File size: 24,105 Bytes
3e21217 |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 |
2023-10-23 15:54:49,408 ----------------------------------------------------------------------------------------------------
2023-10-23 15:54:49,409 Model: "SequenceTagger(
(embeddings): TransformerWordEmbeddings(
(model): BertModel(
(embeddings): BertEmbeddings(
(word_embeddings): Embedding(64001, 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): BertEncoder(
(layer): ModuleList(
(0-11): 12 x BertLayer(
(attention): BertAttention(
(self): BertSelfAttention(
(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): BertSelfOutput(
(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): BertIntermediate(
(dense): Linear(in_features=768, out_features=3072, bias=True)
(intermediate_act_fn): GELUActivation()
)
(output): BertOutput(
(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)
)
)
)
)
(pooler): BertPooler(
(dense): Linear(in_features=768, out_features=768, bias=True)
(activation): Tanh()
)
)
)
(locked_dropout): LockedDropout(p=0.5)
(linear): Linear(in_features=768, out_features=25, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-23 15:54:49,409 ----------------------------------------------------------------------------------------------------
2023-10-23 15:54:49,409 MultiCorpus: 1100 train + 206 dev + 240 test sentences
- NER_HIPE_2022 Corpus: 1100 train + 206 dev + 240 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/ajmc/de/with_doc_seperator
2023-10-23 15:54:49,409 ----------------------------------------------------------------------------------------------------
2023-10-23 15:54:49,409 Train: 1100 sentences
2023-10-23 15:54:49,409 (train_with_dev=False, train_with_test=False)
2023-10-23 15:54:49,409 ----------------------------------------------------------------------------------------------------
2023-10-23 15:54:49,409 Training Params:
2023-10-23 15:54:49,409 - learning_rate: "3e-05"
2023-10-23 15:54:49,409 - mini_batch_size: "8"
2023-10-23 15:54:49,409 - max_epochs: "10"
2023-10-23 15:54:49,410 - shuffle: "True"
2023-10-23 15:54:49,410 ----------------------------------------------------------------------------------------------------
2023-10-23 15:54:49,410 Plugins:
2023-10-23 15:54:49,410 - TensorboardLogger
2023-10-23 15:54:49,410 - LinearScheduler | warmup_fraction: '0.1'
2023-10-23 15:54:49,410 ----------------------------------------------------------------------------------------------------
2023-10-23 15:54:49,410 Final evaluation on model from best epoch (best-model.pt)
2023-10-23 15:54:49,410 - metric: "('micro avg', 'f1-score')"
2023-10-23 15:54:49,410 ----------------------------------------------------------------------------------------------------
2023-10-23 15:54:49,410 Computation:
2023-10-23 15:54:49,410 - compute on device: cuda:0
2023-10-23 15:54:49,410 - embedding storage: none
2023-10-23 15:54:49,410 ----------------------------------------------------------------------------------------------------
2023-10-23 15:54:49,410 Model training base path: "hmbench-ajmc/de-dbmdz/bert-base-historic-multilingual-64k-td-cased-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-5"
2023-10-23 15:54:49,410 ----------------------------------------------------------------------------------------------------
2023-10-23 15:54:49,410 ----------------------------------------------------------------------------------------------------
2023-10-23 15:54:49,410 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-23 15:54:50,141 epoch 1 - iter 13/138 - loss 3.00405917 - time (sec): 0.73 - samples/sec: 2373.99 - lr: 0.000003 - momentum: 0.000000
2023-10-23 15:54:50,875 epoch 1 - iter 26/138 - loss 2.56287802 - time (sec): 1.46 - samples/sec: 2693.60 - lr: 0.000005 - momentum: 0.000000
2023-10-23 15:54:51,628 epoch 1 - iter 39/138 - loss 2.05445276 - time (sec): 2.22 - samples/sec: 2813.21 - lr: 0.000008 - momentum: 0.000000
2023-10-23 15:54:52,397 epoch 1 - iter 52/138 - loss 1.76142833 - time (sec): 2.99 - samples/sec: 2810.56 - lr: 0.000011 - momentum: 0.000000
2023-10-23 15:54:53,182 epoch 1 - iter 65/138 - loss 1.54424671 - time (sec): 3.77 - samples/sec: 2837.08 - lr: 0.000014 - momentum: 0.000000
2023-10-23 15:54:53,971 epoch 1 - iter 78/138 - loss 1.35513027 - time (sec): 4.56 - samples/sec: 2841.14 - lr: 0.000017 - momentum: 0.000000
2023-10-23 15:54:54,733 epoch 1 - iter 91/138 - loss 1.22250128 - time (sec): 5.32 - samples/sec: 2819.82 - lr: 0.000020 - momentum: 0.000000
2023-10-23 15:54:55,485 epoch 1 - iter 104/138 - loss 1.11598652 - time (sec): 6.07 - samples/sec: 2841.53 - lr: 0.000022 - momentum: 0.000000
2023-10-23 15:54:56,239 epoch 1 - iter 117/138 - loss 1.02403952 - time (sec): 6.83 - samples/sec: 2874.76 - lr: 0.000025 - momentum: 0.000000
2023-10-23 15:54:56,997 epoch 1 - iter 130/138 - loss 0.95915259 - time (sec): 7.59 - samples/sec: 2853.35 - lr: 0.000028 - momentum: 0.000000
2023-10-23 15:54:57,457 ----------------------------------------------------------------------------------------------------
2023-10-23 15:54:57,457 EPOCH 1 done: loss 0.9234 - lr: 0.000028
2023-10-23 15:54:58,036 DEV : loss 0.19589945673942566 - f1-score (micro avg) 0.7295
2023-10-23 15:54:58,042 saving best model
2023-10-23 15:54:58,436 ----------------------------------------------------------------------------------------------------
2023-10-23 15:54:59,187 epoch 2 - iter 13/138 - loss 0.24513018 - time (sec): 0.75 - samples/sec: 3157.13 - lr: 0.000030 - momentum: 0.000000
2023-10-23 15:54:59,947 epoch 2 - iter 26/138 - loss 0.23709938 - time (sec): 1.51 - samples/sec: 2974.25 - lr: 0.000029 - momentum: 0.000000
2023-10-23 15:55:00,669 epoch 2 - iter 39/138 - loss 0.20795711 - time (sec): 2.23 - samples/sec: 2891.60 - lr: 0.000029 - momentum: 0.000000
2023-10-23 15:55:01,392 epoch 2 - iter 52/138 - loss 0.20218652 - time (sec): 2.95 - samples/sec: 2802.42 - lr: 0.000029 - momentum: 0.000000
2023-10-23 15:55:02,109 epoch 2 - iter 65/138 - loss 0.19026323 - time (sec): 3.67 - samples/sec: 2809.17 - lr: 0.000028 - momentum: 0.000000
2023-10-23 15:55:02,835 epoch 2 - iter 78/138 - loss 0.18132874 - time (sec): 4.40 - samples/sec: 2844.07 - lr: 0.000028 - momentum: 0.000000
2023-10-23 15:55:03,562 epoch 2 - iter 91/138 - loss 0.17220623 - time (sec): 5.12 - samples/sec: 2839.33 - lr: 0.000028 - momentum: 0.000000
2023-10-23 15:55:04,321 epoch 2 - iter 104/138 - loss 0.16856554 - time (sec): 5.88 - samples/sec: 2842.82 - lr: 0.000028 - momentum: 0.000000
2023-10-23 15:55:05,080 epoch 2 - iter 117/138 - loss 0.17136679 - time (sec): 6.64 - samples/sec: 2867.72 - lr: 0.000027 - momentum: 0.000000
2023-10-23 15:55:05,831 epoch 2 - iter 130/138 - loss 0.16835947 - time (sec): 7.39 - samples/sec: 2904.80 - lr: 0.000027 - momentum: 0.000000
2023-10-23 15:55:06,309 ----------------------------------------------------------------------------------------------------
2023-10-23 15:55:06,309 EPOCH 2 done: loss 0.1688 - lr: 0.000027
2023-10-23 15:55:06,848 DEV : loss 0.1276298463344574 - f1-score (micro avg) 0.8109
2023-10-23 15:55:06,854 saving best model
2023-10-23 15:55:07,392 ----------------------------------------------------------------------------------------------------
2023-10-23 15:55:08,123 epoch 3 - iter 13/138 - loss 0.11073512 - time (sec): 0.73 - samples/sec: 3098.87 - lr: 0.000026 - momentum: 0.000000
2023-10-23 15:55:08,869 epoch 3 - iter 26/138 - loss 0.09498375 - time (sec): 1.47 - samples/sec: 3028.40 - lr: 0.000026 - momentum: 0.000000
2023-10-23 15:55:09,584 epoch 3 - iter 39/138 - loss 0.09514575 - time (sec): 2.19 - samples/sec: 2991.48 - lr: 0.000026 - momentum: 0.000000
2023-10-23 15:55:10,302 epoch 3 - iter 52/138 - loss 0.10425731 - time (sec): 2.91 - samples/sec: 3042.18 - lr: 0.000025 - momentum: 0.000000
2023-10-23 15:55:11,014 epoch 3 - iter 65/138 - loss 0.09912849 - time (sec): 3.62 - samples/sec: 3030.06 - lr: 0.000025 - momentum: 0.000000
2023-10-23 15:55:11,744 epoch 3 - iter 78/138 - loss 0.09280214 - time (sec): 4.35 - samples/sec: 3000.12 - lr: 0.000025 - momentum: 0.000000
2023-10-23 15:55:12,486 epoch 3 - iter 91/138 - loss 0.09344571 - time (sec): 5.09 - samples/sec: 2982.33 - lr: 0.000025 - momentum: 0.000000
2023-10-23 15:55:13,228 epoch 3 - iter 104/138 - loss 0.09045760 - time (sec): 5.83 - samples/sec: 2967.07 - lr: 0.000024 - momentum: 0.000000
2023-10-23 15:55:13,972 epoch 3 - iter 117/138 - loss 0.09303884 - time (sec): 6.58 - samples/sec: 2971.02 - lr: 0.000024 - momentum: 0.000000
2023-10-23 15:55:14,712 epoch 3 - iter 130/138 - loss 0.09075384 - time (sec): 7.32 - samples/sec: 2950.98 - lr: 0.000024 - momentum: 0.000000
2023-10-23 15:55:15,164 ----------------------------------------------------------------------------------------------------
2023-10-23 15:55:15,164 EPOCH 3 done: loss 0.0923 - lr: 0.000024
2023-10-23 15:55:15,702 DEV : loss 0.12066850066184998 - f1-score (micro avg) 0.8379
2023-10-23 15:55:15,708 saving best model
2023-10-23 15:55:16,258 ----------------------------------------------------------------------------------------------------
2023-10-23 15:55:16,991 epoch 4 - iter 13/138 - loss 0.11350050 - time (sec): 0.73 - samples/sec: 3101.74 - lr: 0.000023 - momentum: 0.000000
2023-10-23 15:55:17,700 epoch 4 - iter 26/138 - loss 0.08075514 - time (sec): 1.44 - samples/sec: 3158.96 - lr: 0.000023 - momentum: 0.000000
2023-10-23 15:55:18,428 epoch 4 - iter 39/138 - loss 0.07346125 - time (sec): 2.17 - samples/sec: 3008.78 - lr: 0.000022 - momentum: 0.000000
2023-10-23 15:55:19,126 epoch 4 - iter 52/138 - loss 0.06656305 - time (sec): 2.87 - samples/sec: 2975.49 - lr: 0.000022 - momentum: 0.000000
2023-10-23 15:55:19,869 epoch 4 - iter 65/138 - loss 0.06313594 - time (sec): 3.61 - samples/sec: 2992.76 - lr: 0.000022 - momentum: 0.000000
2023-10-23 15:55:20,597 epoch 4 - iter 78/138 - loss 0.06522584 - time (sec): 4.34 - samples/sec: 3016.50 - lr: 0.000021 - momentum: 0.000000
2023-10-23 15:55:21,310 epoch 4 - iter 91/138 - loss 0.06281046 - time (sec): 5.05 - samples/sec: 3010.97 - lr: 0.000021 - momentum: 0.000000
2023-10-23 15:55:22,054 epoch 4 - iter 104/138 - loss 0.06475346 - time (sec): 5.79 - samples/sec: 3005.96 - lr: 0.000021 - momentum: 0.000000
2023-10-23 15:55:22,749 epoch 4 - iter 117/138 - loss 0.06295527 - time (sec): 6.49 - samples/sec: 2979.84 - lr: 0.000021 - momentum: 0.000000
2023-10-23 15:55:23,481 epoch 4 - iter 130/138 - loss 0.06220209 - time (sec): 7.22 - samples/sec: 2944.77 - lr: 0.000020 - momentum: 0.000000
2023-10-23 15:55:23,918 ----------------------------------------------------------------------------------------------------
2023-10-23 15:55:23,918 EPOCH 4 done: loss 0.0608 - lr: 0.000020
2023-10-23 15:55:24,446 DEV : loss 0.1310967206954956 - f1-score (micro avg) 0.8609
2023-10-23 15:55:24,452 saving best model
2023-10-23 15:55:24,998 ----------------------------------------------------------------------------------------------------
2023-10-23 15:55:25,720 epoch 5 - iter 13/138 - loss 0.05355793 - time (sec): 0.72 - samples/sec: 2992.96 - lr: 0.000020 - momentum: 0.000000
2023-10-23 15:55:26,423 epoch 5 - iter 26/138 - loss 0.06962420 - time (sec): 1.42 - samples/sec: 2987.17 - lr: 0.000019 - momentum: 0.000000
2023-10-23 15:55:27,155 epoch 5 - iter 39/138 - loss 0.05818477 - time (sec): 2.16 - samples/sec: 3006.77 - lr: 0.000019 - momentum: 0.000000
2023-10-23 15:55:27,903 epoch 5 - iter 52/138 - loss 0.04671793 - time (sec): 2.90 - samples/sec: 2959.51 - lr: 0.000019 - momentum: 0.000000
2023-10-23 15:55:28,647 epoch 5 - iter 65/138 - loss 0.04604634 - time (sec): 3.65 - samples/sec: 2997.02 - lr: 0.000018 - momentum: 0.000000
2023-10-23 15:55:29,391 epoch 5 - iter 78/138 - loss 0.04370062 - time (sec): 4.39 - samples/sec: 2928.05 - lr: 0.000018 - momentum: 0.000000
2023-10-23 15:55:30,130 epoch 5 - iter 91/138 - loss 0.04395983 - time (sec): 5.13 - samples/sec: 2938.58 - lr: 0.000018 - momentum: 0.000000
2023-10-23 15:55:30,889 epoch 5 - iter 104/138 - loss 0.04812229 - time (sec): 5.89 - samples/sec: 2925.69 - lr: 0.000018 - momentum: 0.000000
2023-10-23 15:55:31,639 epoch 5 - iter 117/138 - loss 0.04842003 - time (sec): 6.64 - samples/sec: 2891.87 - lr: 0.000017 - momentum: 0.000000
2023-10-23 15:55:32,383 epoch 5 - iter 130/138 - loss 0.04784414 - time (sec): 7.38 - samples/sec: 2887.52 - lr: 0.000017 - momentum: 0.000000
2023-10-23 15:55:32,838 ----------------------------------------------------------------------------------------------------
2023-10-23 15:55:32,838 EPOCH 5 done: loss 0.0472 - lr: 0.000017
2023-10-23 15:55:33,377 DEV : loss 0.13556700944900513 - f1-score (micro avg) 0.881
2023-10-23 15:55:33,383 saving best model
2023-10-23 15:55:33,924 ----------------------------------------------------------------------------------------------------
2023-10-23 15:55:34,699 epoch 6 - iter 13/138 - loss 0.00447347 - time (sec): 0.77 - samples/sec: 3105.62 - lr: 0.000016 - momentum: 0.000000
2023-10-23 15:55:35,433 epoch 6 - iter 26/138 - loss 0.03663128 - time (sec): 1.51 - samples/sec: 2857.73 - lr: 0.000016 - momentum: 0.000000
2023-10-23 15:55:36,190 epoch 6 - iter 39/138 - loss 0.02970269 - time (sec): 2.26 - samples/sec: 2836.41 - lr: 0.000016 - momentum: 0.000000
2023-10-23 15:55:36,929 epoch 6 - iter 52/138 - loss 0.03038221 - time (sec): 3.00 - samples/sec: 2898.43 - lr: 0.000015 - momentum: 0.000000
2023-10-23 15:55:37,666 epoch 6 - iter 65/138 - loss 0.02843939 - time (sec): 3.74 - samples/sec: 2877.33 - lr: 0.000015 - momentum: 0.000000
2023-10-23 15:55:38,403 epoch 6 - iter 78/138 - loss 0.03163609 - time (sec): 4.48 - samples/sec: 2867.41 - lr: 0.000015 - momentum: 0.000000
2023-10-23 15:55:39,145 epoch 6 - iter 91/138 - loss 0.03089174 - time (sec): 5.22 - samples/sec: 2838.67 - lr: 0.000015 - momentum: 0.000000
2023-10-23 15:55:39,889 epoch 6 - iter 104/138 - loss 0.03317619 - time (sec): 5.96 - samples/sec: 2839.11 - lr: 0.000014 - momentum: 0.000000
2023-10-23 15:55:40,628 epoch 6 - iter 117/138 - loss 0.03642630 - time (sec): 6.70 - samples/sec: 2865.20 - lr: 0.000014 - momentum: 0.000000
2023-10-23 15:55:41,372 epoch 6 - iter 130/138 - loss 0.03570597 - time (sec): 7.44 - samples/sec: 2869.61 - lr: 0.000014 - momentum: 0.000000
2023-10-23 15:55:41,829 ----------------------------------------------------------------------------------------------------
2023-10-23 15:55:41,829 EPOCH 6 done: loss 0.0388 - lr: 0.000014
2023-10-23 15:55:42,364 DEV : loss 0.12986908853054047 - f1-score (micro avg) 0.8857
2023-10-23 15:55:42,370 saving best model
2023-10-23 15:55:42,912 ----------------------------------------------------------------------------------------------------
2023-10-23 15:55:43,641 epoch 7 - iter 13/138 - loss 0.04560714 - time (sec): 0.72 - samples/sec: 3046.17 - lr: 0.000013 - momentum: 0.000000
2023-10-23 15:55:44,362 epoch 7 - iter 26/138 - loss 0.04097470 - time (sec): 1.45 - samples/sec: 2967.32 - lr: 0.000013 - momentum: 0.000000
2023-10-23 15:55:45,104 epoch 7 - iter 39/138 - loss 0.03509241 - time (sec): 2.19 - samples/sec: 2861.35 - lr: 0.000012 - momentum: 0.000000
2023-10-23 15:55:45,854 epoch 7 - iter 52/138 - loss 0.03079550 - time (sec): 2.94 - samples/sec: 2872.75 - lr: 0.000012 - momentum: 0.000000
2023-10-23 15:55:46,591 epoch 7 - iter 65/138 - loss 0.02734683 - time (sec): 3.67 - samples/sec: 2913.33 - lr: 0.000012 - momentum: 0.000000
2023-10-23 15:55:47,325 epoch 7 - iter 78/138 - loss 0.02674504 - time (sec): 4.41 - samples/sec: 2904.82 - lr: 0.000012 - momentum: 0.000000
2023-10-23 15:55:48,052 epoch 7 - iter 91/138 - loss 0.02597626 - time (sec): 5.14 - samples/sec: 2898.08 - lr: 0.000011 - momentum: 0.000000
2023-10-23 15:55:48,786 epoch 7 - iter 104/138 - loss 0.02279908 - time (sec): 5.87 - samples/sec: 2938.68 - lr: 0.000011 - momentum: 0.000000
2023-10-23 15:55:49,527 epoch 7 - iter 117/138 - loss 0.02768589 - time (sec): 6.61 - samples/sec: 2959.75 - lr: 0.000011 - momentum: 0.000000
2023-10-23 15:55:50,264 epoch 7 - iter 130/138 - loss 0.02689635 - time (sec): 7.35 - samples/sec: 2936.27 - lr: 0.000010 - momentum: 0.000000
2023-10-23 15:55:50,716 ----------------------------------------------------------------------------------------------------
2023-10-23 15:55:50,716 EPOCH 7 done: loss 0.0261 - lr: 0.000010
2023-10-23 15:55:51,252 DEV : loss 0.14399342238903046 - f1-score (micro avg) 0.8766
2023-10-23 15:55:51,258 ----------------------------------------------------------------------------------------------------
2023-10-23 15:55:52,007 epoch 8 - iter 13/138 - loss 0.03629714 - time (sec): 0.75 - samples/sec: 2708.89 - lr: 0.000010 - momentum: 0.000000
2023-10-23 15:55:52,767 epoch 8 - iter 26/138 - loss 0.02677088 - time (sec): 1.51 - samples/sec: 2954.84 - lr: 0.000009 - momentum: 0.000000
2023-10-23 15:55:53,495 epoch 8 - iter 39/138 - loss 0.03630954 - time (sec): 2.24 - samples/sec: 2889.29 - lr: 0.000009 - momentum: 0.000000
2023-10-23 15:55:54,276 epoch 8 - iter 52/138 - loss 0.03402449 - time (sec): 3.02 - samples/sec: 2934.37 - lr: 0.000009 - momentum: 0.000000
2023-10-23 15:55:55,016 epoch 8 - iter 65/138 - loss 0.03031141 - time (sec): 3.76 - samples/sec: 2962.17 - lr: 0.000009 - momentum: 0.000000
2023-10-23 15:55:55,781 epoch 8 - iter 78/138 - loss 0.02709042 - time (sec): 4.52 - samples/sec: 2964.19 - lr: 0.000008 - momentum: 0.000000
2023-10-23 15:55:56,513 epoch 8 - iter 91/138 - loss 0.02482129 - time (sec): 5.25 - samples/sec: 2943.89 - lr: 0.000008 - momentum: 0.000000
2023-10-23 15:55:57,244 epoch 8 - iter 104/138 - loss 0.02348331 - time (sec): 5.98 - samples/sec: 2909.14 - lr: 0.000008 - momentum: 0.000000
2023-10-23 15:55:58,010 epoch 8 - iter 117/138 - loss 0.02455497 - time (sec): 6.75 - samples/sec: 2883.47 - lr: 0.000007 - momentum: 0.000000
2023-10-23 15:55:58,754 epoch 8 - iter 130/138 - loss 0.02442744 - time (sec): 7.49 - samples/sec: 2877.71 - lr: 0.000007 - momentum: 0.000000
2023-10-23 15:55:59,224 ----------------------------------------------------------------------------------------------------
2023-10-23 15:55:59,224 EPOCH 8 done: loss 0.0243 - lr: 0.000007
2023-10-23 15:55:59,776 DEV : loss 0.14874805510044098 - f1-score (micro avg) 0.8948
2023-10-23 15:55:59,782 saving best model
2023-10-23 15:56:00,334 ----------------------------------------------------------------------------------------------------
2023-10-23 15:56:01,062 epoch 9 - iter 13/138 - loss 0.01544187 - time (sec): 0.72 - samples/sec: 3068.12 - lr: 0.000006 - momentum: 0.000000
2023-10-23 15:56:01,793 epoch 9 - iter 26/138 - loss 0.01943197 - time (sec): 1.45 - samples/sec: 2962.08 - lr: 0.000006 - momentum: 0.000000
2023-10-23 15:56:02,529 epoch 9 - iter 39/138 - loss 0.02702448 - time (sec): 2.19 - samples/sec: 3028.97 - lr: 0.000006 - momentum: 0.000000
2023-10-23 15:56:03,253 epoch 9 - iter 52/138 - loss 0.02759916 - time (sec): 2.91 - samples/sec: 2961.00 - lr: 0.000005 - momentum: 0.000000
2023-10-23 15:56:03,988 epoch 9 - iter 65/138 - loss 0.02404425 - time (sec): 3.65 - samples/sec: 2930.40 - lr: 0.000005 - momentum: 0.000000
2023-10-23 15:56:04,729 epoch 9 - iter 78/138 - loss 0.01987204 - time (sec): 4.39 - samples/sec: 3004.47 - lr: 0.000005 - momentum: 0.000000
2023-10-23 15:56:05,471 epoch 9 - iter 91/138 - loss 0.01814794 - time (sec): 5.13 - samples/sec: 2981.58 - lr: 0.000005 - momentum: 0.000000
2023-10-23 15:56:06,204 epoch 9 - iter 104/138 - loss 0.01732655 - time (sec): 5.86 - samples/sec: 2973.69 - lr: 0.000004 - momentum: 0.000000
2023-10-23 15:56:06,931 epoch 9 - iter 117/138 - loss 0.01633533 - time (sec): 6.59 - samples/sec: 2991.47 - lr: 0.000004 - momentum: 0.000000
2023-10-23 15:56:07,675 epoch 9 - iter 130/138 - loss 0.01746274 - time (sec): 7.33 - samples/sec: 2956.10 - lr: 0.000004 - momentum: 0.000000
2023-10-23 15:56:08,128 ----------------------------------------------------------------------------------------------------
2023-10-23 15:56:08,129 EPOCH 9 done: loss 0.0179 - lr: 0.000004
2023-10-23 15:56:08,663 DEV : loss 0.1484120488166809 - f1-score (micro avg) 0.8852
2023-10-23 15:56:08,669 ----------------------------------------------------------------------------------------------------
2023-10-23 15:56:09,391 epoch 10 - iter 13/138 - loss 0.02130054 - time (sec): 0.72 - samples/sec: 2830.96 - lr: 0.000003 - momentum: 0.000000
2023-10-23 15:56:10,108 epoch 10 - iter 26/138 - loss 0.01334521 - time (sec): 1.44 - samples/sec: 2911.14 - lr: 0.000003 - momentum: 0.000000
2023-10-23 15:56:10,826 epoch 10 - iter 39/138 - loss 0.01106197 - time (sec): 2.16 - samples/sec: 2984.20 - lr: 0.000002 - momentum: 0.000000
2023-10-23 15:56:11,576 epoch 10 - iter 52/138 - loss 0.00961286 - time (sec): 2.91 - samples/sec: 3018.09 - lr: 0.000002 - momentum: 0.000000
2023-10-23 15:56:12,281 epoch 10 - iter 65/138 - loss 0.00864498 - time (sec): 3.61 - samples/sec: 2954.18 - lr: 0.000002 - momentum: 0.000000
2023-10-23 15:56:13,017 epoch 10 - iter 78/138 - loss 0.01211412 - time (sec): 4.35 - samples/sec: 2938.79 - lr: 0.000002 - momentum: 0.000000
2023-10-23 15:56:13,741 epoch 10 - iter 91/138 - loss 0.01154617 - time (sec): 5.07 - samples/sec: 2914.28 - lr: 0.000001 - momentum: 0.000000
2023-10-23 15:56:14,451 epoch 10 - iter 104/138 - loss 0.01059741 - time (sec): 5.78 - samples/sec: 2936.14 - lr: 0.000001 - momentum: 0.000000
2023-10-23 15:56:15,177 epoch 10 - iter 117/138 - loss 0.01141697 - time (sec): 6.51 - samples/sec: 2935.76 - lr: 0.000001 - momentum: 0.000000
2023-10-23 15:56:15,924 epoch 10 - iter 130/138 - loss 0.01581859 - time (sec): 7.25 - samples/sec: 2959.38 - lr: 0.000000 - momentum: 0.000000
2023-10-23 15:56:16,375 ----------------------------------------------------------------------------------------------------
2023-10-23 15:56:16,375 EPOCH 10 done: loss 0.0156 - lr: 0.000000
2023-10-23 15:56:16,904 DEV : loss 0.14890803396701813 - f1-score (micro avg) 0.8918
2023-10-23 15:56:17,307 ----------------------------------------------------------------------------------------------------
2023-10-23 15:56:17,308 Loading model from best epoch ...
2023-10-23 15:56:18,949 SequenceTagger predicts: Dictionary with 25 tags: O, S-scope, B-scope, E-scope, I-scope, S-pers, B-pers, E-pers, I-pers, S-work, B-work, E-work, I-work, S-loc, B-loc, E-loc, I-loc, S-object, B-object, E-object, I-object, S-date, B-date, E-date, I-date
2023-10-23 15:56:19,626
Results:
- F-score (micro) 0.9026
- F-score (macro) 0.8039
- Accuracy 0.8265
By class:
precision recall f1-score support
scope 0.8827 0.8977 0.8901 176
pers 0.9612 0.9688 0.9650 128
work 0.8676 0.7973 0.8310 74
object 1.0000 0.5000 0.6667 2
loc 1.0000 0.5000 0.6667 2
micro avg 0.9074 0.8979 0.9026 382
macro avg 0.9423 0.7328 0.8039 382
weighted avg 0.9073 0.8979 0.9014 382
2023-10-23 15:56:19,626 ----------------------------------------------------------------------------------------------------
|