File size: 24,060 Bytes
a1f4f60 |
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 |
2023-10-25 21:12:45,451 ----------------------------------------------------------------------------------------------------
2023-10-25 21:12:45,452 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=17, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-25 21:12:45,452 ----------------------------------------------------------------------------------------------------
2023-10-25 21:12:45,453 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
2023-10-25 21:12:45,453 ----------------------------------------------------------------------------------------------------
2023-10-25 21:12:45,453 Train: 1166 sentences
2023-10-25 21:12:45,453 (train_with_dev=False, train_with_test=False)
2023-10-25 21:12:45,453 ----------------------------------------------------------------------------------------------------
2023-10-25 21:12:45,453 Training Params:
2023-10-25 21:12:45,453 - learning_rate: "5e-05"
2023-10-25 21:12:45,453 - mini_batch_size: "4"
2023-10-25 21:12:45,453 - max_epochs: "10"
2023-10-25 21:12:45,453 - shuffle: "True"
2023-10-25 21:12:45,453 ----------------------------------------------------------------------------------------------------
2023-10-25 21:12:45,453 Plugins:
2023-10-25 21:12:45,453 - TensorboardLogger
2023-10-25 21:12:45,453 - LinearScheduler | warmup_fraction: '0.1'
2023-10-25 21:12:45,453 ----------------------------------------------------------------------------------------------------
2023-10-25 21:12:45,453 Final evaluation on model from best epoch (best-model.pt)
2023-10-25 21:12:45,453 - metric: "('micro avg', 'f1-score')"
2023-10-25 21:12:45,453 ----------------------------------------------------------------------------------------------------
2023-10-25 21:12:45,453 Computation:
2023-10-25 21:12:45,453 - compute on device: cuda:0
2023-10-25 21:12:45,453 - embedding storage: none
2023-10-25 21:12:45,453 ----------------------------------------------------------------------------------------------------
2023-10-25 21:12:45,453 Model training base path: "hmbench-newseye/fi-dbmdz/bert-base-historic-multilingual-64k-td-cased-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-3"
2023-10-25 21:12:45,453 ----------------------------------------------------------------------------------------------------
2023-10-25 21:12:45,453 ----------------------------------------------------------------------------------------------------
2023-10-25 21:12:45,453 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-25 21:12:46,773 epoch 1 - iter 29/292 - loss 2.60630160 - time (sec): 1.32 - samples/sec: 2620.93 - lr: 0.000005 - momentum: 0.000000
2023-10-25 21:12:48,030 epoch 1 - iter 58/292 - loss 1.73093403 - time (sec): 2.58 - samples/sec: 2898.55 - lr: 0.000010 - momentum: 0.000000
2023-10-25 21:12:49,336 epoch 1 - iter 87/292 - loss 1.29007377 - time (sec): 3.88 - samples/sec: 3112.26 - lr: 0.000015 - momentum: 0.000000
2023-10-25 21:12:50,589 epoch 1 - iter 116/292 - loss 1.08928573 - time (sec): 5.13 - samples/sec: 3136.32 - lr: 0.000020 - momentum: 0.000000
2023-10-25 21:12:51,860 epoch 1 - iter 145/292 - loss 0.93038328 - time (sec): 6.41 - samples/sec: 3201.92 - lr: 0.000025 - momentum: 0.000000
2023-10-25 21:12:53,193 epoch 1 - iter 174/292 - loss 0.83340092 - time (sec): 7.74 - samples/sec: 3228.79 - lr: 0.000030 - momentum: 0.000000
2023-10-25 21:12:54,501 epoch 1 - iter 203/292 - loss 0.74787824 - time (sec): 9.05 - samples/sec: 3289.33 - lr: 0.000035 - momentum: 0.000000
2023-10-25 21:12:55,918 epoch 1 - iter 232/292 - loss 0.68293476 - time (sec): 10.46 - samples/sec: 3333.71 - lr: 0.000040 - momentum: 0.000000
2023-10-25 21:12:57,231 epoch 1 - iter 261/292 - loss 0.61828383 - time (sec): 11.78 - samples/sec: 3395.00 - lr: 0.000045 - momentum: 0.000000
2023-10-25 21:12:58,524 epoch 1 - iter 290/292 - loss 0.57952199 - time (sec): 13.07 - samples/sec: 3387.23 - lr: 0.000049 - momentum: 0.000000
2023-10-25 21:12:58,602 ----------------------------------------------------------------------------------------------------
2023-10-25 21:12:58,602 EPOCH 1 done: loss 0.5801 - lr: 0.000049
2023-10-25 21:12:59,117 DEV : loss 0.1393888294696808 - f1-score (micro avg) 0.6157
2023-10-25 21:12:59,121 saving best model
2023-10-25 21:12:59,633 ----------------------------------------------------------------------------------------------------
2023-10-25 21:13:00,938 epoch 2 - iter 29/292 - loss 0.16641729 - time (sec): 1.30 - samples/sec: 3551.69 - lr: 0.000049 - momentum: 0.000000
2023-10-25 21:13:02,295 epoch 2 - iter 58/292 - loss 0.14870205 - time (sec): 2.66 - samples/sec: 3657.47 - lr: 0.000049 - momentum: 0.000000
2023-10-25 21:13:03,566 epoch 2 - iter 87/292 - loss 0.15224226 - time (sec): 3.93 - samples/sec: 3558.93 - lr: 0.000048 - momentum: 0.000000
2023-10-25 21:13:04,863 epoch 2 - iter 116/292 - loss 0.15195587 - time (sec): 5.23 - samples/sec: 3491.95 - lr: 0.000048 - momentum: 0.000000
2023-10-25 21:13:06,114 epoch 2 - iter 145/292 - loss 0.15173369 - time (sec): 6.48 - samples/sec: 3425.51 - lr: 0.000047 - momentum: 0.000000
2023-10-25 21:13:07,298 epoch 2 - iter 174/292 - loss 0.16302891 - time (sec): 7.66 - samples/sec: 3398.01 - lr: 0.000047 - momentum: 0.000000
2023-10-25 21:13:08,492 epoch 2 - iter 203/292 - loss 0.16090612 - time (sec): 8.86 - samples/sec: 3414.98 - lr: 0.000046 - momentum: 0.000000
2023-10-25 21:13:09,795 epoch 2 - iter 232/292 - loss 0.15301547 - time (sec): 10.16 - samples/sec: 3440.05 - lr: 0.000046 - momentum: 0.000000
2023-10-25 21:13:11,060 epoch 2 - iter 261/292 - loss 0.15185628 - time (sec): 11.43 - samples/sec: 3468.19 - lr: 0.000045 - momentum: 0.000000
2023-10-25 21:13:12,335 epoch 2 - iter 290/292 - loss 0.15180422 - time (sec): 12.70 - samples/sec: 3473.70 - lr: 0.000045 - momentum: 0.000000
2023-10-25 21:13:12,422 ----------------------------------------------------------------------------------------------------
2023-10-25 21:13:12,423 EPOCH 2 done: loss 0.1516 - lr: 0.000045
2023-10-25 21:13:13,340 DEV : loss 0.1067458763718605 - f1-score (micro avg) 0.7253
2023-10-25 21:13:13,344 saving best model
2023-10-25 21:13:14,183 ----------------------------------------------------------------------------------------------------
2023-10-25 21:13:15,560 epoch 3 - iter 29/292 - loss 0.07559153 - time (sec): 1.37 - samples/sec: 3251.25 - lr: 0.000044 - momentum: 0.000000
2023-10-25 21:13:16,805 epoch 3 - iter 58/292 - loss 0.07386580 - time (sec): 2.62 - samples/sec: 2941.45 - lr: 0.000043 - momentum: 0.000000
2023-10-25 21:13:18,136 epoch 3 - iter 87/292 - loss 0.07463057 - time (sec): 3.95 - samples/sec: 3118.77 - lr: 0.000043 - momentum: 0.000000
2023-10-25 21:13:19,407 epoch 3 - iter 116/292 - loss 0.07309791 - time (sec): 5.22 - samples/sec: 3058.69 - lr: 0.000042 - momentum: 0.000000
2023-10-25 21:13:20,812 epoch 3 - iter 145/292 - loss 0.07859240 - time (sec): 6.62 - samples/sec: 3310.07 - lr: 0.000042 - momentum: 0.000000
2023-10-25 21:13:22,125 epoch 3 - iter 174/292 - loss 0.08064284 - time (sec): 7.94 - samples/sec: 3359.83 - lr: 0.000041 - momentum: 0.000000
2023-10-25 21:13:23,411 epoch 3 - iter 203/292 - loss 0.08000187 - time (sec): 9.22 - samples/sec: 3387.08 - lr: 0.000041 - momentum: 0.000000
2023-10-25 21:13:24,634 epoch 3 - iter 232/292 - loss 0.07888199 - time (sec): 10.45 - samples/sec: 3370.63 - lr: 0.000040 - momentum: 0.000000
2023-10-25 21:13:25,866 epoch 3 - iter 261/292 - loss 0.07895832 - time (sec): 11.68 - samples/sec: 3351.73 - lr: 0.000040 - momentum: 0.000000
2023-10-25 21:13:27,159 epoch 3 - iter 290/292 - loss 0.08051650 - time (sec): 12.97 - samples/sec: 3374.19 - lr: 0.000039 - momentum: 0.000000
2023-10-25 21:13:27,266 ----------------------------------------------------------------------------------------------------
2023-10-25 21:13:27,266 EPOCH 3 done: loss 0.0807 - lr: 0.000039
2023-10-25 21:13:28,186 DEV : loss 0.10645782947540283 - f1-score (micro avg) 0.7273
2023-10-25 21:13:28,190 saving best model
2023-10-25 21:13:28,853 ----------------------------------------------------------------------------------------------------
2023-10-25 21:13:30,194 epoch 4 - iter 29/292 - loss 0.05879957 - time (sec): 1.34 - samples/sec: 3783.44 - lr: 0.000038 - momentum: 0.000000
2023-10-25 21:13:31,451 epoch 4 - iter 58/292 - loss 0.05787533 - time (sec): 2.60 - samples/sec: 3488.46 - lr: 0.000038 - momentum: 0.000000
2023-10-25 21:13:32,757 epoch 4 - iter 87/292 - loss 0.05964165 - time (sec): 3.90 - samples/sec: 3404.79 - lr: 0.000037 - momentum: 0.000000
2023-10-25 21:13:34,060 epoch 4 - iter 116/292 - loss 0.05624702 - time (sec): 5.21 - samples/sec: 3327.34 - lr: 0.000037 - momentum: 0.000000
2023-10-25 21:13:35,289 epoch 4 - iter 145/292 - loss 0.05478522 - time (sec): 6.43 - samples/sec: 3263.01 - lr: 0.000036 - momentum: 0.000000
2023-10-25 21:13:36,625 epoch 4 - iter 174/292 - loss 0.05936602 - time (sec): 7.77 - samples/sec: 3299.85 - lr: 0.000036 - momentum: 0.000000
2023-10-25 21:13:37,824 epoch 4 - iter 203/292 - loss 0.05877258 - time (sec): 8.97 - samples/sec: 3309.89 - lr: 0.000035 - momentum: 0.000000
2023-10-25 21:13:39,052 epoch 4 - iter 232/292 - loss 0.05534208 - time (sec): 10.20 - samples/sec: 3293.06 - lr: 0.000035 - momentum: 0.000000
2023-10-25 21:13:40,403 epoch 4 - iter 261/292 - loss 0.05738089 - time (sec): 11.55 - samples/sec: 3384.13 - lr: 0.000034 - momentum: 0.000000
2023-10-25 21:13:41,720 epoch 4 - iter 290/292 - loss 0.05628166 - time (sec): 12.86 - samples/sec: 3443.39 - lr: 0.000033 - momentum: 0.000000
2023-10-25 21:13:41,808 ----------------------------------------------------------------------------------------------------
2023-10-25 21:13:41,808 EPOCH 4 done: loss 0.0561 - lr: 0.000033
2023-10-25 21:13:42,726 DEV : loss 0.13811993598937988 - f1-score (micro avg) 0.7425
2023-10-25 21:13:42,730 saving best model
2023-10-25 21:13:43,402 ----------------------------------------------------------------------------------------------------
2023-10-25 21:13:44,735 epoch 5 - iter 29/292 - loss 0.02917983 - time (sec): 1.33 - samples/sec: 3600.95 - lr: 0.000033 - momentum: 0.000000
2023-10-25 21:13:45,986 epoch 5 - iter 58/292 - loss 0.03003225 - time (sec): 2.58 - samples/sec: 3317.20 - lr: 0.000032 - momentum: 0.000000
2023-10-25 21:13:47,328 epoch 5 - iter 87/292 - loss 0.03965107 - time (sec): 3.92 - samples/sec: 3485.40 - lr: 0.000032 - momentum: 0.000000
2023-10-25 21:13:48,619 epoch 5 - iter 116/292 - loss 0.04015937 - time (sec): 5.21 - samples/sec: 3403.14 - lr: 0.000031 - momentum: 0.000000
2023-10-25 21:13:49,919 epoch 5 - iter 145/292 - loss 0.03738301 - time (sec): 6.51 - samples/sec: 3370.76 - lr: 0.000031 - momentum: 0.000000
2023-10-25 21:13:51,199 epoch 5 - iter 174/292 - loss 0.03512502 - time (sec): 7.79 - samples/sec: 3305.98 - lr: 0.000030 - momentum: 0.000000
2023-10-25 21:13:52,517 epoch 5 - iter 203/292 - loss 0.03795555 - time (sec): 9.11 - samples/sec: 3285.02 - lr: 0.000030 - momentum: 0.000000
2023-10-25 21:13:53,848 epoch 5 - iter 232/292 - loss 0.04076074 - time (sec): 10.44 - samples/sec: 3339.81 - lr: 0.000029 - momentum: 0.000000
2023-10-25 21:13:55,256 epoch 5 - iter 261/292 - loss 0.03934850 - time (sec): 11.85 - samples/sec: 3380.71 - lr: 0.000028 - momentum: 0.000000
2023-10-25 21:13:56,543 epoch 5 - iter 290/292 - loss 0.04048060 - time (sec): 13.14 - samples/sec: 3374.10 - lr: 0.000028 - momentum: 0.000000
2023-10-25 21:13:56,617 ----------------------------------------------------------------------------------------------------
2023-10-25 21:13:56,617 EPOCH 5 done: loss 0.0404 - lr: 0.000028
2023-10-25 21:13:57,540 DEV : loss 0.12971803545951843 - f1-score (micro avg) 0.7451
2023-10-25 21:13:57,544 saving best model
2023-10-25 21:13:58,224 ----------------------------------------------------------------------------------------------------
2023-10-25 21:13:59,472 epoch 6 - iter 29/292 - loss 0.02180752 - time (sec): 1.24 - samples/sec: 3433.49 - lr: 0.000027 - momentum: 0.000000
2023-10-25 21:14:00,795 epoch 6 - iter 58/292 - loss 0.02936813 - time (sec): 2.57 - samples/sec: 3411.13 - lr: 0.000027 - momentum: 0.000000
2023-10-25 21:14:02,143 epoch 6 - iter 87/292 - loss 0.02555129 - time (sec): 3.91 - samples/sec: 3498.75 - lr: 0.000026 - momentum: 0.000000
2023-10-25 21:14:03,427 epoch 6 - iter 116/292 - loss 0.02657167 - time (sec): 5.20 - samples/sec: 3501.91 - lr: 0.000026 - momentum: 0.000000
2023-10-25 21:14:04,721 epoch 6 - iter 145/292 - loss 0.02565577 - time (sec): 6.49 - samples/sec: 3419.27 - lr: 0.000025 - momentum: 0.000000
2023-10-25 21:14:06,062 epoch 6 - iter 174/292 - loss 0.02343463 - time (sec): 7.83 - samples/sec: 3344.78 - lr: 0.000025 - momentum: 0.000000
2023-10-25 21:14:07,348 epoch 6 - iter 203/292 - loss 0.02416227 - time (sec): 9.12 - samples/sec: 3368.83 - lr: 0.000024 - momentum: 0.000000
2023-10-25 21:14:08,679 epoch 6 - iter 232/292 - loss 0.02927410 - time (sec): 10.45 - samples/sec: 3384.05 - lr: 0.000023 - momentum: 0.000000
2023-10-25 21:14:09,995 epoch 6 - iter 261/292 - loss 0.03021667 - time (sec): 11.77 - samples/sec: 3401.67 - lr: 0.000023 - momentum: 0.000000
2023-10-25 21:14:11,280 epoch 6 - iter 290/292 - loss 0.03036653 - time (sec): 13.05 - samples/sec: 3389.93 - lr: 0.000022 - momentum: 0.000000
2023-10-25 21:14:11,367 ----------------------------------------------------------------------------------------------------
2023-10-25 21:14:11,367 EPOCH 6 done: loss 0.0307 - lr: 0.000022
2023-10-25 21:14:12,290 DEV : loss 0.15307584404945374 - f1-score (micro avg) 0.7665
2023-10-25 21:14:12,294 saving best model
2023-10-25 21:14:12,958 ----------------------------------------------------------------------------------------------------
2023-10-25 21:14:14,228 epoch 7 - iter 29/292 - loss 0.02367547 - time (sec): 1.27 - samples/sec: 3400.04 - lr: 0.000022 - momentum: 0.000000
2023-10-25 21:14:15,919 epoch 7 - iter 58/292 - loss 0.02823955 - time (sec): 2.96 - samples/sec: 3480.61 - lr: 0.000021 - momentum: 0.000000
2023-10-25 21:14:17,150 epoch 7 - iter 87/292 - loss 0.02673198 - time (sec): 4.19 - samples/sec: 3329.46 - lr: 0.000021 - momentum: 0.000000
2023-10-25 21:14:18,398 epoch 7 - iter 116/292 - loss 0.02646895 - time (sec): 5.44 - samples/sec: 3258.84 - lr: 0.000020 - momentum: 0.000000
2023-10-25 21:14:19,722 epoch 7 - iter 145/292 - loss 0.02552160 - time (sec): 6.76 - samples/sec: 3321.63 - lr: 0.000020 - momentum: 0.000000
2023-10-25 21:14:21,003 epoch 7 - iter 174/292 - loss 0.02323132 - time (sec): 8.04 - samples/sec: 3367.73 - lr: 0.000019 - momentum: 0.000000
2023-10-25 21:14:22,256 epoch 7 - iter 203/292 - loss 0.02225006 - time (sec): 9.29 - samples/sec: 3393.37 - lr: 0.000018 - momentum: 0.000000
2023-10-25 21:14:23,470 epoch 7 - iter 232/292 - loss 0.02294774 - time (sec): 10.51 - samples/sec: 3343.71 - lr: 0.000018 - momentum: 0.000000
2023-10-25 21:14:24,767 epoch 7 - iter 261/292 - loss 0.02407455 - time (sec): 11.80 - samples/sec: 3361.55 - lr: 0.000017 - momentum: 0.000000
2023-10-25 21:14:26,083 epoch 7 - iter 290/292 - loss 0.02328112 - time (sec): 13.12 - samples/sec: 3374.12 - lr: 0.000017 - momentum: 0.000000
2023-10-25 21:14:26,163 ----------------------------------------------------------------------------------------------------
2023-10-25 21:14:26,163 EPOCH 7 done: loss 0.0232 - lr: 0.000017
2023-10-25 21:14:27,082 DEV : loss 0.1477293223142624 - f1-score (micro avg) 0.787
2023-10-25 21:14:27,087 saving best model
2023-10-25 21:14:27,754 ----------------------------------------------------------------------------------------------------
2023-10-25 21:14:29,068 epoch 8 - iter 29/292 - loss 0.04019118 - time (sec): 1.31 - samples/sec: 3033.76 - lr: 0.000016 - momentum: 0.000000
2023-10-25 21:14:30,486 epoch 8 - iter 58/292 - loss 0.02562495 - time (sec): 2.73 - samples/sec: 3338.48 - lr: 0.000016 - momentum: 0.000000
2023-10-25 21:14:31,835 epoch 8 - iter 87/292 - loss 0.02135462 - time (sec): 4.08 - samples/sec: 3416.50 - lr: 0.000015 - momentum: 0.000000
2023-10-25 21:14:33,120 epoch 8 - iter 116/292 - loss 0.02053731 - time (sec): 5.36 - samples/sec: 3408.96 - lr: 0.000015 - momentum: 0.000000
2023-10-25 21:14:34,425 epoch 8 - iter 145/292 - loss 0.02012399 - time (sec): 6.67 - samples/sec: 3372.65 - lr: 0.000014 - momentum: 0.000000
2023-10-25 21:14:35,694 epoch 8 - iter 174/292 - loss 0.01842217 - time (sec): 7.94 - samples/sec: 3388.27 - lr: 0.000013 - momentum: 0.000000
2023-10-25 21:14:36,985 epoch 8 - iter 203/292 - loss 0.01663993 - time (sec): 9.23 - samples/sec: 3331.69 - lr: 0.000013 - momentum: 0.000000
2023-10-25 21:14:38,289 epoch 8 - iter 232/292 - loss 0.01669885 - time (sec): 10.53 - samples/sec: 3367.99 - lr: 0.000012 - momentum: 0.000000
2023-10-25 21:14:39,605 epoch 8 - iter 261/292 - loss 0.01668062 - time (sec): 11.85 - samples/sec: 3368.50 - lr: 0.000012 - momentum: 0.000000
2023-10-25 21:14:40,885 epoch 8 - iter 290/292 - loss 0.01628557 - time (sec): 13.13 - samples/sec: 3376.27 - lr: 0.000011 - momentum: 0.000000
2023-10-25 21:14:40,973 ----------------------------------------------------------------------------------------------------
2023-10-25 21:14:40,973 EPOCH 8 done: loss 0.0162 - lr: 0.000011
2023-10-25 21:14:41,893 DEV : loss 0.1719065010547638 - f1-score (micro avg) 0.7582
2023-10-25 21:14:41,897 ----------------------------------------------------------------------------------------------------
2023-10-25 21:14:43,194 epoch 9 - iter 29/292 - loss 0.00358527 - time (sec): 1.30 - samples/sec: 3663.02 - lr: 0.000011 - momentum: 0.000000
2023-10-25 21:14:44,499 epoch 9 - iter 58/292 - loss 0.00492909 - time (sec): 2.60 - samples/sec: 3524.73 - lr: 0.000010 - momentum: 0.000000
2023-10-25 21:14:45,755 epoch 9 - iter 87/292 - loss 0.00373125 - time (sec): 3.86 - samples/sec: 3372.51 - lr: 0.000010 - momentum: 0.000000
2023-10-25 21:14:47,134 epoch 9 - iter 116/292 - loss 0.00767643 - time (sec): 5.24 - samples/sec: 3419.71 - lr: 0.000009 - momentum: 0.000000
2023-10-25 21:14:48,469 epoch 9 - iter 145/292 - loss 0.01032308 - time (sec): 6.57 - samples/sec: 3470.78 - lr: 0.000008 - momentum: 0.000000
2023-10-25 21:14:49,777 epoch 9 - iter 174/292 - loss 0.01017440 - time (sec): 7.88 - samples/sec: 3476.45 - lr: 0.000008 - momentum: 0.000000
2023-10-25 21:14:51,085 epoch 9 - iter 203/292 - loss 0.00944404 - time (sec): 9.19 - samples/sec: 3441.81 - lr: 0.000007 - momentum: 0.000000
2023-10-25 21:14:52,353 epoch 9 - iter 232/292 - loss 0.01116023 - time (sec): 10.45 - samples/sec: 3406.07 - lr: 0.000007 - momentum: 0.000000
2023-10-25 21:14:53,666 epoch 9 - iter 261/292 - loss 0.01067018 - time (sec): 11.77 - samples/sec: 3371.12 - lr: 0.000006 - momentum: 0.000000
2023-10-25 21:14:54,957 epoch 9 - iter 290/292 - loss 0.00991839 - time (sec): 13.06 - samples/sec: 3384.07 - lr: 0.000006 - momentum: 0.000000
2023-10-25 21:14:55,046 ----------------------------------------------------------------------------------------------------
2023-10-25 21:14:55,047 EPOCH 9 done: loss 0.0099 - lr: 0.000006
2023-10-25 21:14:55,967 DEV : loss 0.17519107460975647 - f1-score (micro avg) 0.7682
2023-10-25 21:14:55,971 ----------------------------------------------------------------------------------------------------
2023-10-25 21:14:57,251 epoch 10 - iter 29/292 - loss 0.00493698 - time (sec): 1.28 - samples/sec: 3374.74 - lr: 0.000005 - momentum: 0.000000
2023-10-25 21:14:58,512 epoch 10 - iter 58/292 - loss 0.00797778 - time (sec): 2.54 - samples/sec: 3427.40 - lr: 0.000005 - momentum: 0.000000
2023-10-25 21:14:59,849 epoch 10 - iter 87/292 - loss 0.01121526 - time (sec): 3.88 - samples/sec: 3401.36 - lr: 0.000004 - momentum: 0.000000
2023-10-25 21:15:01,143 epoch 10 - iter 116/292 - loss 0.00908355 - time (sec): 5.17 - samples/sec: 3370.72 - lr: 0.000003 - momentum: 0.000000
2023-10-25 21:15:02,436 epoch 10 - iter 145/292 - loss 0.00825735 - time (sec): 6.46 - samples/sec: 3348.68 - lr: 0.000003 - momentum: 0.000000
2023-10-25 21:15:03,716 epoch 10 - iter 174/292 - loss 0.00808191 - time (sec): 7.74 - samples/sec: 3310.53 - lr: 0.000002 - momentum: 0.000000
2023-10-25 21:15:05,138 epoch 10 - iter 203/292 - loss 0.00845788 - time (sec): 9.17 - samples/sec: 3386.32 - lr: 0.000002 - momentum: 0.000000
2023-10-25 21:15:06,420 epoch 10 - iter 232/292 - loss 0.00783064 - time (sec): 10.45 - samples/sec: 3375.44 - lr: 0.000001 - momentum: 0.000000
2023-10-25 21:15:07,819 epoch 10 - iter 261/292 - loss 0.00786645 - time (sec): 11.85 - samples/sec: 3371.73 - lr: 0.000001 - momentum: 0.000000
2023-10-25 21:15:09,104 epoch 10 - iter 290/292 - loss 0.00733907 - time (sec): 13.13 - samples/sec: 3362.91 - lr: 0.000000 - momentum: 0.000000
2023-10-25 21:15:09,186 ----------------------------------------------------------------------------------------------------
2023-10-25 21:15:09,187 EPOCH 10 done: loss 0.0073 - lr: 0.000000
2023-10-25 21:15:10,107 DEV : loss 0.177558034658432 - f1-score (micro avg) 0.7613
2023-10-25 21:15:10,525 ----------------------------------------------------------------------------------------------------
2023-10-25 21:15:10,527 Loading model from best epoch ...
2023-10-25 21:15:12,151 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
2023-10-25 21:15:13,880
Results:
- F-score (micro) 0.7417
- F-score (macro) 0.6377
- Accuracy 0.6109
By class:
precision recall f1-score support
PER 0.7983 0.8190 0.8085 348
LOC 0.6450 0.8352 0.7279 261
ORG 0.4390 0.3462 0.3871 52
HumanProd 0.5517 0.7273 0.6275 22
micro avg 0.7020 0.7862 0.7417 683
macro avg 0.6085 0.6819 0.6377 683
weighted avg 0.7044 0.7862 0.7398 683
2023-10-25 21:15:13,880 ----------------------------------------------------------------------------------------------------
|