File size: 24,113 Bytes
5aa271d |
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 18:19:41,179 ----------------------------------------------------------------------------------------------------
2023-10-23 18:19:41,180 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 18:19:41,181 ----------------------------------------------------------------------------------------------------
2023-10-23 18:19:41,181 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
2023-10-23 18:19:41,181 ----------------------------------------------------------------------------------------------------
2023-10-23 18:19:41,181 Train: 1214 sentences
2023-10-23 18:19:41,181 (train_with_dev=False, train_with_test=False)
2023-10-23 18:19:41,181 ----------------------------------------------------------------------------------------------------
2023-10-23 18:19:41,181 Training Params:
2023-10-23 18:19:41,181 - learning_rate: "3e-05"
2023-10-23 18:19:41,181 - mini_batch_size: "8"
2023-10-23 18:19:41,181 - max_epochs: "10"
2023-10-23 18:19:41,181 - shuffle: "True"
2023-10-23 18:19:41,181 ----------------------------------------------------------------------------------------------------
2023-10-23 18:19:41,181 Plugins:
2023-10-23 18:19:41,181 - TensorboardLogger
2023-10-23 18:19:41,181 - LinearScheduler | warmup_fraction: '0.1'
2023-10-23 18:19:41,181 ----------------------------------------------------------------------------------------------------
2023-10-23 18:19:41,181 Final evaluation on model from best epoch (best-model.pt)
2023-10-23 18:19:41,181 - metric: "('micro avg', 'f1-score')"
2023-10-23 18:19:41,181 ----------------------------------------------------------------------------------------------------
2023-10-23 18:19:41,182 Computation:
2023-10-23 18:19:41,182 - compute on device: cuda:0
2023-10-23 18:19:41,182 - embedding storage: none
2023-10-23 18:19:41,182 ----------------------------------------------------------------------------------------------------
2023-10-23 18:19:41,182 Model training base path: "hmbench-ajmc/en-dbmdz/bert-base-historic-multilingual-64k-td-cased-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-3"
2023-10-23 18:19:41,182 ----------------------------------------------------------------------------------------------------
2023-10-23 18:19:41,182 ----------------------------------------------------------------------------------------------------
2023-10-23 18:19:41,182 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-23 18:19:41,949 epoch 1 - iter 15/152 - loss 3.11385143 - time (sec): 0.77 - samples/sec: 3973.24 - lr: 0.000003 - momentum: 0.000000
2023-10-23 18:19:42,708 epoch 1 - iter 30/152 - loss 2.66819735 - time (sec): 1.53 - samples/sec: 3873.24 - lr: 0.000006 - momentum: 0.000000
2023-10-23 18:19:43,469 epoch 1 - iter 45/152 - loss 2.03026616 - time (sec): 2.29 - samples/sec: 3895.80 - lr: 0.000009 - momentum: 0.000000
2023-10-23 18:19:44,229 epoch 1 - iter 60/152 - loss 1.70709060 - time (sec): 3.05 - samples/sec: 3932.94 - lr: 0.000012 - momentum: 0.000000
2023-10-23 18:19:44,983 epoch 1 - iter 75/152 - loss 1.46041251 - time (sec): 3.80 - samples/sec: 3963.43 - lr: 0.000015 - momentum: 0.000000
2023-10-23 18:19:45,749 epoch 1 - iter 90/152 - loss 1.28362004 - time (sec): 4.57 - samples/sec: 3980.01 - lr: 0.000018 - momentum: 0.000000
2023-10-23 18:19:46,529 epoch 1 - iter 105/152 - loss 1.14421153 - time (sec): 5.35 - samples/sec: 3984.53 - lr: 0.000021 - momentum: 0.000000
2023-10-23 18:19:47,287 epoch 1 - iter 120/152 - loss 1.04230113 - time (sec): 6.10 - samples/sec: 3989.91 - lr: 0.000023 - momentum: 0.000000
2023-10-23 18:19:48,056 epoch 1 - iter 135/152 - loss 0.94143217 - time (sec): 6.87 - samples/sec: 4014.79 - lr: 0.000026 - momentum: 0.000000
2023-10-23 18:19:48,811 epoch 1 - iter 150/152 - loss 0.87310563 - time (sec): 7.63 - samples/sec: 4009.45 - lr: 0.000029 - momentum: 0.000000
2023-10-23 18:19:48,918 ----------------------------------------------------------------------------------------------------
2023-10-23 18:19:48,918 EPOCH 1 done: loss 0.8644 - lr: 0.000029
2023-10-23 18:19:49,725 DEV : loss 0.19825053215026855 - f1-score (micro avg) 0.5945
2023-10-23 18:19:49,733 saving best model
2023-10-23 18:19:50,168 ----------------------------------------------------------------------------------------------------
2023-10-23 18:19:50,956 epoch 2 - iter 15/152 - loss 0.20761390 - time (sec): 0.79 - samples/sec: 3941.37 - lr: 0.000030 - momentum: 0.000000
2023-10-23 18:19:51,725 epoch 2 - iter 30/152 - loss 0.17537014 - time (sec): 1.56 - samples/sec: 4053.34 - lr: 0.000029 - momentum: 0.000000
2023-10-23 18:19:52,510 epoch 2 - iter 45/152 - loss 0.16433025 - time (sec): 2.34 - samples/sec: 3933.86 - lr: 0.000029 - momentum: 0.000000
2023-10-23 18:19:53,280 epoch 2 - iter 60/152 - loss 0.16227219 - time (sec): 3.11 - samples/sec: 3925.61 - lr: 0.000029 - momentum: 0.000000
2023-10-23 18:19:54,051 epoch 2 - iter 75/152 - loss 0.15897537 - time (sec): 3.88 - samples/sec: 3896.99 - lr: 0.000028 - momentum: 0.000000
2023-10-23 18:19:54,821 epoch 2 - iter 90/152 - loss 0.15189330 - time (sec): 4.65 - samples/sec: 3916.47 - lr: 0.000028 - momentum: 0.000000
2023-10-23 18:19:55,588 epoch 2 - iter 105/152 - loss 0.15322637 - time (sec): 5.42 - samples/sec: 3864.70 - lr: 0.000028 - momentum: 0.000000
2023-10-23 18:19:56,352 epoch 2 - iter 120/152 - loss 0.15144396 - time (sec): 6.18 - samples/sec: 3911.47 - lr: 0.000027 - momentum: 0.000000
2023-10-23 18:19:57,182 epoch 2 - iter 135/152 - loss 0.14911332 - time (sec): 7.01 - samples/sec: 3892.71 - lr: 0.000027 - momentum: 0.000000
2023-10-23 18:19:57,954 epoch 2 - iter 150/152 - loss 0.14321232 - time (sec): 7.78 - samples/sec: 3934.72 - lr: 0.000027 - momentum: 0.000000
2023-10-23 18:19:58,052 ----------------------------------------------------------------------------------------------------
2023-10-23 18:19:58,052 EPOCH 2 done: loss 0.1424 - lr: 0.000027
2023-10-23 18:19:58,905 DEV : loss 0.14702850580215454 - f1-score (micro avg) 0.7924
2023-10-23 18:19:58,913 saving best model
2023-10-23 18:19:59,472 ----------------------------------------------------------------------------------------------------
2023-10-23 18:20:00,334 epoch 3 - iter 15/152 - loss 0.06391830 - time (sec): 0.86 - samples/sec: 3370.68 - lr: 0.000026 - momentum: 0.000000
2023-10-23 18:20:01,135 epoch 3 - iter 30/152 - loss 0.06920391 - time (sec): 1.66 - samples/sec: 3759.95 - lr: 0.000026 - momentum: 0.000000
2023-10-23 18:20:01,975 epoch 3 - iter 45/152 - loss 0.08444799 - time (sec): 2.50 - samples/sec: 3780.32 - lr: 0.000026 - momentum: 0.000000
2023-10-23 18:20:02,847 epoch 3 - iter 60/152 - loss 0.08218897 - time (sec): 3.37 - samples/sec: 3685.87 - lr: 0.000025 - momentum: 0.000000
2023-10-23 18:20:03,565 epoch 3 - iter 75/152 - loss 0.07841224 - time (sec): 4.09 - samples/sec: 3839.98 - lr: 0.000025 - momentum: 0.000000
2023-10-23 18:20:04,310 epoch 3 - iter 90/152 - loss 0.07548322 - time (sec): 4.84 - samples/sec: 3852.36 - lr: 0.000025 - momentum: 0.000000
2023-10-23 18:20:05,182 epoch 3 - iter 105/152 - loss 0.07717780 - time (sec): 5.71 - samples/sec: 3766.44 - lr: 0.000024 - momentum: 0.000000
2023-10-23 18:20:06,039 epoch 3 - iter 120/152 - loss 0.07553394 - time (sec): 6.57 - samples/sec: 3780.26 - lr: 0.000024 - momentum: 0.000000
2023-10-23 18:20:06,888 epoch 3 - iter 135/152 - loss 0.07497682 - time (sec): 7.41 - samples/sec: 3727.66 - lr: 0.000024 - momentum: 0.000000
2023-10-23 18:20:07,685 epoch 3 - iter 150/152 - loss 0.08015255 - time (sec): 8.21 - samples/sec: 3735.74 - lr: 0.000023 - momentum: 0.000000
2023-10-23 18:20:07,797 ----------------------------------------------------------------------------------------------------
2023-10-23 18:20:07,797 EPOCH 3 done: loss 0.0807 - lr: 0.000023
2023-10-23 18:20:08,644 DEV : loss 0.14846408367156982 - f1-score (micro avg) 0.7875
2023-10-23 18:20:08,651 ----------------------------------------------------------------------------------------------------
2023-10-23 18:20:09,522 epoch 4 - iter 15/152 - loss 0.04146583 - time (sec): 0.87 - samples/sec: 3878.14 - lr: 0.000023 - momentum: 0.000000
2023-10-23 18:20:10,377 epoch 4 - iter 30/152 - loss 0.04382436 - time (sec): 1.72 - samples/sec: 3770.97 - lr: 0.000023 - momentum: 0.000000
2023-10-23 18:20:11,250 epoch 4 - iter 45/152 - loss 0.05477412 - time (sec): 2.60 - samples/sec: 3588.03 - lr: 0.000022 - momentum: 0.000000
2023-10-23 18:20:12,105 epoch 4 - iter 60/152 - loss 0.04640008 - time (sec): 3.45 - samples/sec: 3617.40 - lr: 0.000022 - momentum: 0.000000
2023-10-23 18:20:12,965 epoch 4 - iter 75/152 - loss 0.04581229 - time (sec): 4.31 - samples/sec: 3590.01 - lr: 0.000022 - momentum: 0.000000
2023-10-23 18:20:13,844 epoch 4 - iter 90/152 - loss 0.05157112 - time (sec): 5.19 - samples/sec: 3570.66 - lr: 0.000021 - momentum: 0.000000
2023-10-23 18:20:14,678 epoch 4 - iter 105/152 - loss 0.06028727 - time (sec): 6.02 - samples/sec: 3591.59 - lr: 0.000021 - momentum: 0.000000
2023-10-23 18:20:15,526 epoch 4 - iter 120/152 - loss 0.05892637 - time (sec): 6.87 - samples/sec: 3587.02 - lr: 0.000021 - momentum: 0.000000
2023-10-23 18:20:16,402 epoch 4 - iter 135/152 - loss 0.05678927 - time (sec): 7.75 - samples/sec: 3571.03 - lr: 0.000020 - momentum: 0.000000
2023-10-23 18:20:17,271 epoch 4 - iter 150/152 - loss 0.05572967 - time (sec): 8.62 - samples/sec: 3548.65 - lr: 0.000020 - momentum: 0.000000
2023-10-23 18:20:17,386 ----------------------------------------------------------------------------------------------------
2023-10-23 18:20:17,386 EPOCH 4 done: loss 0.0552 - lr: 0.000020
2023-10-23 18:20:18,253 DEV : loss 0.1560734659433365 - f1-score (micro avg) 0.8216
2023-10-23 18:20:18,260 saving best model
2023-10-23 18:20:18,829 ----------------------------------------------------------------------------------------------------
2023-10-23 18:20:19,661 epoch 5 - iter 15/152 - loss 0.01895573 - time (sec): 0.83 - samples/sec: 3785.59 - lr: 0.000020 - momentum: 0.000000
2023-10-23 18:20:20,531 epoch 5 - iter 30/152 - loss 0.03738480 - time (sec): 1.70 - samples/sec: 3588.93 - lr: 0.000019 - momentum: 0.000000
2023-10-23 18:20:21,386 epoch 5 - iter 45/152 - loss 0.04663379 - time (sec): 2.56 - samples/sec: 3658.12 - lr: 0.000019 - momentum: 0.000000
2023-10-23 18:20:22,235 epoch 5 - iter 60/152 - loss 0.04280235 - time (sec): 3.40 - samples/sec: 3627.69 - lr: 0.000019 - momentum: 0.000000
2023-10-23 18:20:23,086 epoch 5 - iter 75/152 - loss 0.04025851 - time (sec): 4.25 - samples/sec: 3688.14 - lr: 0.000018 - momentum: 0.000000
2023-10-23 18:20:23,909 epoch 5 - iter 90/152 - loss 0.03707620 - time (sec): 5.08 - samples/sec: 3678.00 - lr: 0.000018 - momentum: 0.000000
2023-10-23 18:20:24,760 epoch 5 - iter 105/152 - loss 0.03528057 - time (sec): 5.93 - samples/sec: 3658.21 - lr: 0.000018 - momentum: 0.000000
2023-10-23 18:20:25,609 epoch 5 - iter 120/152 - loss 0.03517572 - time (sec): 6.78 - samples/sec: 3647.27 - lr: 0.000017 - momentum: 0.000000
2023-10-23 18:20:26,467 epoch 5 - iter 135/152 - loss 0.03659409 - time (sec): 7.64 - samples/sec: 3614.77 - lr: 0.000017 - momentum: 0.000000
2023-10-23 18:20:27,329 epoch 5 - iter 150/152 - loss 0.03748552 - time (sec): 8.50 - samples/sec: 3591.66 - lr: 0.000017 - momentum: 0.000000
2023-10-23 18:20:27,433 ----------------------------------------------------------------------------------------------------
2023-10-23 18:20:27,433 EPOCH 5 done: loss 0.0388 - lr: 0.000017
2023-10-23 18:20:28,329 DEV : loss 0.17505785822868347 - f1-score (micro avg) 0.82
2023-10-23 18:20:28,337 ----------------------------------------------------------------------------------------------------
2023-10-23 18:20:29,209 epoch 6 - iter 15/152 - loss 0.07671918 - time (sec): 0.87 - samples/sec: 3587.02 - lr: 0.000016 - momentum: 0.000000
2023-10-23 18:20:30,069 epoch 6 - iter 30/152 - loss 0.04659880 - time (sec): 1.73 - samples/sec: 3469.36 - lr: 0.000016 - momentum: 0.000000
2023-10-23 18:20:30,926 epoch 6 - iter 45/152 - loss 0.04527842 - time (sec): 2.59 - samples/sec: 3568.63 - lr: 0.000016 - momentum: 0.000000
2023-10-23 18:20:31,782 epoch 6 - iter 60/152 - loss 0.03656474 - time (sec): 3.44 - samples/sec: 3532.81 - lr: 0.000015 - momentum: 0.000000
2023-10-23 18:20:32,631 epoch 6 - iter 75/152 - loss 0.03768485 - time (sec): 4.29 - samples/sec: 3528.16 - lr: 0.000015 - momentum: 0.000000
2023-10-23 18:20:33,496 epoch 6 - iter 90/152 - loss 0.03359681 - time (sec): 5.16 - samples/sec: 3495.68 - lr: 0.000015 - momentum: 0.000000
2023-10-23 18:20:34,370 epoch 6 - iter 105/152 - loss 0.03221236 - time (sec): 6.03 - samples/sec: 3500.09 - lr: 0.000014 - momentum: 0.000000
2023-10-23 18:20:35,205 epoch 6 - iter 120/152 - loss 0.03110637 - time (sec): 6.87 - samples/sec: 3552.19 - lr: 0.000014 - momentum: 0.000000
2023-10-23 18:20:36,059 epoch 6 - iter 135/152 - loss 0.03086084 - time (sec): 7.72 - samples/sec: 3549.93 - lr: 0.000014 - momentum: 0.000000
2023-10-23 18:20:36,922 epoch 6 - iter 150/152 - loss 0.02867945 - time (sec): 8.58 - samples/sec: 3575.30 - lr: 0.000013 - momentum: 0.000000
2023-10-23 18:20:37,023 ----------------------------------------------------------------------------------------------------
2023-10-23 18:20:37,023 EPOCH 6 done: loss 0.0286 - lr: 0.000013
2023-10-23 18:20:37,897 DEV : loss 0.213690847158432 - f1-score (micro avg) 0.8042
2023-10-23 18:20:37,904 ----------------------------------------------------------------------------------------------------
2023-10-23 18:20:38,775 epoch 7 - iter 15/152 - loss 0.01848948 - time (sec): 0.87 - samples/sec: 3605.57 - lr: 0.000013 - momentum: 0.000000
2023-10-23 18:20:39,649 epoch 7 - iter 30/152 - loss 0.02345417 - time (sec): 1.74 - samples/sec: 3534.13 - lr: 0.000013 - momentum: 0.000000
2023-10-23 18:20:40,466 epoch 7 - iter 45/152 - loss 0.01960782 - time (sec): 2.56 - samples/sec: 3630.78 - lr: 0.000012 - momentum: 0.000000
2023-10-23 18:20:41,328 epoch 7 - iter 60/152 - loss 0.01659578 - time (sec): 3.42 - samples/sec: 3677.30 - lr: 0.000012 - momentum: 0.000000
2023-10-23 18:20:42,192 epoch 7 - iter 75/152 - loss 0.01802838 - time (sec): 4.29 - samples/sec: 3632.53 - lr: 0.000012 - momentum: 0.000000
2023-10-23 18:20:43,067 epoch 7 - iter 90/152 - loss 0.01609284 - time (sec): 5.16 - samples/sec: 3606.92 - lr: 0.000011 - momentum: 0.000000
2023-10-23 18:20:44,104 epoch 7 - iter 105/152 - loss 0.01693181 - time (sec): 6.20 - samples/sec: 3481.38 - lr: 0.000011 - momentum: 0.000000
2023-10-23 18:20:44,980 epoch 7 - iter 120/152 - loss 0.02051780 - time (sec): 7.07 - samples/sec: 3447.07 - lr: 0.000011 - momentum: 0.000000
2023-10-23 18:20:45,850 epoch 7 - iter 135/152 - loss 0.02240269 - time (sec): 7.94 - samples/sec: 3500.08 - lr: 0.000010 - momentum: 0.000000
2023-10-23 18:20:46,714 epoch 7 - iter 150/152 - loss 0.02191603 - time (sec): 8.81 - samples/sec: 3484.81 - lr: 0.000010 - momentum: 0.000000
2023-10-23 18:20:46,830 ----------------------------------------------------------------------------------------------------
2023-10-23 18:20:46,830 EPOCH 7 done: loss 0.0217 - lr: 0.000010
2023-10-23 18:20:47,711 DEV : loss 0.20538194477558136 - f1-score (micro avg) 0.8298
2023-10-23 18:20:47,719 saving best model
2023-10-23 18:20:48,300 ----------------------------------------------------------------------------------------------------
2023-10-23 18:20:49,154 epoch 8 - iter 15/152 - loss 0.03230665 - time (sec): 0.85 - samples/sec: 3154.08 - lr: 0.000010 - momentum: 0.000000
2023-10-23 18:20:50,029 epoch 8 - iter 30/152 - loss 0.01668591 - time (sec): 1.73 - samples/sec: 3357.08 - lr: 0.000009 - momentum: 0.000000
2023-10-23 18:20:50,884 epoch 8 - iter 45/152 - loss 0.02251894 - time (sec): 2.58 - samples/sec: 3437.65 - lr: 0.000009 - momentum: 0.000000
2023-10-23 18:20:51,762 epoch 8 - iter 60/152 - loss 0.02137099 - time (sec): 3.46 - samples/sec: 3363.13 - lr: 0.000009 - momentum: 0.000000
2023-10-23 18:20:52,597 epoch 8 - iter 75/152 - loss 0.02250062 - time (sec): 4.30 - samples/sec: 3544.67 - lr: 0.000008 - momentum: 0.000000
2023-10-23 18:20:53,453 epoch 8 - iter 90/152 - loss 0.02719262 - time (sec): 5.15 - samples/sec: 3585.35 - lr: 0.000008 - momentum: 0.000000
2023-10-23 18:20:54,321 epoch 8 - iter 105/152 - loss 0.02531438 - time (sec): 6.02 - samples/sec: 3569.07 - lr: 0.000008 - momentum: 0.000000
2023-10-23 18:20:55,174 epoch 8 - iter 120/152 - loss 0.02258369 - time (sec): 6.87 - samples/sec: 3641.91 - lr: 0.000007 - momentum: 0.000000
2023-10-23 18:20:56,042 epoch 8 - iter 135/152 - loss 0.02136133 - time (sec): 7.74 - samples/sec: 3617.28 - lr: 0.000007 - momentum: 0.000000
2023-10-23 18:20:56,897 epoch 8 - iter 150/152 - loss 0.01996947 - time (sec): 8.60 - samples/sec: 3568.50 - lr: 0.000007 - momentum: 0.000000
2023-10-23 18:20:57,011 ----------------------------------------------------------------------------------------------------
2023-10-23 18:20:57,012 EPOCH 8 done: loss 0.0199 - lr: 0.000007
2023-10-23 18:20:57,869 DEV : loss 0.21447792649269104 - f1-score (micro avg) 0.8333
2023-10-23 18:20:57,877 saving best model
2023-10-23 18:20:58,441 ----------------------------------------------------------------------------------------------------
2023-10-23 18:20:59,283 epoch 9 - iter 15/152 - loss 0.01725436 - time (sec): 0.84 - samples/sec: 3938.89 - lr: 0.000006 - momentum: 0.000000
2023-10-23 18:21:00,148 epoch 9 - iter 30/152 - loss 0.01283757 - time (sec): 1.70 - samples/sec: 3671.09 - lr: 0.000006 - momentum: 0.000000
2023-10-23 18:21:00,986 epoch 9 - iter 45/152 - loss 0.01772803 - time (sec): 2.54 - samples/sec: 3708.49 - lr: 0.000006 - momentum: 0.000000
2023-10-23 18:21:01,843 epoch 9 - iter 60/152 - loss 0.01848425 - time (sec): 3.40 - samples/sec: 3532.67 - lr: 0.000005 - momentum: 0.000000
2023-10-23 18:21:02,696 epoch 9 - iter 75/152 - loss 0.01580778 - time (sec): 4.25 - samples/sec: 3604.89 - lr: 0.000005 - momentum: 0.000000
2023-10-23 18:21:03,553 epoch 9 - iter 90/152 - loss 0.01408170 - time (sec): 5.11 - samples/sec: 3579.45 - lr: 0.000005 - momentum: 0.000000
2023-10-23 18:21:04,389 epoch 9 - iter 105/152 - loss 0.01445924 - time (sec): 5.95 - samples/sec: 3594.29 - lr: 0.000004 - momentum: 0.000000
2023-10-23 18:21:05,258 epoch 9 - iter 120/152 - loss 0.01571336 - time (sec): 6.82 - samples/sec: 3577.20 - lr: 0.000004 - momentum: 0.000000
2023-10-23 18:21:06,133 epoch 9 - iter 135/152 - loss 0.01439284 - time (sec): 7.69 - samples/sec: 3576.32 - lr: 0.000004 - momentum: 0.000000
2023-10-23 18:21:07,008 epoch 9 - iter 150/152 - loss 0.01446179 - time (sec): 8.57 - samples/sec: 3577.22 - lr: 0.000004 - momentum: 0.000000
2023-10-23 18:21:07,117 ----------------------------------------------------------------------------------------------------
2023-10-23 18:21:07,117 EPOCH 9 done: loss 0.0143 - lr: 0.000004
2023-10-23 18:21:08,043 DEV : loss 0.20278561115264893 - f1-score (micro avg) 0.844
2023-10-23 18:21:08,053 saving best model
2023-10-23 18:21:08,617 ----------------------------------------------------------------------------------------------------
2023-10-23 18:21:09,456 epoch 10 - iter 15/152 - loss 0.01131616 - time (sec): 0.84 - samples/sec: 3627.17 - lr: 0.000003 - momentum: 0.000000
2023-10-23 18:21:10,257 epoch 10 - iter 30/152 - loss 0.01668086 - time (sec): 1.64 - samples/sec: 3740.30 - lr: 0.000003 - momentum: 0.000000
2023-10-23 18:21:11,029 epoch 10 - iter 45/152 - loss 0.01181820 - time (sec): 2.41 - samples/sec: 3876.25 - lr: 0.000002 - momentum: 0.000000
2023-10-23 18:21:11,812 epoch 10 - iter 60/152 - loss 0.01505051 - time (sec): 3.19 - samples/sec: 3823.97 - lr: 0.000002 - momentum: 0.000000
2023-10-23 18:21:12,607 epoch 10 - iter 75/152 - loss 0.01291314 - time (sec): 3.99 - samples/sec: 3857.98 - lr: 0.000002 - momentum: 0.000000
2023-10-23 18:21:13,388 epoch 10 - iter 90/152 - loss 0.01284814 - time (sec): 4.77 - samples/sec: 3864.82 - lr: 0.000002 - momentum: 0.000000
2023-10-23 18:21:14,161 epoch 10 - iter 105/152 - loss 0.01282451 - time (sec): 5.54 - samples/sec: 3862.61 - lr: 0.000001 - momentum: 0.000000
2023-10-23 18:21:14,948 epoch 10 - iter 120/152 - loss 0.01439910 - time (sec): 6.33 - samples/sec: 3912.19 - lr: 0.000001 - momentum: 0.000000
2023-10-23 18:21:15,731 epoch 10 - iter 135/152 - loss 0.01333481 - time (sec): 7.11 - samples/sec: 3882.82 - lr: 0.000001 - momentum: 0.000000
2023-10-23 18:21:16,498 epoch 10 - iter 150/152 - loss 0.01248508 - time (sec): 7.88 - samples/sec: 3884.03 - lr: 0.000000 - momentum: 0.000000
2023-10-23 18:21:16,596 ----------------------------------------------------------------------------------------------------
2023-10-23 18:21:16,596 EPOCH 10 done: loss 0.0123 - lr: 0.000000
2023-10-23 18:21:17,550 DEV : loss 0.20392796397209167 - f1-score (micro avg) 0.8443
2023-10-23 18:21:17,559 saving best model
2023-10-23 18:21:18,564 ----------------------------------------------------------------------------------------------------
2023-10-23 18:21:18,566 Loading model from best epoch ...
2023-10-23 18:21:20,461 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
2023-10-23 18:21:21,342
Results:
- F-score (micro) 0.8076
- F-score (macro) 0.6245
- Accuracy 0.6882
By class:
precision recall f1-score support
scope 0.7669 0.8278 0.7962 151
pers 0.7807 0.9271 0.8476 96
work 0.7664 0.8632 0.8119 95
loc 0.6667 0.6667 0.6667 3
date 0.0000 0.0000 0.0000 3
micro avg 0.7641 0.8563 0.8076 348
macro avg 0.5961 0.6569 0.6245 348
weighted avg 0.7631 0.8563 0.8067 348
2023-10-23 18:21:21,342 ----------------------------------------------------------------------------------------------------
|