File size: 23,847 Bytes
553aaf6 |
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 |
2023-10-17 09:50:37,113 ----------------------------------------------------------------------------------------------------
2023-10-17 09:50:37,114 Model: "SequenceTagger(
(embeddings): TransformerWordEmbeddings(
(model): ElectraModel(
(embeddings): ElectraEmbeddings(
(word_embeddings): Embedding(32001, 768)
(position_embeddings): Embedding(512, 768)
(token_type_embeddings): Embedding(2, 768)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(encoder): ElectraEncoder(
(layer): ModuleList(
(0-11): 12 x ElectraLayer(
(attention): ElectraAttention(
(self): ElectraSelfAttention(
(query): Linear(in_features=768, out_features=768, bias=True)
(key): Linear(in_features=768, out_features=768, bias=True)
(value): Linear(in_features=768, out_features=768, bias=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(output): ElectraSelfOutput(
(dense): Linear(in_features=768, out_features=768, bias=True)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
(intermediate): ElectraIntermediate(
(dense): Linear(in_features=768, out_features=3072, bias=True)
(intermediate_act_fn): GELUActivation()
)
(output): ElectraOutput(
(dense): Linear(in_features=3072, out_features=768, bias=True)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
)
)
)
(locked_dropout): LockedDropout(p=0.5)
(linear): Linear(in_features=768, out_features=13, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-17 09:50:37,114 ----------------------------------------------------------------------------------------------------
2023-10-17 09:50:37,115 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
2023-10-17 09:50:37,115 ----------------------------------------------------------------------------------------------------
2023-10-17 09:50:37,115 Train: 6183 sentences
2023-10-17 09:50:37,115 (train_with_dev=False, train_with_test=False)
2023-10-17 09:50:37,115 ----------------------------------------------------------------------------------------------------
2023-10-17 09:50:37,115 Training Params:
2023-10-17 09:50:37,115 - learning_rate: "5e-05"
2023-10-17 09:50:37,115 - mini_batch_size: "8"
2023-10-17 09:50:37,115 - max_epochs: "10"
2023-10-17 09:50:37,115 - shuffle: "True"
2023-10-17 09:50:37,115 ----------------------------------------------------------------------------------------------------
2023-10-17 09:50:37,116 Plugins:
2023-10-17 09:50:37,116 - TensorboardLogger
2023-10-17 09:50:37,116 - LinearScheduler | warmup_fraction: '0.1'
2023-10-17 09:50:37,116 ----------------------------------------------------------------------------------------------------
2023-10-17 09:50:37,116 Final evaluation on model from best epoch (best-model.pt)
2023-10-17 09:50:37,116 - metric: "('micro avg', 'f1-score')"
2023-10-17 09:50:37,116 ----------------------------------------------------------------------------------------------------
2023-10-17 09:50:37,116 Computation:
2023-10-17 09:50:37,116 - compute on device: cuda:0
2023-10-17 09:50:37,116 - embedding storage: none
2023-10-17 09:50:37,116 ----------------------------------------------------------------------------------------------------
2023-10-17 09:50:37,116 Model training base path: "hmbench-topres19th/en-hmteams/teams-base-historic-multilingual-discriminator-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-1"
2023-10-17 09:50:37,116 ----------------------------------------------------------------------------------------------------
2023-10-17 09:50:37,116 ----------------------------------------------------------------------------------------------------
2023-10-17 09:50:37,117 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-17 09:50:44,240 epoch 1 - iter 77/773 - loss 1.99803994 - time (sec): 7.12 - samples/sec: 1805.03 - lr: 0.000005 - momentum: 0.000000
2023-10-17 09:50:51,561 epoch 1 - iter 154/773 - loss 1.14101515 - time (sec): 14.44 - samples/sec: 1736.92 - lr: 0.000010 - momentum: 0.000000
2023-10-17 09:50:58,592 epoch 1 - iter 231/773 - loss 0.81080166 - time (sec): 21.47 - samples/sec: 1742.53 - lr: 0.000015 - momentum: 0.000000
2023-10-17 09:51:05,952 epoch 1 - iter 308/773 - loss 0.63449031 - time (sec): 28.83 - samples/sec: 1748.52 - lr: 0.000020 - momentum: 0.000000
2023-10-17 09:51:13,243 epoch 1 - iter 385/773 - loss 0.53077408 - time (sec): 36.12 - samples/sec: 1734.04 - lr: 0.000025 - momentum: 0.000000
2023-10-17 09:51:20,423 epoch 1 - iter 462/773 - loss 0.46116490 - time (sec): 43.31 - samples/sec: 1730.19 - lr: 0.000030 - momentum: 0.000000
2023-10-17 09:51:28,021 epoch 1 - iter 539/773 - loss 0.41868646 - time (sec): 50.90 - samples/sec: 1703.61 - lr: 0.000035 - momentum: 0.000000
2023-10-17 09:51:35,107 epoch 1 - iter 616/773 - loss 0.38248750 - time (sec): 57.99 - samples/sec: 1703.90 - lr: 0.000040 - momentum: 0.000000
2023-10-17 09:51:42,607 epoch 1 - iter 693/773 - loss 0.34805302 - time (sec): 65.49 - samples/sec: 1703.78 - lr: 0.000045 - momentum: 0.000000
2023-10-17 09:51:49,968 epoch 1 - iter 770/773 - loss 0.32227332 - time (sec): 72.85 - samples/sec: 1702.09 - lr: 0.000050 - momentum: 0.000000
2023-10-17 09:51:50,232 ----------------------------------------------------------------------------------------------------
2023-10-17 09:51:50,232 EPOCH 1 done: loss 0.3216 - lr: 0.000050
2023-10-17 09:51:52,930 DEV : loss 0.05330459401011467 - f1-score (micro avg) 0.7489
2023-10-17 09:51:52,960 saving best model
2023-10-17 09:51:53,514 ----------------------------------------------------------------------------------------------------
2023-10-17 09:52:00,448 epoch 2 - iter 77/773 - loss 0.09677124 - time (sec): 6.93 - samples/sec: 1704.73 - lr: 0.000049 - momentum: 0.000000
2023-10-17 09:52:07,574 epoch 2 - iter 154/773 - loss 0.08008441 - time (sec): 14.06 - samples/sec: 1718.53 - lr: 0.000049 - momentum: 0.000000
2023-10-17 09:52:14,632 epoch 2 - iter 231/773 - loss 0.07559641 - time (sec): 21.12 - samples/sec: 1785.39 - lr: 0.000048 - momentum: 0.000000
2023-10-17 09:52:21,666 epoch 2 - iter 308/773 - loss 0.07646491 - time (sec): 28.15 - samples/sec: 1780.87 - lr: 0.000048 - momentum: 0.000000
2023-10-17 09:52:28,587 epoch 2 - iter 385/773 - loss 0.07753478 - time (sec): 35.07 - samples/sec: 1789.21 - lr: 0.000047 - momentum: 0.000000
2023-10-17 09:52:35,688 epoch 2 - iter 462/773 - loss 0.07853793 - time (sec): 42.17 - samples/sec: 1776.69 - lr: 0.000047 - momentum: 0.000000
2023-10-17 09:52:42,856 epoch 2 - iter 539/773 - loss 0.07587851 - time (sec): 49.34 - samples/sec: 1772.60 - lr: 0.000046 - momentum: 0.000000
2023-10-17 09:52:50,035 epoch 2 - iter 616/773 - loss 0.07502133 - time (sec): 56.52 - samples/sec: 1778.78 - lr: 0.000046 - momentum: 0.000000
2023-10-17 09:52:57,050 epoch 2 - iter 693/773 - loss 0.07354922 - time (sec): 63.53 - samples/sec: 1766.03 - lr: 0.000045 - momentum: 0.000000
2023-10-17 09:53:03,992 epoch 2 - iter 770/773 - loss 0.07444958 - time (sec): 70.48 - samples/sec: 1759.81 - lr: 0.000044 - momentum: 0.000000
2023-10-17 09:53:04,245 ----------------------------------------------------------------------------------------------------
2023-10-17 09:53:04,245 EPOCH 2 done: loss 0.0746 - lr: 0.000044
2023-10-17 09:53:07,171 DEV : loss 0.06132051348686218 - f1-score (micro avg) 0.6713
2023-10-17 09:53:07,199 ----------------------------------------------------------------------------------------------------
2023-10-17 09:53:13,887 epoch 3 - iter 77/773 - loss 0.04795621 - time (sec): 6.69 - samples/sec: 1748.40 - lr: 0.000044 - momentum: 0.000000
2023-10-17 09:53:21,207 epoch 3 - iter 154/773 - loss 0.04943447 - time (sec): 14.01 - samples/sec: 1772.77 - lr: 0.000043 - momentum: 0.000000
2023-10-17 09:53:29,211 epoch 3 - iter 231/773 - loss 0.04886716 - time (sec): 22.01 - samples/sec: 1734.92 - lr: 0.000043 - momentum: 0.000000
2023-10-17 09:53:36,821 epoch 3 - iter 308/773 - loss 0.04640129 - time (sec): 29.62 - samples/sec: 1708.16 - lr: 0.000042 - momentum: 0.000000
2023-10-17 09:53:44,470 epoch 3 - iter 385/773 - loss 0.04816842 - time (sec): 37.27 - samples/sec: 1675.52 - lr: 0.000042 - momentum: 0.000000
2023-10-17 09:53:52,356 epoch 3 - iter 462/773 - loss 0.05098456 - time (sec): 45.15 - samples/sec: 1662.59 - lr: 0.000041 - momentum: 0.000000
2023-10-17 09:53:59,926 epoch 3 - iter 539/773 - loss 0.05101299 - time (sec): 52.73 - samples/sec: 1653.10 - lr: 0.000041 - momentum: 0.000000
2023-10-17 09:54:06,734 epoch 3 - iter 616/773 - loss 0.05078472 - time (sec): 59.53 - samples/sec: 1671.82 - lr: 0.000040 - momentum: 0.000000
2023-10-17 09:54:13,425 epoch 3 - iter 693/773 - loss 0.05251785 - time (sec): 66.22 - samples/sec: 1666.04 - lr: 0.000039 - momentum: 0.000000
2023-10-17 09:54:20,384 epoch 3 - iter 770/773 - loss 0.05247431 - time (sec): 73.18 - samples/sec: 1692.98 - lr: 0.000039 - momentum: 0.000000
2023-10-17 09:54:20,643 ----------------------------------------------------------------------------------------------------
2023-10-17 09:54:20,643 EPOCH 3 done: loss 0.0524 - lr: 0.000039
2023-10-17 09:54:23,554 DEV : loss 0.05568350851535797 - f1-score (micro avg) 0.7886
2023-10-17 09:54:23,582 saving best model
2023-10-17 09:54:24,988 ----------------------------------------------------------------------------------------------------
2023-10-17 09:54:31,745 epoch 4 - iter 77/773 - loss 0.03917114 - time (sec): 6.75 - samples/sec: 1903.55 - lr: 0.000038 - momentum: 0.000000
2023-10-17 09:54:38,162 epoch 4 - iter 154/773 - loss 0.03481858 - time (sec): 13.17 - samples/sec: 1862.13 - lr: 0.000038 - momentum: 0.000000
2023-10-17 09:54:45,104 epoch 4 - iter 231/773 - loss 0.03357236 - time (sec): 20.11 - samples/sec: 1865.14 - lr: 0.000037 - momentum: 0.000000
2023-10-17 09:54:52,050 epoch 4 - iter 308/773 - loss 0.03530401 - time (sec): 27.06 - samples/sec: 1847.93 - lr: 0.000037 - momentum: 0.000000
2023-10-17 09:54:58,597 epoch 4 - iter 385/773 - loss 0.03461690 - time (sec): 33.61 - samples/sec: 1857.33 - lr: 0.000036 - momentum: 0.000000
2023-10-17 09:55:05,499 epoch 4 - iter 462/773 - loss 0.03627980 - time (sec): 40.51 - samples/sec: 1859.14 - lr: 0.000036 - momentum: 0.000000
2023-10-17 09:55:12,971 epoch 4 - iter 539/773 - loss 0.03653929 - time (sec): 47.98 - samples/sec: 1834.41 - lr: 0.000035 - momentum: 0.000000
2023-10-17 09:55:19,826 epoch 4 - iter 616/773 - loss 0.03616690 - time (sec): 54.83 - samples/sec: 1818.25 - lr: 0.000034 - momentum: 0.000000
2023-10-17 09:55:27,192 epoch 4 - iter 693/773 - loss 0.03664448 - time (sec): 62.20 - samples/sec: 1792.69 - lr: 0.000034 - momentum: 0.000000
2023-10-17 09:55:34,633 epoch 4 - iter 770/773 - loss 0.03696210 - time (sec): 69.64 - samples/sec: 1779.84 - lr: 0.000033 - momentum: 0.000000
2023-10-17 09:55:34,898 ----------------------------------------------------------------------------------------------------
2023-10-17 09:55:34,898 EPOCH 4 done: loss 0.0372 - lr: 0.000033
2023-10-17 09:55:37,813 DEV : loss 0.09010311961174011 - f1-score (micro avg) 0.795
2023-10-17 09:55:37,844 saving best model
2023-10-17 09:55:39,252 ----------------------------------------------------------------------------------------------------
2023-10-17 09:55:46,196 epoch 5 - iter 77/773 - loss 0.02882609 - time (sec): 6.94 - samples/sec: 1709.68 - lr: 0.000033 - momentum: 0.000000
2023-10-17 09:55:53,133 epoch 5 - iter 154/773 - loss 0.02484139 - time (sec): 13.88 - samples/sec: 1750.43 - lr: 0.000032 - momentum: 0.000000
2023-10-17 09:56:00,065 epoch 5 - iter 231/773 - loss 0.02494855 - time (sec): 20.81 - samples/sec: 1738.11 - lr: 0.000032 - momentum: 0.000000
2023-10-17 09:56:07,071 epoch 5 - iter 308/773 - loss 0.02438607 - time (sec): 27.81 - samples/sec: 1740.43 - lr: 0.000031 - momentum: 0.000000
2023-10-17 09:56:14,264 epoch 5 - iter 385/773 - loss 0.02602136 - time (sec): 35.01 - samples/sec: 1753.53 - lr: 0.000031 - momentum: 0.000000
2023-10-17 09:56:21,203 epoch 5 - iter 462/773 - loss 0.02629902 - time (sec): 41.95 - samples/sec: 1764.17 - lr: 0.000030 - momentum: 0.000000
2023-10-17 09:56:28,542 epoch 5 - iter 539/773 - loss 0.02576945 - time (sec): 49.28 - samples/sec: 1755.14 - lr: 0.000029 - momentum: 0.000000
2023-10-17 09:56:35,568 epoch 5 - iter 616/773 - loss 0.02604685 - time (sec): 56.31 - samples/sec: 1752.21 - lr: 0.000029 - momentum: 0.000000
2023-10-17 09:56:42,730 epoch 5 - iter 693/773 - loss 0.02654183 - time (sec): 63.47 - samples/sec: 1763.08 - lr: 0.000028 - momentum: 0.000000
2023-10-17 09:56:50,088 epoch 5 - iter 770/773 - loss 0.02657177 - time (sec): 70.83 - samples/sec: 1747.09 - lr: 0.000028 - momentum: 0.000000
2023-10-17 09:56:50,394 ----------------------------------------------------------------------------------------------------
2023-10-17 09:56:50,395 EPOCH 5 done: loss 0.0266 - lr: 0.000028
2023-10-17 09:56:53,269 DEV : loss 0.1054750606417656 - f1-score (micro avg) 0.7785
2023-10-17 09:56:53,298 ----------------------------------------------------------------------------------------------------
2023-10-17 09:57:00,470 epoch 6 - iter 77/773 - loss 0.01388745 - time (sec): 7.17 - samples/sec: 1788.67 - lr: 0.000027 - momentum: 0.000000
2023-10-17 09:57:07,341 epoch 6 - iter 154/773 - loss 0.01126746 - time (sec): 14.04 - samples/sec: 1832.99 - lr: 0.000027 - momentum: 0.000000
2023-10-17 09:57:14,224 epoch 6 - iter 231/773 - loss 0.01379040 - time (sec): 20.92 - samples/sec: 1817.27 - lr: 0.000026 - momentum: 0.000000
2023-10-17 09:57:21,174 epoch 6 - iter 308/773 - loss 0.01635429 - time (sec): 27.87 - samples/sec: 1813.88 - lr: 0.000026 - momentum: 0.000000
2023-10-17 09:57:28,078 epoch 6 - iter 385/773 - loss 0.01738643 - time (sec): 34.78 - samples/sec: 1826.82 - lr: 0.000025 - momentum: 0.000000
2023-10-17 09:57:34,818 epoch 6 - iter 462/773 - loss 0.01857094 - time (sec): 41.52 - samples/sec: 1809.73 - lr: 0.000024 - momentum: 0.000000
2023-10-17 09:57:41,802 epoch 6 - iter 539/773 - loss 0.01798259 - time (sec): 48.50 - samples/sec: 1791.72 - lr: 0.000024 - momentum: 0.000000
2023-10-17 09:57:49,468 epoch 6 - iter 616/773 - loss 0.01750456 - time (sec): 56.17 - samples/sec: 1757.49 - lr: 0.000023 - momentum: 0.000000
2023-10-17 09:57:56,975 epoch 6 - iter 693/773 - loss 0.01733636 - time (sec): 63.68 - samples/sec: 1748.90 - lr: 0.000023 - momentum: 0.000000
2023-10-17 09:58:03,735 epoch 6 - iter 770/773 - loss 0.01777726 - time (sec): 70.44 - samples/sec: 1758.81 - lr: 0.000022 - momentum: 0.000000
2023-10-17 09:58:03,988 ----------------------------------------------------------------------------------------------------
2023-10-17 09:58:03,989 EPOCH 6 done: loss 0.0177 - lr: 0.000022
2023-10-17 09:58:07,150 DEV : loss 0.11583945155143738 - f1-score (micro avg) 0.7778
2023-10-17 09:58:07,202 ----------------------------------------------------------------------------------------------------
2023-10-17 09:58:13,949 epoch 7 - iter 77/773 - loss 0.00456103 - time (sec): 6.74 - samples/sec: 1737.20 - lr: 0.000022 - momentum: 0.000000
2023-10-17 09:58:20,748 epoch 7 - iter 154/773 - loss 0.01048448 - time (sec): 13.54 - samples/sec: 1752.44 - lr: 0.000021 - momentum: 0.000000
2023-10-17 09:58:27,539 epoch 7 - iter 231/773 - loss 0.01223233 - time (sec): 20.33 - samples/sec: 1781.78 - lr: 0.000021 - momentum: 0.000000
2023-10-17 09:58:34,652 epoch 7 - iter 308/773 - loss 0.01163819 - time (sec): 27.45 - samples/sec: 1783.17 - lr: 0.000020 - momentum: 0.000000
2023-10-17 09:58:42,661 epoch 7 - iter 385/773 - loss 0.01140701 - time (sec): 35.46 - samples/sec: 1736.84 - lr: 0.000019 - momentum: 0.000000
2023-10-17 09:58:50,477 epoch 7 - iter 462/773 - loss 0.01053450 - time (sec): 43.27 - samples/sec: 1708.48 - lr: 0.000019 - momentum: 0.000000
2023-10-17 09:58:57,739 epoch 7 - iter 539/773 - loss 0.01010124 - time (sec): 50.53 - samples/sec: 1701.36 - lr: 0.000018 - momentum: 0.000000
2023-10-17 09:59:04,851 epoch 7 - iter 616/773 - loss 0.00993634 - time (sec): 57.65 - samples/sec: 1718.81 - lr: 0.000018 - momentum: 0.000000
2023-10-17 09:59:12,153 epoch 7 - iter 693/773 - loss 0.01039778 - time (sec): 64.95 - samples/sec: 1722.45 - lr: 0.000017 - momentum: 0.000000
2023-10-17 09:59:19,114 epoch 7 - iter 770/773 - loss 0.01064432 - time (sec): 71.91 - samples/sec: 1720.07 - lr: 0.000017 - momentum: 0.000000
2023-10-17 09:59:19,398 ----------------------------------------------------------------------------------------------------
2023-10-17 09:59:19,398 EPOCH 7 done: loss 0.0106 - lr: 0.000017
2023-10-17 09:59:22,806 DEV : loss 0.11872334033250809 - f1-score (micro avg) 0.818
2023-10-17 09:59:22,836 saving best model
2023-10-17 09:59:24,267 ----------------------------------------------------------------------------------------------------
2023-10-17 09:59:31,418 epoch 8 - iter 77/773 - loss 0.01473996 - time (sec): 7.14 - samples/sec: 1732.87 - lr: 0.000016 - momentum: 0.000000
2023-10-17 09:59:39,170 epoch 8 - iter 154/773 - loss 0.01203286 - time (sec): 14.90 - samples/sec: 1696.02 - lr: 0.000016 - momentum: 0.000000
2023-10-17 09:59:46,392 epoch 8 - iter 231/773 - loss 0.01161199 - time (sec): 22.12 - samples/sec: 1689.65 - lr: 0.000015 - momentum: 0.000000
2023-10-17 09:59:53,350 epoch 8 - iter 308/773 - loss 0.00966991 - time (sec): 29.08 - samples/sec: 1702.11 - lr: 0.000014 - momentum: 0.000000
2023-10-17 10:00:00,653 epoch 8 - iter 385/773 - loss 0.00955744 - time (sec): 36.38 - samples/sec: 1690.94 - lr: 0.000014 - momentum: 0.000000
2023-10-17 10:00:08,701 epoch 8 - iter 462/773 - loss 0.00952576 - time (sec): 44.43 - samples/sec: 1680.49 - lr: 0.000013 - momentum: 0.000000
2023-10-17 10:00:16,534 epoch 8 - iter 539/773 - loss 0.00909750 - time (sec): 52.26 - samples/sec: 1676.63 - lr: 0.000013 - momentum: 0.000000
2023-10-17 10:00:24,416 epoch 8 - iter 616/773 - loss 0.00903899 - time (sec): 60.14 - samples/sec: 1654.18 - lr: 0.000012 - momentum: 0.000000
2023-10-17 10:00:31,889 epoch 8 - iter 693/773 - loss 0.00893252 - time (sec): 67.61 - samples/sec: 1641.23 - lr: 0.000012 - momentum: 0.000000
2023-10-17 10:00:39,288 epoch 8 - iter 770/773 - loss 0.00862898 - time (sec): 75.01 - samples/sec: 1652.22 - lr: 0.000011 - momentum: 0.000000
2023-10-17 10:00:39,604 ----------------------------------------------------------------------------------------------------
2023-10-17 10:00:39,605 EPOCH 8 done: loss 0.0086 - lr: 0.000011
2023-10-17 10:00:42,778 DEV : loss 0.12361815571784973 - f1-score (micro avg) 0.778
2023-10-17 10:00:42,809 ----------------------------------------------------------------------------------------------------
2023-10-17 10:00:50,878 epoch 9 - iter 77/773 - loss 0.00424221 - time (sec): 8.07 - samples/sec: 1562.76 - lr: 0.000011 - momentum: 0.000000
2023-10-17 10:00:58,265 epoch 9 - iter 154/773 - loss 0.00382206 - time (sec): 15.45 - samples/sec: 1587.14 - lr: 0.000010 - momentum: 0.000000
2023-10-17 10:01:05,562 epoch 9 - iter 231/773 - loss 0.00494659 - time (sec): 22.75 - samples/sec: 1643.92 - lr: 0.000009 - momentum: 0.000000
2023-10-17 10:01:13,223 epoch 9 - iter 308/773 - loss 0.00538922 - time (sec): 30.41 - samples/sec: 1615.47 - lr: 0.000009 - momentum: 0.000000
2023-10-17 10:01:20,951 epoch 9 - iter 385/773 - loss 0.00585671 - time (sec): 38.14 - samples/sec: 1622.67 - lr: 0.000008 - momentum: 0.000000
2023-10-17 10:01:28,146 epoch 9 - iter 462/773 - loss 0.00592916 - time (sec): 45.33 - samples/sec: 1632.48 - lr: 0.000008 - momentum: 0.000000
2023-10-17 10:01:35,335 epoch 9 - iter 539/773 - loss 0.00543064 - time (sec): 52.52 - samples/sec: 1654.27 - lr: 0.000007 - momentum: 0.000000
2023-10-17 10:01:42,299 epoch 9 - iter 616/773 - loss 0.00549350 - time (sec): 59.49 - samples/sec: 1664.43 - lr: 0.000007 - momentum: 0.000000
2023-10-17 10:01:49,520 epoch 9 - iter 693/773 - loss 0.00513001 - time (sec): 66.71 - samples/sec: 1683.75 - lr: 0.000006 - momentum: 0.000000
2023-10-17 10:01:56,696 epoch 9 - iter 770/773 - loss 0.00549165 - time (sec): 73.89 - samples/sec: 1675.81 - lr: 0.000006 - momentum: 0.000000
2023-10-17 10:01:56,967 ----------------------------------------------------------------------------------------------------
2023-10-17 10:01:56,967 EPOCH 9 done: loss 0.0055 - lr: 0.000006
2023-10-17 10:01:59,842 DEV : loss 0.12155171483755112 - f1-score (micro avg) 0.7975
2023-10-17 10:01:59,871 ----------------------------------------------------------------------------------------------------
2023-10-17 10:02:06,780 epoch 10 - iter 77/773 - loss 0.00669265 - time (sec): 6.91 - samples/sec: 1812.77 - lr: 0.000005 - momentum: 0.000000
2023-10-17 10:02:13,747 epoch 10 - iter 154/773 - loss 0.00465994 - time (sec): 13.87 - samples/sec: 1786.79 - lr: 0.000005 - momentum: 0.000000
2023-10-17 10:02:20,720 epoch 10 - iter 231/773 - loss 0.00329970 - time (sec): 20.85 - samples/sec: 1815.30 - lr: 0.000004 - momentum: 0.000000
2023-10-17 10:02:27,633 epoch 10 - iter 308/773 - loss 0.00309264 - time (sec): 27.76 - samples/sec: 1810.49 - lr: 0.000003 - momentum: 0.000000
2023-10-17 10:02:34,497 epoch 10 - iter 385/773 - loss 0.00320190 - time (sec): 34.62 - samples/sec: 1805.14 - lr: 0.000003 - momentum: 0.000000
2023-10-17 10:02:41,428 epoch 10 - iter 462/773 - loss 0.00367371 - time (sec): 41.56 - samples/sec: 1786.35 - lr: 0.000002 - momentum: 0.000000
2023-10-17 10:02:48,391 epoch 10 - iter 539/773 - loss 0.00352687 - time (sec): 48.52 - samples/sec: 1793.81 - lr: 0.000002 - momentum: 0.000000
2023-10-17 10:02:55,392 epoch 10 - iter 616/773 - loss 0.00364918 - time (sec): 55.52 - samples/sec: 1780.19 - lr: 0.000001 - momentum: 0.000000
2023-10-17 10:03:02,317 epoch 10 - iter 693/773 - loss 0.00341373 - time (sec): 62.44 - samples/sec: 1784.63 - lr: 0.000001 - momentum: 0.000000
2023-10-17 10:03:09,513 epoch 10 - iter 770/773 - loss 0.00316098 - time (sec): 69.64 - samples/sec: 1778.30 - lr: 0.000000 - momentum: 0.000000
2023-10-17 10:03:09,772 ----------------------------------------------------------------------------------------------------
2023-10-17 10:03:09,772 EPOCH 10 done: loss 0.0031 - lr: 0.000000
2023-10-17 10:03:12,617 DEV : loss 0.12296666949987411 - f1-score (micro avg) 0.7942
2023-10-17 10:03:13,213 ----------------------------------------------------------------------------------------------------
2023-10-17 10:03:13,215 Loading model from best epoch ...
2023-10-17 10:03:15,488 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
2023-10-17 10:03:23,976
Results:
- F-score (micro) 0.7995
- F-score (macro) 0.7004
- Accuracy 0.6805
By class:
precision recall f1-score support
LOC 0.8866 0.8182 0.8510 946
BUILDING 0.6207 0.4865 0.5455 185
STREET 0.7551 0.6607 0.7048 56
micro avg 0.8444 0.7591 0.7995 1187
macro avg 0.7541 0.6551 0.7004 1187
weighted avg 0.8390 0.7591 0.7965 1187
2023-10-17 10:03:23,976 ----------------------------------------------------------------------------------------------------
|