File size: 23,927 Bytes
aa4056d |
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 18:01:41,762 ----------------------------------------------------------------------------------------------------
2023-10-17 18:01:41,763 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 18:01:41,763 ----------------------------------------------------------------------------------------------------
2023-10-17 18:01:41,763 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 18:01:41,763 ----------------------------------------------------------------------------------------------------
2023-10-17 18:01:41,763 Train: 1166 sentences
2023-10-17 18:01:41,763 (train_with_dev=False, train_with_test=False)
2023-10-17 18:01:41,763 ----------------------------------------------------------------------------------------------------
2023-10-17 18:01:41,763 Training Params:
2023-10-17 18:01:41,763 - learning_rate: "3e-05"
2023-10-17 18:01:41,763 - mini_batch_size: "8"
2023-10-17 18:01:41,763 - max_epochs: "10"
2023-10-17 18:01:41,763 - shuffle: "True"
2023-10-17 18:01:41,763 ----------------------------------------------------------------------------------------------------
2023-10-17 18:01:41,763 Plugins:
2023-10-17 18:01:41,763 - TensorboardLogger
2023-10-17 18:01:41,763 - LinearScheduler | warmup_fraction: '0.1'
2023-10-17 18:01:41,763 ----------------------------------------------------------------------------------------------------
2023-10-17 18:01:41,764 Final evaluation on model from best epoch (best-model.pt)
2023-10-17 18:01:41,764 - metric: "('micro avg', 'f1-score')"
2023-10-17 18:01:41,764 ----------------------------------------------------------------------------------------------------
2023-10-17 18:01:41,764 Computation:
2023-10-17 18:01:41,764 - compute on device: cuda:0
2023-10-17 18:01:41,764 - embedding storage: none
2023-10-17 18:01:41,764 ----------------------------------------------------------------------------------------------------
2023-10-17 18:01:41,764 Model training base path: "hmbench-newseye/fi-hmteams/teams-base-historic-multilingual-discriminator-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-3"
2023-10-17 18:01:41,764 ----------------------------------------------------------------------------------------------------
2023-10-17 18:01:41,764 ----------------------------------------------------------------------------------------------------
2023-10-17 18:01:41,764 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-17 18:01:43,042 epoch 1 - iter 14/146 - loss 3.53043795 - time (sec): 1.28 - samples/sec: 2887.22 - lr: 0.000003 - momentum: 0.000000
2023-10-17 18:01:44,750 epoch 1 - iter 28/146 - loss 3.22220354 - time (sec): 2.99 - samples/sec: 2903.33 - lr: 0.000006 - momentum: 0.000000
2023-10-17 18:01:46,108 epoch 1 - iter 42/146 - loss 2.82585575 - time (sec): 4.34 - samples/sec: 2974.54 - lr: 0.000008 - momentum: 0.000000
2023-10-17 18:01:47,961 epoch 1 - iter 56/146 - loss 2.33875546 - time (sec): 6.20 - samples/sec: 2863.59 - lr: 0.000011 - momentum: 0.000000
2023-10-17 18:01:49,464 epoch 1 - iter 70/146 - loss 1.98561751 - time (sec): 7.70 - samples/sec: 2859.02 - lr: 0.000014 - momentum: 0.000000
2023-10-17 18:01:51,151 epoch 1 - iter 84/146 - loss 1.73604633 - time (sec): 9.39 - samples/sec: 2822.89 - lr: 0.000017 - momentum: 0.000000
2023-10-17 18:01:52,455 epoch 1 - iter 98/146 - loss 1.56244089 - time (sec): 10.69 - samples/sec: 2869.67 - lr: 0.000020 - momentum: 0.000000
2023-10-17 18:01:53,840 epoch 1 - iter 112/146 - loss 1.45136493 - time (sec): 12.07 - samples/sec: 2856.69 - lr: 0.000023 - momentum: 0.000000
2023-10-17 18:01:55,351 epoch 1 - iter 126/146 - loss 1.32971185 - time (sec): 13.59 - samples/sec: 2849.01 - lr: 0.000026 - momentum: 0.000000
2023-10-17 18:01:56,627 epoch 1 - iter 140/146 - loss 1.24009769 - time (sec): 14.86 - samples/sec: 2874.39 - lr: 0.000029 - momentum: 0.000000
2023-10-17 18:01:57,281 ----------------------------------------------------------------------------------------------------
2023-10-17 18:01:57,282 EPOCH 1 done: loss 1.2042 - lr: 0.000029
2023-10-17 18:01:58,453 DEV : loss 0.2369639128446579 - f1-score (micro avg) 0.3429
2023-10-17 18:01:58,459 saving best model
2023-10-17 18:01:58,784 ----------------------------------------------------------------------------------------------------
2023-10-17 18:02:00,122 epoch 2 - iter 14/146 - loss 0.29832645 - time (sec): 1.34 - samples/sec: 3233.20 - lr: 0.000030 - momentum: 0.000000
2023-10-17 18:02:01,215 epoch 2 - iter 28/146 - loss 0.27090210 - time (sec): 2.43 - samples/sec: 3144.62 - lr: 0.000029 - momentum: 0.000000
2023-10-17 18:02:02,734 epoch 2 - iter 42/146 - loss 0.24774915 - time (sec): 3.95 - samples/sec: 3163.37 - lr: 0.000029 - momentum: 0.000000
2023-10-17 18:02:04,573 epoch 2 - iter 56/146 - loss 0.24935139 - time (sec): 5.79 - samples/sec: 3031.59 - lr: 0.000029 - momentum: 0.000000
2023-10-17 18:02:05,827 epoch 2 - iter 70/146 - loss 0.24406299 - time (sec): 7.04 - samples/sec: 3021.97 - lr: 0.000028 - momentum: 0.000000
2023-10-17 18:02:07,349 epoch 2 - iter 84/146 - loss 0.23566930 - time (sec): 8.56 - samples/sec: 2980.19 - lr: 0.000028 - momentum: 0.000000
2023-10-17 18:02:08,992 epoch 2 - iter 98/146 - loss 0.22785994 - time (sec): 10.21 - samples/sec: 2986.86 - lr: 0.000028 - momentum: 0.000000
2023-10-17 18:02:10,368 epoch 2 - iter 112/146 - loss 0.22262977 - time (sec): 11.58 - samples/sec: 2997.58 - lr: 0.000027 - momentum: 0.000000
2023-10-17 18:02:11,870 epoch 2 - iter 126/146 - loss 0.22621003 - time (sec): 13.08 - samples/sec: 3001.66 - lr: 0.000027 - momentum: 0.000000
2023-10-17 18:02:13,129 epoch 2 - iter 140/146 - loss 0.22404336 - time (sec): 14.34 - samples/sec: 2985.96 - lr: 0.000027 - momentum: 0.000000
2023-10-17 18:02:13,639 ----------------------------------------------------------------------------------------------------
2023-10-17 18:02:13,639 EPOCH 2 done: loss 0.2206 - lr: 0.000027
2023-10-17 18:02:14,920 DEV : loss 0.1299794614315033 - f1-score (micro avg) 0.5919
2023-10-17 18:02:14,926 saving best model
2023-10-17 18:02:15,371 ----------------------------------------------------------------------------------------------------
2023-10-17 18:02:16,875 epoch 3 - iter 14/146 - loss 0.12717687 - time (sec): 1.50 - samples/sec: 3121.96 - lr: 0.000026 - momentum: 0.000000
2023-10-17 18:02:18,340 epoch 3 - iter 28/146 - loss 0.13016870 - time (sec): 2.97 - samples/sec: 3101.19 - lr: 0.000026 - momentum: 0.000000
2023-10-17 18:02:19,758 epoch 3 - iter 42/146 - loss 0.14516907 - time (sec): 4.39 - samples/sec: 3022.70 - lr: 0.000026 - momentum: 0.000000
2023-10-17 18:02:21,273 epoch 3 - iter 56/146 - loss 0.13349387 - time (sec): 5.90 - samples/sec: 2944.83 - lr: 0.000025 - momentum: 0.000000
2023-10-17 18:02:22,750 epoch 3 - iter 70/146 - loss 0.12864265 - time (sec): 7.38 - samples/sec: 2953.89 - lr: 0.000025 - momentum: 0.000000
2023-10-17 18:02:24,312 epoch 3 - iter 84/146 - loss 0.13230498 - time (sec): 8.94 - samples/sec: 2942.26 - lr: 0.000025 - momentum: 0.000000
2023-10-17 18:02:25,476 epoch 3 - iter 98/146 - loss 0.12924129 - time (sec): 10.10 - samples/sec: 2969.36 - lr: 0.000024 - momentum: 0.000000
2023-10-17 18:02:27,038 epoch 3 - iter 112/146 - loss 0.12219386 - time (sec): 11.66 - samples/sec: 2977.73 - lr: 0.000024 - momentum: 0.000000
2023-10-17 18:02:28,564 epoch 3 - iter 126/146 - loss 0.12044637 - time (sec): 13.19 - samples/sec: 2959.39 - lr: 0.000024 - momentum: 0.000000
2023-10-17 18:02:30,111 epoch 3 - iter 140/146 - loss 0.12007643 - time (sec): 14.74 - samples/sec: 2913.51 - lr: 0.000024 - momentum: 0.000000
2023-10-17 18:02:30,652 ----------------------------------------------------------------------------------------------------
2023-10-17 18:02:30,652 EPOCH 3 done: loss 0.1233 - lr: 0.000024
2023-10-17 18:02:32,289 DEV : loss 0.10894083976745605 - f1-score (micro avg) 0.7187
2023-10-17 18:02:32,299 saving best model
2023-10-17 18:02:32,723 ----------------------------------------------------------------------------------------------------
2023-10-17 18:02:34,231 epoch 4 - iter 14/146 - loss 0.07416896 - time (sec): 1.51 - samples/sec: 3010.11 - lr: 0.000023 - momentum: 0.000000
2023-10-17 18:02:35,856 epoch 4 - iter 28/146 - loss 0.07141417 - time (sec): 3.13 - samples/sec: 2907.94 - lr: 0.000023 - momentum: 0.000000
2023-10-17 18:02:37,152 epoch 4 - iter 42/146 - loss 0.08970989 - time (sec): 4.43 - samples/sec: 2908.80 - lr: 0.000022 - momentum: 0.000000
2023-10-17 18:02:38,867 epoch 4 - iter 56/146 - loss 0.07963099 - time (sec): 6.14 - samples/sec: 2874.00 - lr: 0.000022 - momentum: 0.000000
2023-10-17 18:02:40,463 epoch 4 - iter 70/146 - loss 0.07752524 - time (sec): 7.74 - samples/sec: 2873.54 - lr: 0.000022 - momentum: 0.000000
2023-10-17 18:02:41,955 epoch 4 - iter 84/146 - loss 0.08042752 - time (sec): 9.23 - samples/sec: 2870.29 - lr: 0.000021 - momentum: 0.000000
2023-10-17 18:02:43,558 epoch 4 - iter 98/146 - loss 0.08106679 - time (sec): 10.83 - samples/sec: 2818.94 - lr: 0.000021 - momentum: 0.000000
2023-10-17 18:02:45,073 epoch 4 - iter 112/146 - loss 0.08397046 - time (sec): 12.35 - samples/sec: 2831.20 - lr: 0.000021 - momentum: 0.000000
2023-10-17 18:02:46,566 epoch 4 - iter 126/146 - loss 0.08434407 - time (sec): 13.84 - samples/sec: 2828.54 - lr: 0.000021 - momentum: 0.000000
2023-10-17 18:02:47,841 epoch 4 - iter 140/146 - loss 0.08375380 - time (sec): 15.12 - samples/sec: 2812.99 - lr: 0.000020 - momentum: 0.000000
2023-10-17 18:02:48,440 ----------------------------------------------------------------------------------------------------
2023-10-17 18:02:48,440 EPOCH 4 done: loss 0.0837 - lr: 0.000020
2023-10-17 18:02:49,738 DEV : loss 0.11354158073663712 - f1-score (micro avg) 0.7364
2023-10-17 18:02:49,744 saving best model
2023-10-17 18:02:50,202 ----------------------------------------------------------------------------------------------------
2023-10-17 18:02:51,702 epoch 5 - iter 14/146 - loss 0.05382494 - time (sec): 1.50 - samples/sec: 2810.34 - lr: 0.000020 - momentum: 0.000000
2023-10-17 18:02:53,009 epoch 5 - iter 28/146 - loss 0.06071119 - time (sec): 2.80 - samples/sec: 2969.28 - lr: 0.000019 - momentum: 0.000000
2023-10-17 18:02:54,527 epoch 5 - iter 42/146 - loss 0.05523519 - time (sec): 4.32 - samples/sec: 3056.07 - lr: 0.000019 - momentum: 0.000000
2023-10-17 18:02:55,986 epoch 5 - iter 56/146 - loss 0.05750211 - time (sec): 5.78 - samples/sec: 2996.72 - lr: 0.000019 - momentum: 0.000000
2023-10-17 18:02:57,546 epoch 5 - iter 70/146 - loss 0.06410589 - time (sec): 7.34 - samples/sec: 2872.91 - lr: 0.000018 - momentum: 0.000000
2023-10-17 18:02:59,083 epoch 5 - iter 84/146 - loss 0.06215775 - time (sec): 8.88 - samples/sec: 2907.99 - lr: 0.000018 - momentum: 0.000000
2023-10-17 18:03:00,391 epoch 5 - iter 98/146 - loss 0.05920684 - time (sec): 10.18 - samples/sec: 2912.85 - lr: 0.000018 - momentum: 0.000000
2023-10-17 18:03:01,906 epoch 5 - iter 112/146 - loss 0.05729003 - time (sec): 11.70 - samples/sec: 2882.89 - lr: 0.000018 - momentum: 0.000000
2023-10-17 18:03:03,453 epoch 5 - iter 126/146 - loss 0.05501850 - time (sec): 13.25 - samples/sec: 2905.99 - lr: 0.000017 - momentum: 0.000000
2023-10-17 18:03:05,100 epoch 5 - iter 140/146 - loss 0.05545585 - time (sec): 14.89 - samples/sec: 2890.13 - lr: 0.000017 - momentum: 0.000000
2023-10-17 18:03:05,589 ----------------------------------------------------------------------------------------------------
2023-10-17 18:03:05,589 EPOCH 5 done: loss 0.0549 - lr: 0.000017
2023-10-17 18:03:06,918 DEV : loss 0.11634857952594757 - f1-score (micro avg) 0.7364
2023-10-17 18:03:06,925 saving best model
2023-10-17 18:03:07,379 ----------------------------------------------------------------------------------------------------
2023-10-17 18:03:08,799 epoch 6 - iter 14/146 - loss 0.04238142 - time (sec): 1.41 - samples/sec: 2913.49 - lr: 0.000016 - momentum: 0.000000
2023-10-17 18:03:10,367 epoch 6 - iter 28/146 - loss 0.03655978 - time (sec): 2.98 - samples/sec: 2978.87 - lr: 0.000016 - momentum: 0.000000
2023-10-17 18:03:11,678 epoch 6 - iter 42/146 - loss 0.04211034 - time (sec): 4.29 - samples/sec: 2890.83 - lr: 0.000016 - momentum: 0.000000
2023-10-17 18:03:13,075 epoch 6 - iter 56/146 - loss 0.04107637 - time (sec): 5.69 - samples/sec: 2815.81 - lr: 0.000015 - momentum: 0.000000
2023-10-17 18:03:14,598 epoch 6 - iter 70/146 - loss 0.03835445 - time (sec): 7.21 - samples/sec: 2834.14 - lr: 0.000015 - momentum: 0.000000
2023-10-17 18:03:16,185 epoch 6 - iter 84/146 - loss 0.04034492 - time (sec): 8.80 - samples/sec: 2893.99 - lr: 0.000015 - momentum: 0.000000
2023-10-17 18:03:17,290 epoch 6 - iter 98/146 - loss 0.03978206 - time (sec): 9.90 - samples/sec: 2918.91 - lr: 0.000015 - momentum: 0.000000
2023-10-17 18:03:18,683 epoch 6 - iter 112/146 - loss 0.04031278 - time (sec): 11.30 - samples/sec: 2921.99 - lr: 0.000014 - momentum: 0.000000
2023-10-17 18:03:20,095 epoch 6 - iter 126/146 - loss 0.04005854 - time (sec): 12.71 - samples/sec: 2955.67 - lr: 0.000014 - momentum: 0.000000
2023-10-17 18:03:21,608 epoch 6 - iter 140/146 - loss 0.03903056 - time (sec): 14.22 - samples/sec: 2987.59 - lr: 0.000014 - momentum: 0.000000
2023-10-17 18:03:22,326 ----------------------------------------------------------------------------------------------------
2023-10-17 18:03:22,326 EPOCH 6 done: loss 0.0384 - lr: 0.000014
2023-10-17 18:03:23,864 DEV : loss 0.11619787663221359 - f1-score (micro avg) 0.7478
2023-10-17 18:03:23,869 saving best model
2023-10-17 18:03:24,312 ----------------------------------------------------------------------------------------------------
2023-10-17 18:03:25,733 epoch 7 - iter 14/146 - loss 0.03971526 - time (sec): 1.41 - samples/sec: 2868.48 - lr: 0.000013 - momentum: 0.000000
2023-10-17 18:03:26,974 epoch 7 - iter 28/146 - loss 0.04113722 - time (sec): 2.65 - samples/sec: 2856.67 - lr: 0.000013 - momentum: 0.000000
2023-10-17 18:03:28,350 epoch 7 - iter 42/146 - loss 0.03581246 - time (sec): 4.03 - samples/sec: 2894.28 - lr: 0.000012 - momentum: 0.000000
2023-10-17 18:03:29,725 epoch 7 - iter 56/146 - loss 0.03125811 - time (sec): 5.41 - samples/sec: 2955.57 - lr: 0.000012 - momentum: 0.000000
2023-10-17 18:03:31,234 epoch 7 - iter 70/146 - loss 0.03390483 - time (sec): 6.92 - samples/sec: 2998.62 - lr: 0.000012 - momentum: 0.000000
2023-10-17 18:03:32,727 epoch 7 - iter 84/146 - loss 0.03434154 - time (sec): 8.41 - samples/sec: 2922.72 - lr: 0.000012 - momentum: 0.000000
2023-10-17 18:03:34,202 epoch 7 - iter 98/146 - loss 0.03222323 - time (sec): 9.88 - samples/sec: 2921.38 - lr: 0.000011 - momentum: 0.000000
2023-10-17 18:03:35,723 epoch 7 - iter 112/146 - loss 0.03097942 - time (sec): 11.40 - samples/sec: 2894.68 - lr: 0.000011 - momentum: 0.000000
2023-10-17 18:03:37,396 epoch 7 - iter 126/146 - loss 0.03133453 - time (sec): 13.08 - samples/sec: 2874.13 - lr: 0.000011 - momentum: 0.000000
2023-10-17 18:03:38,840 epoch 7 - iter 140/146 - loss 0.02975401 - time (sec): 14.52 - samples/sec: 2914.63 - lr: 0.000010 - momentum: 0.000000
2023-10-17 18:03:39,546 ----------------------------------------------------------------------------------------------------
2023-10-17 18:03:39,547 EPOCH 7 done: loss 0.0289 - lr: 0.000010
2023-10-17 18:03:40,808 DEV : loss 0.12387975305318832 - f1-score (micro avg) 0.7793
2023-10-17 18:03:40,813 saving best model
2023-10-17 18:03:41,234 ----------------------------------------------------------------------------------------------------
2023-10-17 18:03:42,601 epoch 8 - iter 14/146 - loss 0.02000203 - time (sec): 1.37 - samples/sec: 3044.44 - lr: 0.000010 - momentum: 0.000000
2023-10-17 18:03:44,171 epoch 8 - iter 28/146 - loss 0.01956968 - time (sec): 2.94 - samples/sec: 2897.32 - lr: 0.000009 - momentum: 0.000000
2023-10-17 18:03:45,590 epoch 8 - iter 42/146 - loss 0.01928448 - time (sec): 4.35 - samples/sec: 2916.95 - lr: 0.000009 - momentum: 0.000000
2023-10-17 18:03:47,084 epoch 8 - iter 56/146 - loss 0.02405329 - time (sec): 5.85 - samples/sec: 2973.71 - lr: 0.000009 - momentum: 0.000000
2023-10-17 18:03:48,535 epoch 8 - iter 70/146 - loss 0.02559984 - time (sec): 7.30 - samples/sec: 2977.54 - lr: 0.000009 - momentum: 0.000000
2023-10-17 18:03:49,980 epoch 8 - iter 84/146 - loss 0.02430293 - time (sec): 8.74 - samples/sec: 2997.30 - lr: 0.000008 - momentum: 0.000000
2023-10-17 18:03:51,652 epoch 8 - iter 98/146 - loss 0.02276596 - time (sec): 10.42 - samples/sec: 2971.35 - lr: 0.000008 - momentum: 0.000000
2023-10-17 18:03:52,912 epoch 8 - iter 112/146 - loss 0.02276090 - time (sec): 11.68 - samples/sec: 2946.39 - lr: 0.000008 - momentum: 0.000000
2023-10-17 18:03:54,513 epoch 8 - iter 126/146 - loss 0.02259700 - time (sec): 13.28 - samples/sec: 2960.07 - lr: 0.000007 - momentum: 0.000000
2023-10-17 18:03:55,891 epoch 8 - iter 140/146 - loss 0.02211640 - time (sec): 14.66 - samples/sec: 2938.42 - lr: 0.000007 - momentum: 0.000000
2023-10-17 18:03:56,387 ----------------------------------------------------------------------------------------------------
2023-10-17 18:03:56,387 EPOCH 8 done: loss 0.0219 - lr: 0.000007
2023-10-17 18:03:57,620 DEV : loss 0.12991590797901154 - f1-score (micro avg) 0.7723
2023-10-17 18:03:57,625 ----------------------------------------------------------------------------------------------------
2023-10-17 18:03:58,998 epoch 9 - iter 14/146 - loss 0.02121112 - time (sec): 1.37 - samples/sec: 2778.46 - lr: 0.000006 - momentum: 0.000000
2023-10-17 18:04:00,753 epoch 9 - iter 28/146 - loss 0.02062558 - time (sec): 3.13 - samples/sec: 2796.60 - lr: 0.000006 - momentum: 0.000000
2023-10-17 18:04:02,551 epoch 9 - iter 42/146 - loss 0.02599310 - time (sec): 4.92 - samples/sec: 2863.48 - lr: 0.000006 - momentum: 0.000000
2023-10-17 18:04:03,939 epoch 9 - iter 56/146 - loss 0.02190832 - time (sec): 6.31 - samples/sec: 2853.47 - lr: 0.000006 - momentum: 0.000000
2023-10-17 18:04:05,179 epoch 9 - iter 70/146 - loss 0.02164107 - time (sec): 7.55 - samples/sec: 2878.04 - lr: 0.000005 - momentum: 0.000000
2023-10-17 18:04:06,666 epoch 9 - iter 84/146 - loss 0.02212824 - time (sec): 9.04 - samples/sec: 2896.08 - lr: 0.000005 - momentum: 0.000000
2023-10-17 18:04:08,036 epoch 9 - iter 98/146 - loss 0.02065961 - time (sec): 10.41 - samples/sec: 2874.83 - lr: 0.000005 - momentum: 0.000000
2023-10-17 18:04:09,463 epoch 9 - iter 112/146 - loss 0.02049381 - time (sec): 11.84 - samples/sec: 2930.78 - lr: 0.000004 - momentum: 0.000000
2023-10-17 18:04:10,874 epoch 9 - iter 126/146 - loss 0.01964867 - time (sec): 13.25 - samples/sec: 2891.82 - lr: 0.000004 - momentum: 0.000000
2023-10-17 18:04:12,171 epoch 9 - iter 140/146 - loss 0.01888946 - time (sec): 14.54 - samples/sec: 2895.64 - lr: 0.000004 - momentum: 0.000000
2023-10-17 18:04:12,832 ----------------------------------------------------------------------------------------------------
2023-10-17 18:04:12,832 EPOCH 9 done: loss 0.0180 - lr: 0.000004
2023-10-17 18:04:14,602 DEV : loss 0.1360742151737213 - f1-score (micro avg) 0.7604
2023-10-17 18:04:14,609 ----------------------------------------------------------------------------------------------------
2023-10-17 18:04:16,465 epoch 10 - iter 14/146 - loss 0.02531276 - time (sec): 1.85 - samples/sec: 2708.78 - lr: 0.000003 - momentum: 0.000000
2023-10-17 18:04:18,092 epoch 10 - iter 28/146 - loss 0.01920590 - time (sec): 3.48 - samples/sec: 2742.03 - lr: 0.000003 - momentum: 0.000000
2023-10-17 18:04:19,521 epoch 10 - iter 42/146 - loss 0.01738091 - time (sec): 4.91 - samples/sec: 2769.85 - lr: 0.000003 - momentum: 0.000000
2023-10-17 18:04:20,878 epoch 10 - iter 56/146 - loss 0.01529737 - time (sec): 6.27 - samples/sec: 2803.00 - lr: 0.000002 - momentum: 0.000000
2023-10-17 18:04:22,307 epoch 10 - iter 70/146 - loss 0.01440336 - time (sec): 7.70 - samples/sec: 2783.17 - lr: 0.000002 - momentum: 0.000000
2023-10-17 18:04:23,876 epoch 10 - iter 84/146 - loss 0.01454689 - time (sec): 9.27 - samples/sec: 2737.19 - lr: 0.000002 - momentum: 0.000000
2023-10-17 18:04:25,406 epoch 10 - iter 98/146 - loss 0.01355592 - time (sec): 10.80 - samples/sec: 2754.28 - lr: 0.000001 - momentum: 0.000000
2023-10-17 18:04:27,080 epoch 10 - iter 112/146 - loss 0.01381817 - time (sec): 12.47 - samples/sec: 2767.65 - lr: 0.000001 - momentum: 0.000000
2023-10-17 18:04:28,326 epoch 10 - iter 126/146 - loss 0.01558305 - time (sec): 13.72 - samples/sec: 2798.28 - lr: 0.000001 - momentum: 0.000000
2023-10-17 18:04:29,771 epoch 10 - iter 140/146 - loss 0.01583718 - time (sec): 15.16 - samples/sec: 2800.43 - lr: 0.000000 - momentum: 0.000000
2023-10-17 18:04:30,590 ----------------------------------------------------------------------------------------------------
2023-10-17 18:04:30,591 EPOCH 10 done: loss 0.0152 - lr: 0.000000
2023-10-17 18:04:31,868 DEV : loss 0.1429719179868698 - f1-score (micro avg) 0.7533
2023-10-17 18:04:32,223 ----------------------------------------------------------------------------------------------------
2023-10-17 18:04:32,224 Loading model from best epoch ...
2023-10-17 18:04:33,624 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 18:04:36,114
Results:
- F-score (micro) 0.7598
- F-score (macro) 0.6737
- Accuracy 0.6318
By class:
precision recall f1-score support
PER 0.8260 0.8592 0.8423 348
LOC 0.6483 0.8123 0.7211 261
ORG 0.4250 0.3269 0.3696 52
HumanProd 0.8000 0.7273 0.7619 22
micro avg 0.7263 0.7965 0.7598 683
macro avg 0.6748 0.6814 0.6737 683
weighted avg 0.7267 0.7965 0.7574 683
2023-10-17 18:04:36,114 ----------------------------------------------------------------------------------------------------
|