File size: 24,224 Bytes
a508f10 |
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 |
2023-10-20 09:06:35,409 ----------------------------------------------------------------------------------------------------
2023-10-20 09:06:35,409 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=13, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-20 09:06:35,410 ----------------------------------------------------------------------------------------------------
2023-10-20 09:06:35,410 MultiCorpus: 6183 train + 680 dev + 2113 test sentences
- NER_HIPE_2022 Corpus: 6183 train + 680 dev + 2113 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/topres19th/en/with_doc_seperator
2023-10-20 09:06:35,410 ----------------------------------------------------------------------------------------------------
2023-10-20 09:06:35,410 Train: 6183 sentences
2023-10-20 09:06:35,410 (train_with_dev=False, train_with_test=False)
2023-10-20 09:06:35,410 ----------------------------------------------------------------------------------------------------
2023-10-20 09:06:35,410 Training Params:
2023-10-20 09:06:35,410 - learning_rate: "3e-05"
2023-10-20 09:06:35,410 - mini_batch_size: "4"
2023-10-20 09:06:35,410 - max_epochs: "10"
2023-10-20 09:06:35,410 - shuffle: "True"
2023-10-20 09:06:35,410 ----------------------------------------------------------------------------------------------------
2023-10-20 09:06:35,411 Plugins:
2023-10-20 09:06:35,411 - TensorboardLogger
2023-10-20 09:06:35,411 - LinearScheduler | warmup_fraction: '0.1'
2023-10-20 09:06:35,411 ----------------------------------------------------------------------------------------------------
2023-10-20 09:06:35,411 Final evaluation on model from best epoch (best-model.pt)
2023-10-20 09:06:35,411 - metric: "('micro avg', 'f1-score')"
2023-10-20 09:06:35,411 ----------------------------------------------------------------------------------------------------
2023-10-20 09:06:35,411 Computation:
2023-10-20 09:06:35,411 - compute on device: cuda:0
2023-10-20 09:06:35,411 - embedding storage: none
2023-10-20 09:06:35,411 ----------------------------------------------------------------------------------------------------
2023-10-20 09:06:35,411 Model training base path: "hmbench-topres19th/en-dbmdz/bert-tiny-historic-multilingual-cased-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-1"
2023-10-20 09:06:35,411 ----------------------------------------------------------------------------------------------------
2023-10-20 09:06:35,411 ----------------------------------------------------------------------------------------------------
2023-10-20 09:06:35,411 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-20 09:06:37,833 epoch 1 - iter 154/1546 - loss 3.28397095 - time (sec): 2.42 - samples/sec: 5244.37 - lr: 0.000003 - momentum: 0.000000
2023-10-20 09:06:40,154 epoch 1 - iter 308/1546 - loss 2.99105285 - time (sec): 4.74 - samples/sec: 5139.46 - lr: 0.000006 - momentum: 0.000000
2023-10-20 09:06:42,531 epoch 1 - iter 462/1546 - loss 2.52293414 - time (sec): 7.12 - samples/sec: 5133.11 - lr: 0.000009 - momentum: 0.000000
2023-10-20 09:06:44,753 epoch 1 - iter 616/1546 - loss 2.02560749 - time (sec): 9.34 - samples/sec: 5282.45 - lr: 0.000012 - momentum: 0.000000
2023-10-20 09:06:46,974 epoch 1 - iter 770/1546 - loss 1.67950423 - time (sec): 11.56 - samples/sec: 5315.86 - lr: 0.000015 - momentum: 0.000000
2023-10-20 09:06:49,149 epoch 1 - iter 924/1546 - loss 1.45624081 - time (sec): 13.74 - samples/sec: 5319.42 - lr: 0.000018 - momentum: 0.000000
2023-10-20 09:06:51,518 epoch 1 - iter 1078/1546 - loss 1.28121677 - time (sec): 16.11 - samples/sec: 5324.77 - lr: 0.000021 - momentum: 0.000000
2023-10-20 09:06:53,979 epoch 1 - iter 1232/1546 - loss 1.14682597 - time (sec): 18.57 - samples/sec: 5319.74 - lr: 0.000024 - momentum: 0.000000
2023-10-20 09:06:56,359 epoch 1 - iter 1386/1546 - loss 1.05075802 - time (sec): 20.95 - samples/sec: 5283.56 - lr: 0.000027 - momentum: 0.000000
2023-10-20 09:06:58,781 epoch 1 - iter 1540/1546 - loss 0.96746300 - time (sec): 23.37 - samples/sec: 5294.97 - lr: 0.000030 - momentum: 0.000000
2023-10-20 09:06:58,888 ----------------------------------------------------------------------------------------------------
2023-10-20 09:06:58,888 EPOCH 1 done: loss 0.9635 - lr: 0.000030
2023-10-20 09:06:59,562 DEV : loss 0.1460493505001068 - f1-score (micro avg) 0.0
2023-10-20 09:06:59,573 ----------------------------------------------------------------------------------------------------
2023-10-20 09:07:01,954 epoch 2 - iter 154/1546 - loss 0.23058634 - time (sec): 2.38 - samples/sec: 5219.28 - lr: 0.000030 - momentum: 0.000000
2023-10-20 09:07:04,312 epoch 2 - iter 308/1546 - loss 0.22254356 - time (sec): 4.74 - samples/sec: 5059.52 - lr: 0.000029 - momentum: 0.000000
2023-10-20 09:07:06,666 epoch 2 - iter 462/1546 - loss 0.22366210 - time (sec): 7.09 - samples/sec: 4999.49 - lr: 0.000029 - momentum: 0.000000
2023-10-20 09:07:09,084 epoch 2 - iter 616/1546 - loss 0.21092520 - time (sec): 9.51 - samples/sec: 5084.24 - lr: 0.000029 - momentum: 0.000000
2023-10-20 09:07:11,446 epoch 2 - iter 770/1546 - loss 0.20806216 - time (sec): 11.87 - samples/sec: 5152.85 - lr: 0.000028 - momentum: 0.000000
2023-10-20 09:07:13,762 epoch 2 - iter 924/1546 - loss 0.20526902 - time (sec): 14.19 - samples/sec: 5138.44 - lr: 0.000028 - momentum: 0.000000
2023-10-20 09:07:16,170 epoch 2 - iter 1078/1546 - loss 0.20379508 - time (sec): 16.60 - samples/sec: 5111.99 - lr: 0.000028 - momentum: 0.000000
2023-10-20 09:07:18,549 epoch 2 - iter 1232/1546 - loss 0.20172155 - time (sec): 18.98 - samples/sec: 5138.82 - lr: 0.000027 - momentum: 0.000000
2023-10-20 09:07:20,996 epoch 2 - iter 1386/1546 - loss 0.20086049 - time (sec): 21.42 - samples/sec: 5126.79 - lr: 0.000027 - momentum: 0.000000
2023-10-20 09:07:23,418 epoch 2 - iter 1540/1546 - loss 0.19704649 - time (sec): 23.84 - samples/sec: 5183.51 - lr: 0.000027 - momentum: 0.000000
2023-10-20 09:07:23,513 ----------------------------------------------------------------------------------------------------
2023-10-20 09:07:23,513 EPOCH 2 done: loss 0.1964 - lr: 0.000027
2023-10-20 09:07:24,832 DEV : loss 0.09642348438501358 - f1-score (micro avg) 0.452
2023-10-20 09:07:24,844 saving best model
2023-10-20 09:07:24,878 ----------------------------------------------------------------------------------------------------
2023-10-20 09:07:27,163 epoch 3 - iter 154/1546 - loss 0.16756513 - time (sec): 2.28 - samples/sec: 5008.14 - lr: 0.000026 - momentum: 0.000000
2023-10-20 09:07:29,473 epoch 3 - iter 308/1546 - loss 0.15306420 - time (sec): 4.59 - samples/sec: 5243.02 - lr: 0.000026 - momentum: 0.000000
2023-10-20 09:07:31,759 epoch 3 - iter 462/1546 - loss 0.14828552 - time (sec): 6.88 - samples/sec: 5298.73 - lr: 0.000026 - momentum: 0.000000
2023-10-20 09:07:34,118 epoch 3 - iter 616/1546 - loss 0.15847543 - time (sec): 9.24 - samples/sec: 5349.69 - lr: 0.000025 - momentum: 0.000000
2023-10-20 09:07:36,476 epoch 3 - iter 770/1546 - loss 0.15836327 - time (sec): 11.60 - samples/sec: 5290.33 - lr: 0.000025 - momentum: 0.000000
2023-10-20 09:07:38,811 epoch 3 - iter 924/1546 - loss 0.15990269 - time (sec): 13.93 - samples/sec: 5374.68 - lr: 0.000025 - momentum: 0.000000
2023-10-20 09:07:41,185 epoch 3 - iter 1078/1546 - loss 0.16137110 - time (sec): 16.31 - samples/sec: 5359.58 - lr: 0.000024 - momentum: 0.000000
2023-10-20 09:07:43,553 epoch 3 - iter 1232/1546 - loss 0.16066108 - time (sec): 18.67 - samples/sec: 5353.33 - lr: 0.000024 - momentum: 0.000000
2023-10-20 09:07:45,922 epoch 3 - iter 1386/1546 - loss 0.16038985 - time (sec): 21.04 - samples/sec: 5281.58 - lr: 0.000024 - momentum: 0.000000
2023-10-20 09:07:48,314 epoch 3 - iter 1540/1546 - loss 0.15944440 - time (sec): 23.44 - samples/sec: 5277.46 - lr: 0.000023 - momentum: 0.000000
2023-10-20 09:07:48,405 ----------------------------------------------------------------------------------------------------
2023-10-20 09:07:48,405 EPOCH 3 done: loss 0.1593 - lr: 0.000023
2023-10-20 09:07:49,466 DEV : loss 0.0896943062543869 - f1-score (micro avg) 0.5099
2023-10-20 09:07:49,477 saving best model
2023-10-20 09:07:49,512 ----------------------------------------------------------------------------------------------------
2023-10-20 09:07:51,999 epoch 4 - iter 154/1546 - loss 0.15769143 - time (sec): 2.49 - samples/sec: 5070.16 - lr: 0.000023 - momentum: 0.000000
2023-10-20 09:07:54,306 epoch 4 - iter 308/1546 - loss 0.14433493 - time (sec): 4.79 - samples/sec: 5182.99 - lr: 0.000023 - momentum: 0.000000
2023-10-20 09:07:56,624 epoch 4 - iter 462/1546 - loss 0.15210379 - time (sec): 7.11 - samples/sec: 5030.59 - lr: 0.000022 - momentum: 0.000000
2023-10-20 09:07:58,995 epoch 4 - iter 616/1546 - loss 0.15295122 - time (sec): 9.48 - samples/sec: 5109.69 - lr: 0.000022 - momentum: 0.000000
2023-10-20 09:08:01,502 epoch 4 - iter 770/1546 - loss 0.15434920 - time (sec): 11.99 - samples/sec: 5061.65 - lr: 0.000022 - momentum: 0.000000
2023-10-20 09:08:03,980 epoch 4 - iter 924/1546 - loss 0.15086170 - time (sec): 14.47 - samples/sec: 5039.38 - lr: 0.000021 - momentum: 0.000000
2023-10-20 09:08:06,343 epoch 4 - iter 1078/1546 - loss 0.14718520 - time (sec): 16.83 - samples/sec: 5118.47 - lr: 0.000021 - momentum: 0.000000
2023-10-20 09:08:08,710 epoch 4 - iter 1232/1546 - loss 0.14718548 - time (sec): 19.20 - samples/sec: 5163.73 - lr: 0.000021 - momentum: 0.000000
2023-10-20 09:08:11,129 epoch 4 - iter 1386/1546 - loss 0.14835468 - time (sec): 21.62 - samples/sec: 5143.52 - lr: 0.000020 - momentum: 0.000000
2023-10-20 09:08:13,586 epoch 4 - iter 1540/1546 - loss 0.14859824 - time (sec): 24.07 - samples/sec: 5144.69 - lr: 0.000020 - momentum: 0.000000
2023-10-20 09:08:13,668 ----------------------------------------------------------------------------------------------------
2023-10-20 09:08:13,669 EPOCH 4 done: loss 0.1485 - lr: 0.000020
2023-10-20 09:08:14,728 DEV : loss 0.08766192942857742 - f1-score (micro avg) 0.5747
2023-10-20 09:08:14,738 saving best model
2023-10-20 09:08:14,771 ----------------------------------------------------------------------------------------------------
2023-10-20 09:08:17,083 epoch 5 - iter 154/1546 - loss 0.12769709 - time (sec): 2.31 - samples/sec: 5337.69 - lr: 0.000020 - momentum: 0.000000
2023-10-20 09:08:19,442 epoch 5 - iter 308/1546 - loss 0.13094414 - time (sec): 4.67 - samples/sec: 5128.06 - lr: 0.000019 - momentum: 0.000000
2023-10-20 09:08:21,893 epoch 5 - iter 462/1546 - loss 0.13300073 - time (sec): 7.12 - samples/sec: 5051.31 - lr: 0.000019 - momentum: 0.000000
2023-10-20 09:08:24,311 epoch 5 - iter 616/1546 - loss 0.13238196 - time (sec): 9.54 - samples/sec: 5160.07 - lr: 0.000019 - momentum: 0.000000
2023-10-20 09:08:26,690 epoch 5 - iter 770/1546 - loss 0.13428010 - time (sec): 11.92 - samples/sec: 5219.68 - lr: 0.000018 - momentum: 0.000000
2023-10-20 09:08:29,059 epoch 5 - iter 924/1546 - loss 0.13243603 - time (sec): 14.29 - samples/sec: 5203.51 - lr: 0.000018 - momentum: 0.000000
2023-10-20 09:08:31,427 epoch 5 - iter 1078/1546 - loss 0.13095759 - time (sec): 16.66 - samples/sec: 5184.83 - lr: 0.000018 - momentum: 0.000000
2023-10-20 09:08:33,859 epoch 5 - iter 1232/1546 - loss 0.13464576 - time (sec): 19.09 - samples/sec: 5174.45 - lr: 0.000017 - momentum: 0.000000
2023-10-20 09:08:36,291 epoch 5 - iter 1386/1546 - loss 0.13608735 - time (sec): 21.52 - samples/sec: 5199.06 - lr: 0.000017 - momentum: 0.000000
2023-10-20 09:08:38,636 epoch 5 - iter 1540/1546 - loss 0.13658634 - time (sec): 23.86 - samples/sec: 5187.95 - lr: 0.000017 - momentum: 0.000000
2023-10-20 09:08:38,723 ----------------------------------------------------------------------------------------------------
2023-10-20 09:08:38,723 EPOCH 5 done: loss 0.1363 - lr: 0.000017
2023-10-20 09:08:39,806 DEV : loss 0.08561883121728897 - f1-score (micro avg) 0.588
2023-10-20 09:08:39,817 saving best model
2023-10-20 09:08:39,849 ----------------------------------------------------------------------------------------------------
2023-10-20 09:08:42,216 epoch 6 - iter 154/1546 - loss 0.10872815 - time (sec): 2.37 - samples/sec: 5045.83 - lr: 0.000016 - momentum: 0.000000
2023-10-20 09:08:44,590 epoch 6 - iter 308/1546 - loss 0.11960726 - time (sec): 4.74 - samples/sec: 5003.13 - lr: 0.000016 - momentum: 0.000000
2023-10-20 09:08:46,973 epoch 6 - iter 462/1546 - loss 0.13441599 - time (sec): 7.12 - samples/sec: 5024.85 - lr: 0.000016 - momentum: 0.000000
2023-10-20 09:08:49,131 epoch 6 - iter 616/1546 - loss 0.13530848 - time (sec): 9.28 - samples/sec: 5247.57 - lr: 0.000015 - momentum: 0.000000
2023-10-20 09:08:51,443 epoch 6 - iter 770/1546 - loss 0.14041532 - time (sec): 11.59 - samples/sec: 5202.06 - lr: 0.000015 - momentum: 0.000000
2023-10-20 09:08:53,814 epoch 6 - iter 924/1546 - loss 0.13528183 - time (sec): 13.96 - samples/sec: 5249.94 - lr: 0.000015 - momentum: 0.000000
2023-10-20 09:08:56,177 epoch 6 - iter 1078/1546 - loss 0.13227395 - time (sec): 16.33 - samples/sec: 5253.32 - lr: 0.000014 - momentum: 0.000000
2023-10-20 09:08:58,522 epoch 6 - iter 1232/1546 - loss 0.13103216 - time (sec): 18.67 - samples/sec: 5297.52 - lr: 0.000014 - momentum: 0.000000
2023-10-20 09:09:00,904 epoch 6 - iter 1386/1546 - loss 0.12991021 - time (sec): 21.05 - samples/sec: 5249.58 - lr: 0.000014 - momentum: 0.000000
2023-10-20 09:09:03,307 epoch 6 - iter 1540/1546 - loss 0.13230097 - time (sec): 23.46 - samples/sec: 5276.90 - lr: 0.000013 - momentum: 0.000000
2023-10-20 09:09:03,406 ----------------------------------------------------------------------------------------------------
2023-10-20 09:09:03,406 EPOCH 6 done: loss 0.1321 - lr: 0.000013
2023-10-20 09:09:04,491 DEV : loss 0.08819162845611572 - f1-score (micro avg) 0.6039
2023-10-20 09:09:04,502 saving best model
2023-10-20 09:09:04,536 ----------------------------------------------------------------------------------------------------
2023-10-20 09:09:07,006 epoch 7 - iter 154/1546 - loss 0.11723149 - time (sec): 2.47 - samples/sec: 5437.44 - lr: 0.000013 - momentum: 0.000000
2023-10-20 09:09:09,369 epoch 7 - iter 308/1546 - loss 0.11870071 - time (sec): 4.83 - samples/sec: 5146.87 - lr: 0.000013 - momentum: 0.000000
2023-10-20 09:09:11,751 epoch 7 - iter 462/1546 - loss 0.11534894 - time (sec): 7.21 - samples/sec: 5251.42 - lr: 0.000012 - momentum: 0.000000
2023-10-20 09:09:14,090 epoch 7 - iter 616/1546 - loss 0.12686834 - time (sec): 9.55 - samples/sec: 5179.35 - lr: 0.000012 - momentum: 0.000000
2023-10-20 09:09:16,467 epoch 7 - iter 770/1546 - loss 0.12653362 - time (sec): 11.93 - samples/sec: 5208.68 - lr: 0.000012 - momentum: 0.000000
2023-10-20 09:09:18,835 epoch 7 - iter 924/1546 - loss 0.12736348 - time (sec): 14.30 - samples/sec: 5228.86 - lr: 0.000011 - momentum: 0.000000
2023-10-20 09:09:21,220 epoch 7 - iter 1078/1546 - loss 0.12994923 - time (sec): 16.68 - samples/sec: 5236.09 - lr: 0.000011 - momentum: 0.000000
2023-10-20 09:09:23,602 epoch 7 - iter 1232/1546 - loss 0.12775059 - time (sec): 19.06 - samples/sec: 5245.07 - lr: 0.000011 - momentum: 0.000000
2023-10-20 09:09:26,092 epoch 7 - iter 1386/1546 - loss 0.12761801 - time (sec): 21.56 - samples/sec: 5181.77 - lr: 0.000010 - momentum: 0.000000
2023-10-20 09:09:28,442 epoch 7 - iter 1540/1546 - loss 0.12680413 - time (sec): 23.91 - samples/sec: 5179.14 - lr: 0.000010 - momentum: 0.000000
2023-10-20 09:09:28,532 ----------------------------------------------------------------------------------------------------
2023-10-20 09:09:28,532 EPOCH 7 done: loss 0.1266 - lr: 0.000010
2023-10-20 09:09:29,594 DEV : loss 0.08853663504123688 - f1-score (micro avg) 0.5949
2023-10-20 09:09:29,605 ----------------------------------------------------------------------------------------------------
2023-10-20 09:09:31,902 epoch 8 - iter 154/1546 - loss 0.10560379 - time (sec): 2.30 - samples/sec: 5298.60 - lr: 0.000010 - momentum: 0.000000
2023-10-20 09:09:34,349 epoch 8 - iter 308/1546 - loss 0.12872546 - time (sec): 4.74 - samples/sec: 5252.11 - lr: 0.000009 - momentum: 0.000000
2023-10-20 09:09:36,720 epoch 8 - iter 462/1546 - loss 0.13119470 - time (sec): 7.11 - samples/sec: 5187.26 - lr: 0.000009 - momentum: 0.000000
2023-10-20 09:09:39,051 epoch 8 - iter 616/1546 - loss 0.12502116 - time (sec): 9.45 - samples/sec: 5215.10 - lr: 0.000009 - momentum: 0.000000
2023-10-20 09:09:41,413 epoch 8 - iter 770/1546 - loss 0.12112850 - time (sec): 11.81 - samples/sec: 5278.34 - lr: 0.000008 - momentum: 0.000000
2023-10-20 09:09:43,859 epoch 8 - iter 924/1546 - loss 0.12463981 - time (sec): 14.25 - samples/sec: 5309.41 - lr: 0.000008 - momentum: 0.000000
2023-10-20 09:09:46,162 epoch 8 - iter 1078/1546 - loss 0.12423906 - time (sec): 16.56 - samples/sec: 5263.09 - lr: 0.000008 - momentum: 0.000000
2023-10-20 09:09:48,482 epoch 8 - iter 1232/1546 - loss 0.12613505 - time (sec): 18.88 - samples/sec: 5225.70 - lr: 0.000007 - momentum: 0.000000
2023-10-20 09:09:50,918 epoch 8 - iter 1386/1546 - loss 0.12382574 - time (sec): 21.31 - samples/sec: 5199.99 - lr: 0.000007 - momentum: 0.000000
2023-10-20 09:09:53,355 epoch 8 - iter 1540/1546 - loss 0.12351904 - time (sec): 23.75 - samples/sec: 5219.53 - lr: 0.000007 - momentum: 0.000000
2023-10-20 09:09:53,443 ----------------------------------------------------------------------------------------------------
2023-10-20 09:09:53,443 EPOCH 8 done: loss 0.1233 - lr: 0.000007
2023-10-20 09:09:54,533 DEV : loss 0.09046540409326553 - f1-score (micro avg) 0.6062
2023-10-20 09:09:54,545 saving best model
2023-10-20 09:09:54,583 ----------------------------------------------------------------------------------------------------
2023-10-20 09:09:56,973 epoch 9 - iter 154/1546 - loss 0.12165880 - time (sec): 2.39 - samples/sec: 5120.10 - lr: 0.000006 - momentum: 0.000000
2023-10-20 09:09:59,354 epoch 9 - iter 308/1546 - loss 0.11720672 - time (sec): 4.77 - samples/sec: 5174.59 - lr: 0.000006 - momentum: 0.000000
2023-10-20 09:10:01,652 epoch 9 - iter 462/1546 - loss 0.10950130 - time (sec): 7.07 - samples/sec: 5397.14 - lr: 0.000006 - momentum: 0.000000
2023-10-20 09:10:03,813 epoch 9 - iter 616/1546 - loss 0.11189291 - time (sec): 9.23 - samples/sec: 5423.30 - lr: 0.000005 - momentum: 0.000000
2023-10-20 09:10:05,958 epoch 9 - iter 770/1546 - loss 0.11566478 - time (sec): 11.37 - samples/sec: 5567.08 - lr: 0.000005 - momentum: 0.000000
2023-10-20 09:10:08,178 epoch 9 - iter 924/1546 - loss 0.12061263 - time (sec): 13.59 - samples/sec: 5535.08 - lr: 0.000005 - momentum: 0.000000
2023-10-20 09:10:10,589 epoch 9 - iter 1078/1546 - loss 0.12124174 - time (sec): 16.01 - samples/sec: 5494.20 - lr: 0.000004 - momentum: 0.000000
2023-10-20 09:10:12,970 epoch 9 - iter 1232/1546 - loss 0.12235867 - time (sec): 18.39 - samples/sec: 5445.86 - lr: 0.000004 - momentum: 0.000000
2023-10-20 09:10:15,300 epoch 9 - iter 1386/1546 - loss 0.12037485 - time (sec): 20.72 - samples/sec: 5384.65 - lr: 0.000004 - momentum: 0.000000
2023-10-20 09:10:17,700 epoch 9 - iter 1540/1546 - loss 0.11940803 - time (sec): 23.12 - samples/sec: 5359.10 - lr: 0.000003 - momentum: 0.000000
2023-10-20 09:10:17,787 ----------------------------------------------------------------------------------------------------
2023-10-20 09:10:17,787 EPOCH 9 done: loss 0.1193 - lr: 0.000003
2023-10-20 09:10:18,860 DEV : loss 0.09189001470804214 - f1-score (micro avg) 0.6117
2023-10-20 09:10:18,871 saving best model
2023-10-20 09:10:18,902 ----------------------------------------------------------------------------------------------------
2023-10-20 09:10:21,118 epoch 10 - iter 154/1546 - loss 0.12687807 - time (sec): 2.22 - samples/sec: 5422.87 - lr: 0.000003 - momentum: 0.000000
2023-10-20 09:10:23,371 epoch 10 - iter 308/1546 - loss 0.12068074 - time (sec): 4.47 - samples/sec: 5568.10 - lr: 0.000003 - momentum: 0.000000
2023-10-20 09:10:25,575 epoch 10 - iter 462/1546 - loss 0.12030419 - time (sec): 6.67 - samples/sec: 5713.42 - lr: 0.000002 - momentum: 0.000000
2023-10-20 09:10:27,713 epoch 10 - iter 616/1546 - loss 0.11605372 - time (sec): 8.81 - samples/sec: 5773.02 - lr: 0.000002 - momentum: 0.000000
2023-10-20 09:10:30,067 epoch 10 - iter 770/1546 - loss 0.11801476 - time (sec): 11.16 - samples/sec: 5649.92 - lr: 0.000002 - momentum: 0.000000
2023-10-20 09:10:32,425 epoch 10 - iter 924/1546 - loss 0.11442404 - time (sec): 13.52 - samples/sec: 5565.71 - lr: 0.000001 - momentum: 0.000000
2023-10-20 09:10:34,785 epoch 10 - iter 1078/1546 - loss 0.11142937 - time (sec): 15.88 - samples/sec: 5542.35 - lr: 0.000001 - momentum: 0.000000
2023-10-20 09:10:37,138 epoch 10 - iter 1232/1546 - loss 0.11098365 - time (sec): 18.24 - samples/sec: 5458.00 - lr: 0.000001 - momentum: 0.000000
2023-10-20 09:10:39,490 epoch 10 - iter 1386/1546 - loss 0.11535574 - time (sec): 20.59 - samples/sec: 5430.96 - lr: 0.000000 - momentum: 0.000000
2023-10-20 09:10:41,875 epoch 10 - iter 1540/1546 - loss 0.11683715 - time (sec): 22.97 - samples/sec: 5397.34 - lr: 0.000000 - momentum: 0.000000
2023-10-20 09:10:41,963 ----------------------------------------------------------------------------------------------------
2023-10-20 09:10:41,963 EPOCH 10 done: loss 0.1167 - lr: 0.000000
2023-10-20 09:10:43,032 DEV : loss 0.0931963250041008 - f1-score (micro avg) 0.6154
2023-10-20 09:10:43,043 saving best model
2023-10-20 09:10:43,106 ----------------------------------------------------------------------------------------------------
2023-10-20 09:10:43,106 Loading model from best epoch ...
2023-10-20 09:10:43,182 SequenceTagger predicts: Dictionary with 13 tags: O, S-LOC, B-LOC, E-LOC, I-LOC, S-BUILDING, B-BUILDING, E-BUILDING, I-BUILDING, S-STREET, B-STREET, E-STREET, I-STREET
2023-10-20 09:10:46,022
Results:
- F-score (micro) 0.5552
- F-score (macro) 0.2272
- Accuracy 0.4001
By class:
precision recall f1-score support
LOC 0.6223 0.6321 0.6272 946
BUILDING 0.1333 0.0108 0.0200 185
STREET 0.5000 0.0179 0.0345 56
micro avg 0.6145 0.5063 0.5552 1187
macro avg 0.4185 0.2203 0.2272 1187
weighted avg 0.5403 0.5063 0.5046 1187
2023-10-20 09:10:46,022 ----------------------------------------------------------------------------------------------------
|