023-10-17 17:49:46,384 epoch 7 - iter 70/146 - loss 0.03083187 - time (sec): 7.26 - samples/sec: 2964.58 - lr: 0.000012 - momentum: 0.000000
2023-10-17 17:49:47,960 epoch 7 - iter 84/146 - loss 0.02860470 - time (sec): 8.84 - samples/sec: 2885.77 - lr: 0.000012 - momentum: 0.000000
2023-10-17 17:49:49,345 epoch 7 - iter 98/146 - loss 0.03056844 - time (sec): 10.22 - samples/sec: 2915.50 - lr: 0.000011 - momentum: 0.000000
2023-10-17 17:49:50,648 epoch 7 - iter 112/146 - loss 0.03330335 - time (sec): 11.53 - samples/sec: 2959.55 - lr: 0.000011 - momentum: 0.000000
2023-10-17 17:49:52,247 epoch 7 - iter 126/146 - loss 0.03165719 - time (sec): 13.12 - samples/sec: 2938.43 - lr: 0.000011 - momentum: 0.000000
2023-10-17 17:49:53,929 epoch 7 - iter 140/146 - loss 0.03178463 - time (sec): 14.81 - samples/sec: 2905.25 - lr: 0.000010 - momentum: 0.000000
2023-10-17 17:49:54,424 ----------------------------------------------------------------------------------------------------
2023-10-17 17:49:54,424 EPOCH 7 done: loss 0.0322 - lr: 0.000010
2023-10-17 17:49:55,741 DEV : loss 0.12387555837631226 - f1-score (micro avg) 0.7565
2023-10-17 17:49:55,747 ----------------------------------------------------------------------------------------------------
2023-10-17 17:49:57,081 epoch 8 - iter 14/146 - loss 0.03722011 - time (sec): 1.33 - samples/sec: 2811.37 - lr: 0.000010 - momentum: 0.000000
2023-10-17 17:49:58,664 epoch 8 - iter 28/146 - loss 0.03633970 - time (sec): 2.92 - samples/sec: 2805.10 - lr: 0.000009 - momentum: 0.000000
2023-10-17 17:50:00,042 epoch 8 - iter 42/146 - loss 0.03006267 - time (sec): 4.29 - samples/sec: 2771.47 - lr: 0.000009 - momentum: 0.000000
2023-10-17 17:50:01,404 epoch 8 - iter 56/146 - loss 0.02940184 - time (sec): 5.66 - samples/sec: 2809.56 - lr: 0.000009 - momentum: 0.000000
2023-10-17 17:50:02,743 epoch 8 - iter 70/146 - loss 0.02738822 - time (sec): 6.99 - samples/sec: 2896.54 - lr: 0.000009 - momentum: 0.000000
2023-10-17 17:50:04,234 epoch 8 - iter 84/146 - loss 0.02622802 - time (sec): 8.49 - samples/sec: 2951.33 - lr: 0.000008 - momentum: 0.000000
2023-10-17 17:50:05,468 epoch 8 - iter 98/146 - loss 0.02554764 - time (sec): 9.72 - samples/sec: 2961.07 - lr: 0.000008 - momentum: 0.000000
2023-10-17 17:50:07,005 epoch 8 - iter 112/146 - loss 0.02539222 - time (sec): 11.26 - samples/sec: 2977.63 - lr: 0.000008 - momentum: 0.000000
2023-10-17 17:50:08,351 epoch 8 - iter 126/146 - loss 0.02497153 - time (sec): 12.60 - samples/sec: 2975.95 - lr: 0.000007 - momentum: 0.000000
2023-10-17 17:50:09,996 epoch 8 - iter 140/146 - loss 0.02398278 - time (sec): 14.25 - samples/sec: 2994.84 - lr: 0.000007 - momentum: 0.000000
2023-10-17 17:50:10,711 ----------------------------------------------------------------------------------------------------
2023-10-17 17:50:10,711 EPOCH 8 done: loss 0.0239 - lr: 0.000007
2023-10-17 17:50:11,971 DEV : loss 0.12609358131885529 - f1-score (micro avg) 0.7699
2023-10-17 17:50:11,976 saving best model
2023-10-17 17:50:12,414 ----------------------------------------------------------------------------------------------------
2023-10-17 17:50:13,900 epoch 9 - iter 14/146 - loss 0.01567117 - time (sec): 1.48 - samples/sec: 3033.06 - lr: 0.000006 - momentum: 0.000000
2023-10-17 17:50:15,389 epoch 9 - iter 28/146 - loss 0.01839902 - time (sec): 2.97 - samples/sec: 2909.24 - lr: 0.000006 - momentum: 0.000000
2023-10-17 17:50:17,035 epoch 9 - iter 42/146 - loss 0.01793049 - time (sec): 4.61 - samples/sec: 2875.68 - lr: 0.000006 - momentum: 0.000000
2023-10-17 17:50:18,804 epoch 9 - iter 56/146 - loss 0.02188313 - time (sec): 6.38 - samples/sec: 2798.01 - lr: 0.000006 - momentum: 0.000000
2023-10-17 17:50:20,408 epoch 9 - iter 70/146 - loss 0.02090357 - time (sec): 7.98 - samples/sec: 2803.78 - lr: 0.000005 - momentum: 0.000000
2023-10-17 17:50:21,786 epoch 9 - iter 84/146 - loss 0.01866954 - time (sec): 9.36 - samples/sec: 2816.37 - lr: 0.000005 - momentum: 0.000000
2023-10-17 17:50:23,480 epoch 9 - iter 98/146 - loss 0.01734549 - time (sec): 11.06 - samples/sec: 2784.86 - lr: 0.000005 - momentum: 0.000000
2023-10-17 17:50:24,712 epoch 9 - iter 112/146 - loss 0.01775145 - time (sec): 12.29 - samples/sec: 2801.24 - lr: 0.000004 - momentum: 0.000000
2023-10-17 17:50:26,122 epoch 9 - iter 126/146 - loss 0.01676461 - time (sec): 13.70 - samples/sec: 2807.04 - lr: 0.000004 - momentum: 0.000000
2023-10-17 17:50:27,462 epoch 9 - iter 140/146 - loss 0.01669329 - time (sec): 15.04 - samples/sec: 2848.20 - lr: 0.000004 - momentum: 0.000000
2023-10-17 17:50:28,138 ----------------------------------------------------------------------------------------------------
2023-10-17 17:50:28,138 EPOCH 9 done: loss 0.0172 - lr: 0.000004
2023-10-17 17:50:29,406 DEV : loss 0.13158701360225677 - f1-score (micro avg) 0.7722
2023-10-17 17:50:29,412 saving best model
2023-10-17 17:50:29,850 ----------------------------------------------------------------------------------------------------
2023-10-17 17:50:31,171 epoch 10 - iter 14/146 - loss 0.01451792 - time (sec): 1.32 - samples/sec: 3077.53 - lr: 0.000003 - momentum: 0.000000
2023-10-17 17:50:32,784 epoch 10 - iter 28/146 - loss 0.01705737 - time (sec): 2.93 - samples/sec: 2778.86 - lr: 0.000003 - momentum: 0.000000
2023-10-17 17:50:34,412 epoch 10 - iter 42/146 - loss 0.01983646 - time (sec): 4.56 - samples/sec: 2737.42 - lr: 0.000003 - momentum: 0.000000
2023-10-17 17:50:35,780 epoch 10 - iter 56/146 - loss 0.01703870 - time (sec): 5.92 - samples/sec: 2850.19 - lr: 0.000002 - momentum: 0.000000
2023-10-17 17:50:37,301 epoch 10 - iter 70/146 - loss 0.01574640 - time (sec): 7.45 - samples/sec: 2902.74 - lr: 0.000002 - momentum: 0.000000
2023-10-17 17:50:38,843 epoch 10 - iter 84/146 - loss 0.01431043 - time (sec): 8.99 - samples/sec: 2883.94 - lr: 0.000002 - momentum: 0.000000
2023-10-17 17:50:40,568 epoch 10 - iter 98/146 - loss 0.01549304 - time (sec): 10.71 - samples/sec: 2825.21 - lr: 0.000001 - momentum: 0.000000
2023-10-17 17:50:42,069 epoch 10 - iter 112/146 - loss 0.01473409 - time (sec): 12.21 - samples/sec: 2846.18 - lr: 0.000001 - momentum: 0.000000
2023-10-17 17:50:43,565 epoch 10 - iter 126/146 - loss 0.01434044 - time (sec): 13.71 - samples/sec: 2843.59 - lr: 0.000001 - momentum: 0.000000
2023-10-17 17:50:44,833 epoch 10 - iter 140/146 - loss 0.01637173 - time (sec): 14.98 - samples/sec: 2850.68 - lr: 0.000000 - momentum: 0.000000
2023-10-17 17:50:45,380 ----------------------------------------------------------------------------------------------------
2023-10-17 17:50:45,380 EPOCH 10 done: loss 0.0159 - lr: 0.000000
2023-10-17 17:50:46,646 DEV : loss 0.13107363879680634 - f1-score (micro avg) 0.7766
2023-10-17 17:50:46,651 saving best model
2023-10-17 17:50:47,431 ----------------------------------------------------------------------------------------------------
2023-10-17 17:50:47,432 Loading model from best epoch ...
2023-10-17 17:50:48,810 SequenceTagger predicts: Dictionary with 17 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, S-HumanProd, B-HumanProd, E-HumanProd, I-HumanProd
2023-10-17 17:50:51,565
Results:
- F-score (micro) 0.7693
- F-score (macro) 0.7008
- Accuracy 0.6408
By class:
precision recall f1-score support
PER 0.8154 0.8506 0.8326 348
LOC 0.6588 0.8582 0.7454 261
ORG 0.4898 0.4615 0.4752 52
HumanProd 0.6923 0.8182 0.7500 22
micro avg 0.7224 0.8228 0.7693 683
macro avg 0.6641 0.7471 0.7008 683
weighted avg 0.7268 0.8228 0.7694 683
2023-10-17 17:50:51,565 ----------------------------------------------------------------------------------------------------
|