File size: 24,010 Bytes
8e0c0aa |
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 |
2023-10-17 09:36:26,101 ----------------------------------------------------------------------------------------------------
2023-10-17 09:36:26,102 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=25, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-17 09:36:26,102 ----------------------------------------------------------------------------------------------------
2023-10-17 09:36:26,102 MultiCorpus: 1214 train + 266 dev + 251 test sentences
- NER_HIPE_2022 Corpus: 1214 train + 266 dev + 251 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/ajmc/en/with_doc_seperator
2023-10-17 09:36:26,102 ----------------------------------------------------------------------------------------------------
2023-10-17 09:36:26,102 Train: 1214 sentences
2023-10-17 09:36:26,102 (train_with_dev=False, train_with_test=False)
2023-10-17 09:36:26,102 ----------------------------------------------------------------------------------------------------
2023-10-17 09:36:26,103 Training Params:
2023-10-17 09:36:26,103 - learning_rate: "3e-05"
2023-10-17 09:36:26,103 - mini_batch_size: "8"
2023-10-17 09:36:26,103 - max_epochs: "10"
2023-10-17 09:36:26,103 - shuffle: "True"
2023-10-17 09:36:26,103 ----------------------------------------------------------------------------------------------------
2023-10-17 09:36:26,103 Plugins:
2023-10-17 09:36:26,103 - TensorboardLogger
2023-10-17 09:36:26,103 - LinearScheduler | warmup_fraction: '0.1'
2023-10-17 09:36:26,103 ----------------------------------------------------------------------------------------------------
2023-10-17 09:36:26,103 Final evaluation on model from best epoch (best-model.pt)
2023-10-17 09:36:26,103 - metric: "('micro avg', 'f1-score')"
2023-10-17 09:36:26,103 ----------------------------------------------------------------------------------------------------
2023-10-17 09:36:26,103 Computation:
2023-10-17 09:36:26,103 - compute on device: cuda:0
2023-10-17 09:36:26,103 - embedding storage: none
2023-10-17 09:36:26,103 ----------------------------------------------------------------------------------------------------
2023-10-17 09:36:26,103 Model training base path: "hmbench-ajmc/en-hmteams/teams-base-historic-multilingual-discriminator-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-1"
2023-10-17 09:36:26,103 ----------------------------------------------------------------------------------------------------
2023-10-17 09:36:26,103 ----------------------------------------------------------------------------------------------------
2023-10-17 09:36:26,103 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-17 09:36:26,932 epoch 1 - iter 15/152 - loss 3.35898714 - time (sec): 0.83 - samples/sec: 3632.65 - lr: 0.000003 - momentum: 0.000000
2023-10-17 09:36:27,788 epoch 1 - iter 30/152 - loss 3.02212465 - time (sec): 1.68 - samples/sec: 3534.27 - lr: 0.000006 - momentum: 0.000000
2023-10-17 09:36:28,663 epoch 1 - iter 45/152 - loss 2.45612354 - time (sec): 2.56 - samples/sec: 3670.98 - lr: 0.000009 - momentum: 0.000000
2023-10-17 09:36:29,535 epoch 1 - iter 60/152 - loss 1.99749694 - time (sec): 3.43 - samples/sec: 3698.47 - lr: 0.000012 - momentum: 0.000000
2023-10-17 09:36:30,327 epoch 1 - iter 75/152 - loss 1.73641236 - time (sec): 4.22 - samples/sec: 3660.69 - lr: 0.000015 - momentum: 0.000000
2023-10-17 09:36:31,127 epoch 1 - iter 90/152 - loss 1.52634135 - time (sec): 5.02 - samples/sec: 3636.89 - lr: 0.000018 - momentum: 0.000000
2023-10-17 09:36:32,009 epoch 1 - iter 105/152 - loss 1.36260959 - time (sec): 5.90 - samples/sec: 3626.11 - lr: 0.000021 - momentum: 0.000000
2023-10-17 09:36:32,868 epoch 1 - iter 120/152 - loss 1.24452509 - time (sec): 6.76 - samples/sec: 3614.67 - lr: 0.000023 - momentum: 0.000000
2023-10-17 09:36:33,728 epoch 1 - iter 135/152 - loss 1.14524406 - time (sec): 7.62 - samples/sec: 3601.31 - lr: 0.000026 - momentum: 0.000000
2023-10-17 09:36:34,591 epoch 1 - iter 150/152 - loss 1.05621100 - time (sec): 8.49 - samples/sec: 3611.55 - lr: 0.000029 - momentum: 0.000000
2023-10-17 09:36:34,694 ----------------------------------------------------------------------------------------------------
2023-10-17 09:36:34,694 EPOCH 1 done: loss 1.0473 - lr: 0.000029
2023-10-17 09:36:35,450 DEV : loss 0.23443591594696045 - f1-score (micro avg) 0.5184
2023-10-17 09:36:35,456 saving best model
2023-10-17 09:36:35,781 ----------------------------------------------------------------------------------------------------
2023-10-17 09:36:36,672 epoch 2 - iter 15/152 - loss 0.24169486 - time (sec): 0.89 - samples/sec: 3386.25 - lr: 0.000030 - momentum: 0.000000
2023-10-17 09:36:37,515 epoch 2 - iter 30/152 - loss 0.23715876 - time (sec): 1.73 - samples/sec: 3525.57 - lr: 0.000029 - momentum: 0.000000
2023-10-17 09:36:38,343 epoch 2 - iter 45/152 - loss 0.21104596 - time (sec): 2.56 - samples/sec: 3598.89 - lr: 0.000029 - momentum: 0.000000
2023-10-17 09:36:39,220 epoch 2 - iter 60/152 - loss 0.20097868 - time (sec): 3.44 - samples/sec: 3562.82 - lr: 0.000029 - momentum: 0.000000
2023-10-17 09:36:40,138 epoch 2 - iter 75/152 - loss 0.19466306 - time (sec): 4.35 - samples/sec: 3541.60 - lr: 0.000028 - momentum: 0.000000
2023-10-17 09:36:40,980 epoch 2 - iter 90/152 - loss 0.18556494 - time (sec): 5.20 - samples/sec: 3523.73 - lr: 0.000028 - momentum: 0.000000
2023-10-17 09:36:41,807 epoch 2 - iter 105/152 - loss 0.18053346 - time (sec): 6.02 - samples/sec: 3513.61 - lr: 0.000028 - momentum: 0.000000
2023-10-17 09:36:42,670 epoch 2 - iter 120/152 - loss 0.18092533 - time (sec): 6.89 - samples/sec: 3560.07 - lr: 0.000027 - momentum: 0.000000
2023-10-17 09:36:43,530 epoch 2 - iter 135/152 - loss 0.17278558 - time (sec): 7.75 - samples/sec: 3580.87 - lr: 0.000027 - momentum: 0.000000
2023-10-17 09:36:44,376 epoch 2 - iter 150/152 - loss 0.17079577 - time (sec): 8.59 - samples/sec: 3571.90 - lr: 0.000027 - momentum: 0.000000
2023-10-17 09:36:44,475 ----------------------------------------------------------------------------------------------------
2023-10-17 09:36:44,475 EPOCH 2 done: loss 0.1696 - lr: 0.000027
2023-10-17 09:36:45,392 DEV : loss 0.13970671594142914 - f1-score (micro avg) 0.7763
2023-10-17 09:36:45,398 saving best model
2023-10-17 09:36:45,846 ----------------------------------------------------------------------------------------------------
2023-10-17 09:36:46,710 epoch 3 - iter 15/152 - loss 0.09358125 - time (sec): 0.86 - samples/sec: 3368.12 - lr: 0.000026 - momentum: 0.000000
2023-10-17 09:36:47,568 epoch 3 - iter 30/152 - loss 0.09697182 - time (sec): 1.72 - samples/sec: 3416.85 - lr: 0.000026 - momentum: 0.000000
2023-10-17 09:36:48,437 epoch 3 - iter 45/152 - loss 0.09722146 - time (sec): 2.59 - samples/sec: 3381.80 - lr: 0.000026 - momentum: 0.000000
2023-10-17 09:36:49,280 epoch 3 - iter 60/152 - loss 0.09059189 - time (sec): 3.43 - samples/sec: 3414.31 - lr: 0.000025 - momentum: 0.000000
2023-10-17 09:36:50,175 epoch 3 - iter 75/152 - loss 0.08518541 - time (sec): 4.33 - samples/sec: 3498.34 - lr: 0.000025 - momentum: 0.000000
2023-10-17 09:36:51,076 epoch 3 - iter 90/152 - loss 0.09661947 - time (sec): 5.23 - samples/sec: 3505.34 - lr: 0.000025 - momentum: 0.000000
2023-10-17 09:36:51,931 epoch 3 - iter 105/152 - loss 0.10037094 - time (sec): 6.08 - samples/sec: 3524.40 - lr: 0.000024 - momentum: 0.000000
2023-10-17 09:36:52,760 epoch 3 - iter 120/152 - loss 0.09624042 - time (sec): 6.91 - samples/sec: 3540.82 - lr: 0.000024 - momentum: 0.000000
2023-10-17 09:36:53,578 epoch 3 - iter 135/152 - loss 0.09198926 - time (sec): 7.73 - samples/sec: 3538.43 - lr: 0.000024 - momentum: 0.000000
2023-10-17 09:36:54,489 epoch 3 - iter 150/152 - loss 0.08944775 - time (sec): 8.64 - samples/sec: 3547.63 - lr: 0.000023 - momentum: 0.000000
2023-10-17 09:36:54,591 ----------------------------------------------------------------------------------------------------
2023-10-17 09:36:54,591 EPOCH 3 done: loss 0.0897 - lr: 0.000023
2023-10-17 09:36:55,531 DEV : loss 0.13742879033088684 - f1-score (micro avg) 0.8245
2023-10-17 09:36:55,538 saving best model
2023-10-17 09:36:56,022 ----------------------------------------------------------------------------------------------------
2023-10-17 09:36:56,879 epoch 4 - iter 15/152 - loss 0.03956629 - time (sec): 0.86 - samples/sec: 3634.93 - lr: 0.000023 - momentum: 0.000000
2023-10-17 09:36:57,724 epoch 4 - iter 30/152 - loss 0.05596716 - time (sec): 1.70 - samples/sec: 3605.37 - lr: 0.000023 - momentum: 0.000000
2023-10-17 09:36:58,546 epoch 4 - iter 45/152 - loss 0.06316567 - time (sec): 2.52 - samples/sec: 3561.13 - lr: 0.000022 - momentum: 0.000000
2023-10-17 09:36:59,461 epoch 4 - iter 60/152 - loss 0.06216934 - time (sec): 3.44 - samples/sec: 3598.96 - lr: 0.000022 - momentum: 0.000000
2023-10-17 09:37:00,333 epoch 4 - iter 75/152 - loss 0.05941341 - time (sec): 4.31 - samples/sec: 3549.57 - lr: 0.000022 - momentum: 0.000000
2023-10-17 09:37:01,339 epoch 4 - iter 90/152 - loss 0.05979395 - time (sec): 5.32 - samples/sec: 3425.34 - lr: 0.000021 - momentum: 0.000000
2023-10-17 09:37:02,210 epoch 4 - iter 105/152 - loss 0.06163013 - time (sec): 6.19 - samples/sec: 3460.81 - lr: 0.000021 - momentum: 0.000000
2023-10-17 09:37:03,064 epoch 4 - iter 120/152 - loss 0.06016428 - time (sec): 7.04 - samples/sec: 3488.96 - lr: 0.000021 - momentum: 0.000000
2023-10-17 09:37:03,947 epoch 4 - iter 135/152 - loss 0.06300382 - time (sec): 7.92 - samples/sec: 3490.01 - lr: 0.000020 - momentum: 0.000000
2023-10-17 09:37:04,785 epoch 4 - iter 150/152 - loss 0.06228708 - time (sec): 8.76 - samples/sec: 3504.62 - lr: 0.000020 - momentum: 0.000000
2023-10-17 09:37:04,891 ----------------------------------------------------------------------------------------------------
2023-10-17 09:37:04,891 EPOCH 4 done: loss 0.0623 - lr: 0.000020
2023-10-17 09:37:05,831 DEV : loss 0.1494779735803604 - f1-score (micro avg) 0.8282
2023-10-17 09:37:05,837 saving best model
2023-10-17 09:37:06,323 ----------------------------------------------------------------------------------------------------
2023-10-17 09:37:07,248 epoch 5 - iter 15/152 - loss 0.05545218 - time (sec): 0.92 - samples/sec: 3654.78 - lr: 0.000020 - momentum: 0.000000
2023-10-17 09:37:08,170 epoch 5 - iter 30/152 - loss 0.04238699 - time (sec): 1.85 - samples/sec: 3436.50 - lr: 0.000019 - momentum: 0.000000
2023-10-17 09:37:09,096 epoch 5 - iter 45/152 - loss 0.04725187 - time (sec): 2.77 - samples/sec: 3492.98 - lr: 0.000019 - momentum: 0.000000
2023-10-17 09:37:09,967 epoch 5 - iter 60/152 - loss 0.04013426 - time (sec): 3.64 - samples/sec: 3489.30 - lr: 0.000019 - momentum: 0.000000
2023-10-17 09:37:10,857 epoch 5 - iter 75/152 - loss 0.03634974 - time (sec): 4.53 - samples/sec: 3481.04 - lr: 0.000018 - momentum: 0.000000
2023-10-17 09:37:11,741 epoch 5 - iter 90/152 - loss 0.03798504 - time (sec): 5.42 - samples/sec: 3448.37 - lr: 0.000018 - momentum: 0.000000
2023-10-17 09:37:12,629 epoch 5 - iter 105/152 - loss 0.03712308 - time (sec): 6.30 - samples/sec: 3439.13 - lr: 0.000018 - momentum: 0.000000
2023-10-17 09:37:13,452 epoch 5 - iter 120/152 - loss 0.04227013 - time (sec): 7.13 - samples/sec: 3428.06 - lr: 0.000017 - momentum: 0.000000
2023-10-17 09:37:14,334 epoch 5 - iter 135/152 - loss 0.04112595 - time (sec): 8.01 - samples/sec: 3442.29 - lr: 0.000017 - momentum: 0.000000
2023-10-17 09:37:15,191 epoch 5 - iter 150/152 - loss 0.04377806 - time (sec): 8.87 - samples/sec: 3462.64 - lr: 0.000017 - momentum: 0.000000
2023-10-17 09:37:15,289 ----------------------------------------------------------------------------------------------------
2023-10-17 09:37:15,289 EPOCH 5 done: loss 0.0436 - lr: 0.000017
2023-10-17 09:37:16,234 DEV : loss 0.1704065203666687 - f1-score (micro avg) 0.8434
2023-10-17 09:37:16,241 saving best model
2023-10-17 09:37:16,697 ----------------------------------------------------------------------------------------------------
2023-10-17 09:37:17,588 epoch 6 - iter 15/152 - loss 0.04317289 - time (sec): 0.88 - samples/sec: 3518.38 - lr: 0.000016 - momentum: 0.000000
2023-10-17 09:37:18,416 epoch 6 - iter 30/152 - loss 0.03691780 - time (sec): 1.71 - samples/sec: 3462.62 - lr: 0.000016 - momentum: 0.000000
2023-10-17 09:37:19,300 epoch 6 - iter 45/152 - loss 0.03511291 - time (sec): 2.59 - samples/sec: 3430.87 - lr: 0.000016 - momentum: 0.000000
2023-10-17 09:37:20,183 epoch 6 - iter 60/152 - loss 0.03753407 - time (sec): 3.48 - samples/sec: 3432.90 - lr: 0.000015 - momentum: 0.000000
2023-10-17 09:37:21,052 epoch 6 - iter 75/152 - loss 0.03621585 - time (sec): 4.34 - samples/sec: 3482.89 - lr: 0.000015 - momentum: 0.000000
2023-10-17 09:37:21,890 epoch 6 - iter 90/152 - loss 0.03430368 - time (sec): 5.18 - samples/sec: 3495.93 - lr: 0.000015 - momentum: 0.000000
2023-10-17 09:37:22,787 epoch 6 - iter 105/152 - loss 0.04059258 - time (sec): 6.08 - samples/sec: 3504.70 - lr: 0.000014 - momentum: 0.000000
2023-10-17 09:37:23,646 epoch 6 - iter 120/152 - loss 0.03799652 - time (sec): 6.94 - samples/sec: 3498.07 - lr: 0.000014 - momentum: 0.000000
2023-10-17 09:37:24,486 epoch 6 - iter 135/152 - loss 0.03759158 - time (sec): 7.78 - samples/sec: 3520.77 - lr: 0.000014 - momentum: 0.000000
2023-10-17 09:37:25,379 epoch 6 - iter 150/152 - loss 0.03898740 - time (sec): 8.67 - samples/sec: 3535.40 - lr: 0.000013 - momentum: 0.000000
2023-10-17 09:37:25,472 ----------------------------------------------------------------------------------------------------
2023-10-17 09:37:25,472 EPOCH 6 done: loss 0.0392 - lr: 0.000013
2023-10-17 09:37:26,445 DEV : loss 0.17794041335582733 - f1-score (micro avg) 0.839
2023-10-17 09:37:26,453 ----------------------------------------------------------------------------------------------------
2023-10-17 09:37:27,356 epoch 7 - iter 15/152 - loss 0.01465965 - time (sec): 0.90 - samples/sec: 3112.14 - lr: 0.000013 - momentum: 0.000000
2023-10-17 09:37:28,224 epoch 7 - iter 30/152 - loss 0.01342575 - time (sec): 1.77 - samples/sec: 3298.14 - lr: 0.000013 - momentum: 0.000000
2023-10-17 09:37:29,082 epoch 7 - iter 45/152 - loss 0.01948712 - time (sec): 2.63 - samples/sec: 3395.28 - lr: 0.000012 - momentum: 0.000000
2023-10-17 09:37:29,937 epoch 7 - iter 60/152 - loss 0.02268211 - time (sec): 3.48 - samples/sec: 3401.47 - lr: 0.000012 - momentum: 0.000000
2023-10-17 09:37:30,759 epoch 7 - iter 75/152 - loss 0.02031838 - time (sec): 4.30 - samples/sec: 3481.55 - lr: 0.000012 - momentum: 0.000000
2023-10-17 09:37:31,605 epoch 7 - iter 90/152 - loss 0.01853079 - time (sec): 5.15 - samples/sec: 3495.72 - lr: 0.000011 - momentum: 0.000000
2023-10-17 09:37:32,460 epoch 7 - iter 105/152 - loss 0.02027409 - time (sec): 6.00 - samples/sec: 3505.39 - lr: 0.000011 - momentum: 0.000000
2023-10-17 09:37:33,390 epoch 7 - iter 120/152 - loss 0.02176488 - time (sec): 6.93 - samples/sec: 3522.89 - lr: 0.000011 - momentum: 0.000000
2023-10-17 09:37:34,284 epoch 7 - iter 135/152 - loss 0.02344965 - time (sec): 7.83 - samples/sec: 3542.75 - lr: 0.000010 - momentum: 0.000000
2023-10-17 09:37:35,098 epoch 7 - iter 150/152 - loss 0.02710869 - time (sec): 8.64 - samples/sec: 3544.50 - lr: 0.000010 - momentum: 0.000000
2023-10-17 09:37:35,204 ----------------------------------------------------------------------------------------------------
2023-10-17 09:37:35,204 EPOCH 7 done: loss 0.0273 - lr: 0.000010
2023-10-17 09:37:36,183 DEV : loss 0.18339850008487701 - f1-score (micro avg) 0.8421
2023-10-17 09:37:36,191 ----------------------------------------------------------------------------------------------------
2023-10-17 09:37:37,052 epoch 8 - iter 15/152 - loss 0.01167967 - time (sec): 0.86 - samples/sec: 3917.73 - lr: 0.000010 - momentum: 0.000000
2023-10-17 09:37:37,918 epoch 8 - iter 30/152 - loss 0.01022357 - time (sec): 1.73 - samples/sec: 3699.27 - lr: 0.000009 - momentum: 0.000000
2023-10-17 09:37:38,811 epoch 8 - iter 45/152 - loss 0.01816524 - time (sec): 2.62 - samples/sec: 3626.56 - lr: 0.000009 - momentum: 0.000000
2023-10-17 09:37:39,640 epoch 8 - iter 60/152 - loss 0.01610311 - time (sec): 3.45 - samples/sec: 3532.16 - lr: 0.000009 - momentum: 0.000000
2023-10-17 09:37:40,532 epoch 8 - iter 75/152 - loss 0.01399500 - time (sec): 4.34 - samples/sec: 3577.81 - lr: 0.000008 - momentum: 0.000000
2023-10-17 09:37:41,328 epoch 8 - iter 90/152 - loss 0.01508785 - time (sec): 5.14 - samples/sec: 3587.85 - lr: 0.000008 - momentum: 0.000000
2023-10-17 09:37:42,196 epoch 8 - iter 105/152 - loss 0.01972651 - time (sec): 6.00 - samples/sec: 3563.80 - lr: 0.000008 - momentum: 0.000000
2023-10-17 09:37:43,084 epoch 8 - iter 120/152 - loss 0.02152978 - time (sec): 6.89 - samples/sec: 3557.71 - lr: 0.000007 - momentum: 0.000000
2023-10-17 09:37:43,960 epoch 8 - iter 135/152 - loss 0.02068304 - time (sec): 7.77 - samples/sec: 3555.58 - lr: 0.000007 - momentum: 0.000000
2023-10-17 09:37:44,816 epoch 8 - iter 150/152 - loss 0.02075768 - time (sec): 8.62 - samples/sec: 3550.64 - lr: 0.000007 - momentum: 0.000000
2023-10-17 09:37:44,924 ----------------------------------------------------------------------------------------------------
2023-10-17 09:37:44,924 EPOCH 8 done: loss 0.0205 - lr: 0.000007
2023-10-17 09:37:45,870 DEV : loss 0.19150298833847046 - f1-score (micro avg) 0.8436
2023-10-17 09:37:45,876 saving best model
2023-10-17 09:37:46,342 ----------------------------------------------------------------------------------------------------
2023-10-17 09:37:47,201 epoch 9 - iter 15/152 - loss 0.01688113 - time (sec): 0.86 - samples/sec: 3585.02 - lr: 0.000006 - momentum: 0.000000
2023-10-17 09:37:48,035 epoch 9 - iter 30/152 - loss 0.00892330 - time (sec): 1.69 - samples/sec: 3535.34 - lr: 0.000006 - momentum: 0.000000
2023-10-17 09:37:48,912 epoch 9 - iter 45/152 - loss 0.01524391 - time (sec): 2.57 - samples/sec: 3566.72 - lr: 0.000006 - momentum: 0.000000
2023-10-17 09:37:49,759 epoch 9 - iter 60/152 - loss 0.01537673 - time (sec): 3.42 - samples/sec: 3525.94 - lr: 0.000005 - momentum: 0.000000
2023-10-17 09:37:50,636 epoch 9 - iter 75/152 - loss 0.01573533 - time (sec): 4.29 - samples/sec: 3549.94 - lr: 0.000005 - momentum: 0.000000
2023-10-17 09:37:51,497 epoch 9 - iter 90/152 - loss 0.01569790 - time (sec): 5.15 - samples/sec: 3561.83 - lr: 0.000005 - momentum: 0.000000
2023-10-17 09:37:52,387 epoch 9 - iter 105/152 - loss 0.01417753 - time (sec): 6.04 - samples/sec: 3527.35 - lr: 0.000004 - momentum: 0.000000
2023-10-17 09:37:53,184 epoch 9 - iter 120/152 - loss 0.01521950 - time (sec): 6.84 - samples/sec: 3546.82 - lr: 0.000004 - momentum: 0.000000
2023-10-17 09:37:54,034 epoch 9 - iter 135/152 - loss 0.01501499 - time (sec): 7.69 - samples/sec: 3567.29 - lr: 0.000004 - momentum: 0.000000
2023-10-17 09:37:54,900 epoch 9 - iter 150/152 - loss 0.01607239 - time (sec): 8.56 - samples/sec: 3582.08 - lr: 0.000004 - momentum: 0.000000
2023-10-17 09:37:55,000 ----------------------------------------------------------------------------------------------------
2023-10-17 09:37:55,000 EPOCH 9 done: loss 0.0159 - lr: 0.000004
2023-10-17 09:37:55,939 DEV : loss 0.19435930252075195 - f1-score (micro avg) 0.8479
2023-10-17 09:37:55,945 saving best model
2023-10-17 09:37:56,410 ----------------------------------------------------------------------------------------------------
2023-10-17 09:37:57,241 epoch 10 - iter 15/152 - loss 0.01021824 - time (sec): 0.83 - samples/sec: 3542.39 - lr: 0.000003 - momentum: 0.000000
2023-10-17 09:37:58,114 epoch 10 - iter 30/152 - loss 0.01670466 - time (sec): 1.70 - samples/sec: 3509.32 - lr: 0.000003 - momentum: 0.000000
2023-10-17 09:37:58,988 epoch 10 - iter 45/152 - loss 0.01789261 - time (sec): 2.58 - samples/sec: 3597.82 - lr: 0.000002 - momentum: 0.000000
2023-10-17 09:37:59,849 epoch 10 - iter 60/152 - loss 0.02154099 - time (sec): 3.44 - samples/sec: 3547.31 - lr: 0.000002 - momentum: 0.000000
2023-10-17 09:38:00,669 epoch 10 - iter 75/152 - loss 0.01827148 - time (sec): 4.26 - samples/sec: 3556.10 - lr: 0.000002 - momentum: 0.000000
2023-10-17 09:38:01,580 epoch 10 - iter 90/152 - loss 0.01773986 - time (sec): 5.17 - samples/sec: 3509.28 - lr: 0.000002 - momentum: 0.000000
2023-10-17 09:38:02,467 epoch 10 - iter 105/152 - loss 0.01550723 - time (sec): 6.06 - samples/sec: 3526.34 - lr: 0.000001 - momentum: 0.000000
2023-10-17 09:38:03,330 epoch 10 - iter 120/152 - loss 0.01531261 - time (sec): 6.92 - samples/sec: 3517.18 - lr: 0.000001 - momentum: 0.000000
2023-10-17 09:38:04,227 epoch 10 - iter 135/152 - loss 0.01532677 - time (sec): 7.82 - samples/sec: 3505.78 - lr: 0.000001 - momentum: 0.000000
2023-10-17 09:38:05,120 epoch 10 - iter 150/152 - loss 0.01646360 - time (sec): 8.71 - samples/sec: 3515.50 - lr: 0.000000 - momentum: 0.000000
2023-10-17 09:38:05,232 ----------------------------------------------------------------------------------------------------
2023-10-17 09:38:05,232 EPOCH 10 done: loss 0.0164 - lr: 0.000000
2023-10-17 09:38:06,206 DEV : loss 0.19792957603931427 - f1-score (micro avg) 0.8419
2023-10-17 09:38:06,568 ----------------------------------------------------------------------------------------------------
2023-10-17 09:38:06,570 Loading model from best epoch ...
2023-10-17 09:38:08,115 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-date, B-date, E-date, I-date, S-object, B-object, E-object, I-object
2023-10-17 09:38:09,020
Results:
- F-score (micro) 0.8368
- F-score (macro) 0.67
- Accuracy 0.7282
By class:
precision recall f1-score support
scope 0.7898 0.8212 0.8052 151
work 0.7810 0.8632 0.8200 95
pers 0.8932 0.9583 0.9246 96
loc 1.0000 0.6667 0.8000 3
date 0.0000 0.0000 0.0000 3
micro avg 0.8130 0.8621 0.8368 348
macro avg 0.6928 0.6619 0.6700 348
weighted avg 0.8109 0.8621 0.8352 348
2023-10-17 09:38:09,021 ----------------------------------------------------------------------------------------------------
|