File size: 24,101 Bytes
b1f715a |
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 |
2023-10-18 17:51:47,354 ----------------------------------------------------------------------------------------------------
2023-10-18 17:51:47,354 Model: "SequenceTagger(
(embeddings): TransformerWordEmbeddings(
(model): BertModel(
(embeddings): BertEmbeddings(
(word_embeddings): Embedding(32001, 128)
(position_embeddings): Embedding(512, 128)
(token_type_embeddings): Embedding(2, 128)
(LayerNorm): LayerNorm((128,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(encoder): BertEncoder(
(layer): ModuleList(
(0-1): 2 x BertLayer(
(attention): BertAttention(
(self): BertSelfAttention(
(query): Linear(in_features=128, out_features=128, bias=True)
(key): Linear(in_features=128, out_features=128, bias=True)
(value): Linear(in_features=128, out_features=128, bias=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(output): BertSelfOutput(
(dense): Linear(in_features=128, out_features=128, bias=True)
(LayerNorm): LayerNorm((128,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
(intermediate): BertIntermediate(
(dense): Linear(in_features=128, out_features=512, bias=True)
(intermediate_act_fn): GELUActivation()
)
(output): BertOutput(
(dense): Linear(in_features=512, out_features=128, bias=True)
(LayerNorm): LayerNorm((128,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
)
(pooler): BertPooler(
(dense): Linear(in_features=128, out_features=128, bias=True)
(activation): Tanh()
)
)
)
(locked_dropout): LockedDropout(p=0.5)
(linear): Linear(in_features=128, out_features=21, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-18 17:51:47,354 ----------------------------------------------------------------------------------------------------
2023-10-18 17:51:47,354 MultiCorpus: 3575 train + 1235 dev + 1266 test sentences
- NER_HIPE_2022 Corpus: 3575 train + 1235 dev + 1266 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/hipe2020/de/with_doc_seperator
2023-10-18 17:51:47,354 ----------------------------------------------------------------------------------------------------
2023-10-18 17:51:47,354 Train: 3575 sentences
2023-10-18 17:51:47,354 (train_with_dev=False, train_with_test=False)
2023-10-18 17:51:47,354 ----------------------------------------------------------------------------------------------------
2023-10-18 17:51:47,354 Training Params:
2023-10-18 17:51:47,355 - learning_rate: "5e-05"
2023-10-18 17:51:47,355 - mini_batch_size: "4"
2023-10-18 17:51:47,355 - max_epochs: "10"
2023-10-18 17:51:47,355 - shuffle: "True"
2023-10-18 17:51:47,355 ----------------------------------------------------------------------------------------------------
2023-10-18 17:51:47,355 Plugins:
2023-10-18 17:51:47,355 - TensorboardLogger
2023-10-18 17:51:47,355 - LinearScheduler | warmup_fraction: '0.1'
2023-10-18 17:51:47,355 ----------------------------------------------------------------------------------------------------
2023-10-18 17:51:47,355 Final evaluation on model from best epoch (best-model.pt)
2023-10-18 17:51:47,355 - metric: "('micro avg', 'f1-score')"
2023-10-18 17:51:47,355 ----------------------------------------------------------------------------------------------------
2023-10-18 17:51:47,355 Computation:
2023-10-18 17:51:47,355 - compute on device: cuda:0
2023-10-18 17:51:47,355 - embedding storage: none
2023-10-18 17:51:47,355 ----------------------------------------------------------------------------------------------------
2023-10-18 17:51:47,355 Model training base path: "hmbench-hipe2020/de-dbmdz/bert-tiny-historic-multilingual-cased-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-2"
2023-10-18 17:51:47,355 ----------------------------------------------------------------------------------------------------
2023-10-18 17:51:47,355 ----------------------------------------------------------------------------------------------------
2023-10-18 17:51:47,355 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-18 17:51:48,543 epoch 1 - iter 89/894 - loss 3.15511888 - time (sec): 1.19 - samples/sec: 7651.58 - lr: 0.000005 - momentum: 0.000000
2023-10-18 17:51:49,821 epoch 1 - iter 178/894 - loss 2.80037201 - time (sec): 2.47 - samples/sec: 7645.46 - lr: 0.000010 - momentum: 0.000000
2023-10-18 17:51:51,177 epoch 1 - iter 267/894 - loss 2.47260416 - time (sec): 3.82 - samples/sec: 6949.16 - lr: 0.000015 - momentum: 0.000000
2023-10-18 17:51:52,559 epoch 1 - iter 356/894 - loss 2.10896505 - time (sec): 5.20 - samples/sec: 6568.91 - lr: 0.000020 - momentum: 0.000000
2023-10-18 17:51:53,949 epoch 1 - iter 445/894 - loss 1.83239594 - time (sec): 6.59 - samples/sec: 6413.69 - lr: 0.000025 - momentum: 0.000000
2023-10-18 17:51:55,393 epoch 1 - iter 534/894 - loss 1.63156121 - time (sec): 8.04 - samples/sec: 6332.03 - lr: 0.000030 - momentum: 0.000000
2023-10-18 17:51:56,861 epoch 1 - iter 623/894 - loss 1.45542562 - time (sec): 9.51 - samples/sec: 6423.68 - lr: 0.000035 - momentum: 0.000000
2023-10-18 17:51:58,227 epoch 1 - iter 712/894 - loss 1.33899230 - time (sec): 10.87 - samples/sec: 6400.13 - lr: 0.000040 - momentum: 0.000000
2023-10-18 17:51:59,651 epoch 1 - iter 801/894 - loss 1.25186435 - time (sec): 12.30 - samples/sec: 6339.08 - lr: 0.000045 - momentum: 0.000000
2023-10-18 17:52:01,030 epoch 1 - iter 890/894 - loss 1.18860703 - time (sec): 13.67 - samples/sec: 6310.36 - lr: 0.000050 - momentum: 0.000000
2023-10-18 17:52:01,090 ----------------------------------------------------------------------------------------------------
2023-10-18 17:52:01,090 EPOCH 1 done: loss 1.1875 - lr: 0.000050
2023-10-18 17:52:03,341 DEV : loss 0.3861956000328064 - f1-score (micro avg) 0.0
2023-10-18 17:52:03,366 ----------------------------------------------------------------------------------------------------
2023-10-18 17:52:04,722 epoch 2 - iter 89/894 - loss 0.46859686 - time (sec): 1.36 - samples/sec: 6233.38 - lr: 0.000049 - momentum: 0.000000
2023-10-18 17:52:06,106 epoch 2 - iter 178/894 - loss 0.48826135 - time (sec): 2.74 - samples/sec: 6366.22 - lr: 0.000049 - momentum: 0.000000
2023-10-18 17:52:07,448 epoch 2 - iter 267/894 - loss 0.47572411 - time (sec): 4.08 - samples/sec: 6180.80 - lr: 0.000048 - momentum: 0.000000
2023-10-18 17:52:08,844 epoch 2 - iter 356/894 - loss 0.47148961 - time (sec): 5.48 - samples/sec: 6084.52 - lr: 0.000048 - momentum: 0.000000
2023-10-18 17:52:10,261 epoch 2 - iter 445/894 - loss 0.46961239 - time (sec): 6.89 - samples/sec: 6246.17 - lr: 0.000047 - momentum: 0.000000
2023-10-18 17:52:11,623 epoch 2 - iter 534/894 - loss 0.46231896 - time (sec): 8.26 - samples/sec: 6237.81 - lr: 0.000047 - momentum: 0.000000
2023-10-18 17:52:13,056 epoch 2 - iter 623/894 - loss 0.46322114 - time (sec): 9.69 - samples/sec: 6365.55 - lr: 0.000046 - momentum: 0.000000
2023-10-18 17:52:14,389 epoch 2 - iter 712/894 - loss 0.45970767 - time (sec): 11.02 - samples/sec: 6278.75 - lr: 0.000046 - momentum: 0.000000
2023-10-18 17:52:15,768 epoch 2 - iter 801/894 - loss 0.45510621 - time (sec): 12.40 - samples/sec: 6268.29 - lr: 0.000045 - momentum: 0.000000
2023-10-18 17:52:17,137 epoch 2 - iter 890/894 - loss 0.45141144 - time (sec): 13.77 - samples/sec: 6265.98 - lr: 0.000044 - momentum: 0.000000
2023-10-18 17:52:17,191 ----------------------------------------------------------------------------------------------------
2023-10-18 17:52:17,191 EPOCH 2 done: loss 0.4526 - lr: 0.000044
2023-10-18 17:52:22,463 DEV : loss 0.3269258439540863 - f1-score (micro avg) 0.2848
2023-10-18 17:52:22,490 saving best model
2023-10-18 17:52:22,526 ----------------------------------------------------------------------------------------------------
2023-10-18 17:52:23,961 epoch 3 - iter 89/894 - loss 0.42146752 - time (sec): 1.43 - samples/sec: 6369.86 - lr: 0.000044 - momentum: 0.000000
2023-10-18 17:52:25,413 epoch 3 - iter 178/894 - loss 0.41385378 - time (sec): 2.89 - samples/sec: 6155.04 - lr: 0.000043 - momentum: 0.000000
2023-10-18 17:52:26,812 epoch 3 - iter 267/894 - loss 0.39703685 - time (sec): 4.29 - samples/sec: 6193.72 - lr: 0.000043 - momentum: 0.000000
2023-10-18 17:52:28,162 epoch 3 - iter 356/894 - loss 0.40727453 - time (sec): 5.64 - samples/sec: 6079.29 - lr: 0.000042 - momentum: 0.000000
2023-10-18 17:52:29,517 epoch 3 - iter 445/894 - loss 0.39317668 - time (sec): 6.99 - samples/sec: 6060.54 - lr: 0.000042 - momentum: 0.000000
2023-10-18 17:52:30,907 epoch 3 - iter 534/894 - loss 0.39255571 - time (sec): 8.38 - samples/sec: 6086.50 - lr: 0.000041 - momentum: 0.000000
2023-10-18 17:52:32,257 epoch 3 - iter 623/894 - loss 0.38537576 - time (sec): 9.73 - samples/sec: 6120.95 - lr: 0.000041 - momentum: 0.000000
2023-10-18 17:52:33,603 epoch 3 - iter 712/894 - loss 0.38546544 - time (sec): 11.08 - samples/sec: 6197.79 - lr: 0.000040 - momentum: 0.000000
2023-10-18 17:52:35,022 epoch 3 - iter 801/894 - loss 0.37977830 - time (sec): 12.50 - samples/sec: 6220.92 - lr: 0.000039 - momentum: 0.000000
2023-10-18 17:52:36,393 epoch 3 - iter 890/894 - loss 0.37792977 - time (sec): 13.87 - samples/sec: 6217.99 - lr: 0.000039 - momentum: 0.000000
2023-10-18 17:52:36,454 ----------------------------------------------------------------------------------------------------
2023-10-18 17:52:36,454 EPOCH 3 done: loss 0.3781 - lr: 0.000039
2023-10-18 17:52:41,761 DEV : loss 0.33094078302383423 - f1-score (micro avg) 0.3422
2023-10-18 17:52:41,787 saving best model
2023-10-18 17:52:41,826 ----------------------------------------------------------------------------------------------------
2023-10-18 17:52:43,299 epoch 4 - iter 89/894 - loss 0.36629004 - time (sec): 1.47 - samples/sec: 5534.51 - lr: 0.000038 - momentum: 0.000000
2023-10-18 17:52:44,808 epoch 4 - iter 178/894 - loss 0.33855096 - time (sec): 2.98 - samples/sec: 6134.47 - lr: 0.000038 - momentum: 0.000000
2023-10-18 17:52:46,217 epoch 4 - iter 267/894 - loss 0.33721312 - time (sec): 4.39 - samples/sec: 6122.15 - lr: 0.000037 - momentum: 0.000000
2023-10-18 17:52:47,629 epoch 4 - iter 356/894 - loss 0.34572340 - time (sec): 5.80 - samples/sec: 6140.01 - lr: 0.000037 - momentum: 0.000000
2023-10-18 17:52:49,067 epoch 4 - iter 445/894 - loss 0.33658431 - time (sec): 7.24 - samples/sec: 6141.39 - lr: 0.000036 - momentum: 0.000000
2023-10-18 17:52:50,434 epoch 4 - iter 534/894 - loss 0.33650082 - time (sec): 8.61 - samples/sec: 6133.57 - lr: 0.000036 - momentum: 0.000000
2023-10-18 17:52:51,821 epoch 4 - iter 623/894 - loss 0.33284655 - time (sec): 9.99 - samples/sec: 6131.40 - lr: 0.000035 - momentum: 0.000000
2023-10-18 17:52:53,229 epoch 4 - iter 712/894 - loss 0.33587300 - time (sec): 11.40 - samples/sec: 6117.26 - lr: 0.000034 - momentum: 0.000000
2023-10-18 17:52:54,614 epoch 4 - iter 801/894 - loss 0.33537409 - time (sec): 12.79 - samples/sec: 6083.85 - lr: 0.000034 - momentum: 0.000000
2023-10-18 17:52:56,013 epoch 4 - iter 890/894 - loss 0.33482722 - time (sec): 14.19 - samples/sec: 6073.37 - lr: 0.000033 - momentum: 0.000000
2023-10-18 17:52:56,079 ----------------------------------------------------------------------------------------------------
2023-10-18 17:52:56,079 EPOCH 4 done: loss 0.3335 - lr: 0.000033
2023-10-18 17:53:01,130 DEV : loss 0.2942203879356384 - f1-score (micro avg) 0.3633
2023-10-18 17:53:01,157 saving best model
2023-10-18 17:53:01,197 ----------------------------------------------------------------------------------------------------
2023-10-18 17:53:02,724 epoch 5 - iter 89/894 - loss 0.32299854 - time (sec): 1.53 - samples/sec: 5726.16 - lr: 0.000033 - momentum: 0.000000
2023-10-18 17:53:04,138 epoch 5 - iter 178/894 - loss 0.29471132 - time (sec): 2.94 - samples/sec: 6180.01 - lr: 0.000032 - momentum: 0.000000
2023-10-18 17:53:05,500 epoch 5 - iter 267/894 - loss 0.30092530 - time (sec): 4.30 - samples/sec: 6008.61 - lr: 0.000032 - momentum: 0.000000
2023-10-18 17:53:06,897 epoch 5 - iter 356/894 - loss 0.29863113 - time (sec): 5.70 - samples/sec: 6060.52 - lr: 0.000031 - momentum: 0.000000
2023-10-18 17:53:08,305 epoch 5 - iter 445/894 - loss 0.30227620 - time (sec): 7.11 - samples/sec: 5986.26 - lr: 0.000031 - momentum: 0.000000
2023-10-18 17:53:09,673 epoch 5 - iter 534/894 - loss 0.30941454 - time (sec): 8.48 - samples/sec: 5982.95 - lr: 0.000030 - momentum: 0.000000
2023-10-18 17:53:11,395 epoch 5 - iter 623/894 - loss 0.30942619 - time (sec): 10.20 - samples/sec: 5877.15 - lr: 0.000029 - momentum: 0.000000
2023-10-18 17:53:12,804 epoch 5 - iter 712/894 - loss 0.31087058 - time (sec): 11.61 - samples/sec: 5970.10 - lr: 0.000029 - momentum: 0.000000
2023-10-18 17:53:14,197 epoch 5 - iter 801/894 - loss 0.30816482 - time (sec): 13.00 - samples/sec: 5989.37 - lr: 0.000028 - momentum: 0.000000
2023-10-18 17:53:15,554 epoch 5 - iter 890/894 - loss 0.30195323 - time (sec): 14.36 - samples/sec: 6006.69 - lr: 0.000028 - momentum: 0.000000
2023-10-18 17:53:15,615 ----------------------------------------------------------------------------------------------------
2023-10-18 17:53:15,615 EPOCH 5 done: loss 0.3014 - lr: 0.000028
2023-10-18 17:53:20,620 DEV : loss 0.30749502778053284 - f1-score (micro avg) 0.3628
2023-10-18 17:53:20,646 ----------------------------------------------------------------------------------------------------
2023-10-18 17:53:22,067 epoch 6 - iter 89/894 - loss 0.28575379 - time (sec): 1.42 - samples/sec: 6046.45 - lr: 0.000027 - momentum: 0.000000
2023-10-18 17:53:23,422 epoch 6 - iter 178/894 - loss 0.27960058 - time (sec): 2.78 - samples/sec: 5968.25 - lr: 0.000027 - momentum: 0.000000
2023-10-18 17:53:24,818 epoch 6 - iter 267/894 - loss 0.26798805 - time (sec): 4.17 - samples/sec: 5811.20 - lr: 0.000026 - momentum: 0.000000
2023-10-18 17:53:26,205 epoch 6 - iter 356/894 - loss 0.28685018 - time (sec): 5.56 - samples/sec: 5849.95 - lr: 0.000026 - momentum: 0.000000
2023-10-18 17:53:27,592 epoch 6 - iter 445/894 - loss 0.28493924 - time (sec): 6.95 - samples/sec: 5887.50 - lr: 0.000025 - momentum: 0.000000
2023-10-18 17:53:29,034 epoch 6 - iter 534/894 - loss 0.29289459 - time (sec): 8.39 - samples/sec: 6068.56 - lr: 0.000024 - momentum: 0.000000
2023-10-18 17:53:30,425 epoch 6 - iter 623/894 - loss 0.28609675 - time (sec): 9.78 - samples/sec: 6099.55 - lr: 0.000024 - momentum: 0.000000
2023-10-18 17:53:31,825 epoch 6 - iter 712/894 - loss 0.27834340 - time (sec): 11.18 - samples/sec: 6125.73 - lr: 0.000023 - momentum: 0.000000
2023-10-18 17:53:33,251 epoch 6 - iter 801/894 - loss 0.27981318 - time (sec): 12.60 - samples/sec: 6170.26 - lr: 0.000023 - momentum: 0.000000
2023-10-18 17:53:34,537 epoch 6 - iter 890/894 - loss 0.27587057 - time (sec): 13.89 - samples/sec: 6206.85 - lr: 0.000022 - momentum: 0.000000
2023-10-18 17:53:34,592 ----------------------------------------------------------------------------------------------------
2023-10-18 17:53:34,592 EPOCH 6 done: loss 0.2759 - lr: 0.000022
2023-10-18 17:53:39,941 DEV : loss 0.2958523631095886 - f1-score (micro avg) 0.3801
2023-10-18 17:53:39,967 saving best model
2023-10-18 17:53:40,004 ----------------------------------------------------------------------------------------------------
2023-10-18 17:53:41,290 epoch 7 - iter 89/894 - loss 0.25107818 - time (sec): 1.29 - samples/sec: 6966.43 - lr: 0.000022 - momentum: 0.000000
2023-10-18 17:53:42,673 epoch 7 - iter 178/894 - loss 0.26751214 - time (sec): 2.67 - samples/sec: 6486.38 - lr: 0.000021 - momentum: 0.000000
2023-10-18 17:53:44,052 epoch 7 - iter 267/894 - loss 0.26367724 - time (sec): 4.05 - samples/sec: 6364.62 - lr: 0.000021 - momentum: 0.000000
2023-10-18 17:53:45,431 epoch 7 - iter 356/894 - loss 0.26557215 - time (sec): 5.43 - samples/sec: 6229.57 - lr: 0.000020 - momentum: 0.000000
2023-10-18 17:53:46,936 epoch 7 - iter 445/894 - loss 0.25680326 - time (sec): 6.93 - samples/sec: 6101.11 - lr: 0.000019 - momentum: 0.000000
2023-10-18 17:53:48,357 epoch 7 - iter 534/894 - loss 0.25808523 - time (sec): 8.35 - samples/sec: 6174.89 - lr: 0.000019 - momentum: 0.000000
2023-10-18 17:53:49,803 epoch 7 - iter 623/894 - loss 0.26017038 - time (sec): 9.80 - samples/sec: 6166.19 - lr: 0.000018 - momentum: 0.000000
2023-10-18 17:53:51,177 epoch 7 - iter 712/894 - loss 0.26202742 - time (sec): 11.17 - samples/sec: 6238.15 - lr: 0.000018 - momentum: 0.000000
2023-10-18 17:53:52,566 epoch 7 - iter 801/894 - loss 0.26124153 - time (sec): 12.56 - samples/sec: 6242.21 - lr: 0.000017 - momentum: 0.000000
2023-10-18 17:53:53,941 epoch 7 - iter 890/894 - loss 0.25949025 - time (sec): 13.94 - samples/sec: 6186.49 - lr: 0.000017 - momentum: 0.000000
2023-10-18 17:53:54,000 ----------------------------------------------------------------------------------------------------
2023-10-18 17:53:54,000 EPOCH 7 done: loss 0.2602 - lr: 0.000017
2023-10-18 17:53:59,392 DEV : loss 0.29916736483573914 - f1-score (micro avg) 0.3861
2023-10-18 17:53:59,420 saving best model
2023-10-18 17:53:59,462 ----------------------------------------------------------------------------------------------------
2023-10-18 17:54:00,850 epoch 8 - iter 89/894 - loss 0.28048937 - time (sec): 1.39 - samples/sec: 6383.64 - lr: 0.000016 - momentum: 0.000000
2023-10-18 17:54:02,245 epoch 8 - iter 178/894 - loss 0.26462728 - time (sec): 2.78 - samples/sec: 6586.99 - lr: 0.000016 - momentum: 0.000000
2023-10-18 17:54:03,619 epoch 8 - iter 267/894 - loss 0.26224666 - time (sec): 4.16 - samples/sec: 6344.89 - lr: 0.000015 - momentum: 0.000000
2023-10-18 17:54:05,047 epoch 8 - iter 356/894 - loss 0.25592339 - time (sec): 5.58 - samples/sec: 6308.85 - lr: 0.000014 - momentum: 0.000000
2023-10-18 17:54:06,456 epoch 8 - iter 445/894 - loss 0.24937114 - time (sec): 6.99 - samples/sec: 6381.76 - lr: 0.000014 - momentum: 0.000000
2023-10-18 17:54:07,811 epoch 8 - iter 534/894 - loss 0.24766456 - time (sec): 8.35 - samples/sec: 6353.10 - lr: 0.000013 - momentum: 0.000000
2023-10-18 17:54:09,183 epoch 8 - iter 623/894 - loss 0.25014946 - time (sec): 9.72 - samples/sec: 6266.67 - lr: 0.000013 - momentum: 0.000000
2023-10-18 17:54:10,616 epoch 8 - iter 712/894 - loss 0.24560857 - time (sec): 11.15 - samples/sec: 6296.58 - lr: 0.000012 - momentum: 0.000000
2023-10-18 17:54:12,031 epoch 8 - iter 801/894 - loss 0.25300894 - time (sec): 12.57 - samples/sec: 6242.10 - lr: 0.000012 - momentum: 0.000000
2023-10-18 17:54:13,396 epoch 8 - iter 890/894 - loss 0.24861475 - time (sec): 13.93 - samples/sec: 6181.84 - lr: 0.000011 - momentum: 0.000000
2023-10-18 17:54:13,454 ----------------------------------------------------------------------------------------------------
2023-10-18 17:54:13,454 EPOCH 8 done: loss 0.2481 - lr: 0.000011
2023-10-18 17:54:18,827 DEV : loss 0.3004680871963501 - f1-score (micro avg) 0.398
2023-10-18 17:54:18,853 saving best model
2023-10-18 17:54:18,893 ----------------------------------------------------------------------------------------------------
2023-10-18 17:54:20,299 epoch 9 - iter 89/894 - loss 0.18561501 - time (sec): 1.41 - samples/sec: 6001.06 - lr: 0.000011 - momentum: 0.000000
2023-10-18 17:54:21,683 epoch 9 - iter 178/894 - loss 0.21725744 - time (sec): 2.79 - samples/sec: 5961.01 - lr: 0.000010 - momentum: 0.000000
2023-10-18 17:54:23,101 epoch 9 - iter 267/894 - loss 0.22441472 - time (sec): 4.21 - samples/sec: 6224.00 - lr: 0.000009 - momentum: 0.000000
2023-10-18 17:54:24,487 epoch 9 - iter 356/894 - loss 0.23638987 - time (sec): 5.59 - samples/sec: 6324.92 - lr: 0.000009 - momentum: 0.000000
2023-10-18 17:54:25,859 epoch 9 - iter 445/894 - loss 0.23826232 - time (sec): 6.97 - samples/sec: 6205.80 - lr: 0.000008 - momentum: 0.000000
2023-10-18 17:54:27,231 epoch 9 - iter 534/894 - loss 0.23872170 - time (sec): 8.34 - samples/sec: 6188.22 - lr: 0.000008 - momentum: 0.000000
2023-10-18 17:54:28,611 epoch 9 - iter 623/894 - loss 0.23973153 - time (sec): 9.72 - samples/sec: 6203.64 - lr: 0.000007 - momentum: 0.000000
2023-10-18 17:54:29,994 epoch 9 - iter 712/894 - loss 0.23340771 - time (sec): 11.10 - samples/sec: 6298.23 - lr: 0.000007 - momentum: 0.000000
2023-10-18 17:54:31,367 epoch 9 - iter 801/894 - loss 0.23972144 - time (sec): 12.47 - samples/sec: 6249.30 - lr: 0.000006 - momentum: 0.000000
2023-10-18 17:54:32,776 epoch 9 - iter 890/894 - loss 0.24028937 - time (sec): 13.88 - samples/sec: 6208.28 - lr: 0.000006 - momentum: 0.000000
2023-10-18 17:54:32,837 ----------------------------------------------------------------------------------------------------
2023-10-18 17:54:32,837 EPOCH 9 done: loss 0.2396 - lr: 0.000006
2023-10-18 17:54:37,940 DEV : loss 0.29927825927734375 - f1-score (micro avg) 0.3978
2023-10-18 17:54:37,969 ----------------------------------------------------------------------------------------------------
2023-10-18 17:54:39,383 epoch 10 - iter 89/894 - loss 0.23021562 - time (sec): 1.41 - samples/sec: 6946.18 - lr: 0.000005 - momentum: 0.000000
2023-10-18 17:54:40,731 epoch 10 - iter 178/894 - loss 0.23527595 - time (sec): 2.76 - samples/sec: 6556.86 - lr: 0.000004 - momentum: 0.000000
2023-10-18 17:54:42,095 epoch 10 - iter 267/894 - loss 0.22389138 - time (sec): 4.13 - samples/sec: 6417.79 - lr: 0.000004 - momentum: 0.000000
2023-10-18 17:54:43,542 epoch 10 - iter 356/894 - loss 0.23573274 - time (sec): 5.57 - samples/sec: 6372.62 - lr: 0.000003 - momentum: 0.000000
2023-10-18 17:54:44,896 epoch 10 - iter 445/894 - loss 0.23234303 - time (sec): 6.93 - samples/sec: 6222.73 - lr: 0.000003 - momentum: 0.000000
2023-10-18 17:54:46,283 epoch 10 - iter 534/894 - loss 0.23247333 - time (sec): 8.31 - samples/sec: 6296.22 - lr: 0.000002 - momentum: 0.000000
2023-10-18 17:54:47,678 epoch 10 - iter 623/894 - loss 0.24172654 - time (sec): 9.71 - samples/sec: 6374.96 - lr: 0.000002 - momentum: 0.000000
2023-10-18 17:54:49,076 epoch 10 - iter 712/894 - loss 0.24388043 - time (sec): 11.11 - samples/sec: 6272.26 - lr: 0.000001 - momentum: 0.000000
2023-10-18 17:54:50,370 epoch 10 - iter 801/894 - loss 0.23766827 - time (sec): 12.40 - samples/sec: 6290.49 - lr: 0.000001 - momentum: 0.000000
2023-10-18 17:54:51,731 epoch 10 - iter 890/894 - loss 0.23736563 - time (sec): 13.76 - samples/sec: 6236.95 - lr: 0.000000 - momentum: 0.000000
2023-10-18 17:54:51,808 ----------------------------------------------------------------------------------------------------
2023-10-18 17:54:51,809 EPOCH 10 done: loss 0.2369 - lr: 0.000000
2023-10-18 17:54:57,262 DEV : loss 0.30093932151794434 - f1-score (micro avg) 0.3971
2023-10-18 17:54:57,319 ----------------------------------------------------------------------------------------------------
2023-10-18 17:54:57,320 Loading model from best epoch ...
2023-10-18 17:54:57,400 SequenceTagger predicts: Dictionary with 21 tags: O, S-loc, B-loc, E-loc, I-loc, S-pers, B-pers, E-pers, I-pers, S-org, B-org, E-org, I-org, S-prod, B-prod, E-prod, I-prod, S-time, B-time, E-time, I-time
2023-10-18 17:54:59,821
Results:
- F-score (micro) 0.4054
- F-score (macro) 0.207
- Accuracy 0.2654
By class:
precision recall f1-score support
loc 0.5597 0.5822 0.5707 596
pers 0.2388 0.3213 0.2740 333
org 0.0000 0.0000 0.0000 132
time 0.2286 0.1633 0.1905 49
prod 0.0000 0.0000 0.0000 66
micro avg 0.4189 0.3929 0.4054 1176
macro avg 0.2054 0.2134 0.2070 1176
weighted avg 0.3608 0.3929 0.3748 1176
2023-10-18 17:54:59,821 ----------------------------------------------------------------------------------------------------
|