File size: 24,048 Bytes
3b41ee8 |
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 |
2023-10-17 10:37:03,892 ----------------------------------------------------------------------------------------------------
2023-10-17 10:37:03,893 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=25, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-17 10:37:03,894 ----------------------------------------------------------------------------------------------------
2023-10-17 10:37:03,894 MultiCorpus: 966 train + 219 dev + 204 test sentences
- NER_HIPE_2022 Corpus: 966 train + 219 dev + 204 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/ajmc/fr/with_doc_seperator
2023-10-17 10:37:03,894 ----------------------------------------------------------------------------------------------------
2023-10-17 10:37:03,894 Train: 966 sentences
2023-10-17 10:37:03,894 (train_with_dev=False, train_with_test=False)
2023-10-17 10:37:03,894 ----------------------------------------------------------------------------------------------------
2023-10-17 10:37:03,894 Training Params:
2023-10-17 10:37:03,894 - learning_rate: "5e-05"
2023-10-17 10:37:03,894 - mini_batch_size: "4"
2023-10-17 10:37:03,894 - max_epochs: "10"
2023-10-17 10:37:03,894 - shuffle: "True"
2023-10-17 10:37:03,894 ----------------------------------------------------------------------------------------------------
2023-10-17 10:37:03,894 Plugins:
2023-10-17 10:37:03,894 - TensorboardLogger
2023-10-17 10:37:03,894 - LinearScheduler | warmup_fraction: '0.1'
2023-10-17 10:37:03,894 ----------------------------------------------------------------------------------------------------
2023-10-17 10:37:03,894 Final evaluation on model from best epoch (best-model.pt)
2023-10-17 10:37:03,894 - metric: "('micro avg', 'f1-score')"
2023-10-17 10:37:03,894 ----------------------------------------------------------------------------------------------------
2023-10-17 10:37:03,894 Computation:
2023-10-17 10:37:03,894 - compute on device: cuda:0
2023-10-17 10:37:03,894 - embedding storage: none
2023-10-17 10:37:03,895 ----------------------------------------------------------------------------------------------------
2023-10-17 10:37:03,895 Model training base path: "hmbench-ajmc/fr-hmteams/teams-base-historic-multilingual-discriminator-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-2"
2023-10-17 10:37:03,895 ----------------------------------------------------------------------------------------------------
2023-10-17 10:37:03,895 ----------------------------------------------------------------------------------------------------
2023-10-17 10:37:03,895 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-17 10:37:05,003 epoch 1 - iter 24/242 - loss 4.26705654 - time (sec): 1.11 - samples/sec: 2025.41 - lr: 0.000005 - momentum: 0.000000
2023-10-17 10:37:06,141 epoch 1 - iter 48/242 - loss 3.16252160 - time (sec): 2.24 - samples/sec: 2080.28 - lr: 0.000010 - momentum: 0.000000
2023-10-17 10:37:07,354 epoch 1 - iter 72/242 - loss 2.25262588 - time (sec): 3.46 - samples/sec: 2122.50 - lr: 0.000015 - momentum: 0.000000
2023-10-17 10:37:08,523 epoch 1 - iter 96/242 - loss 1.85353559 - time (sec): 4.63 - samples/sec: 2089.09 - lr: 0.000020 - momentum: 0.000000
2023-10-17 10:37:09,707 epoch 1 - iter 120/242 - loss 1.57271846 - time (sec): 5.81 - samples/sec: 2110.01 - lr: 0.000025 - momentum: 0.000000
2023-10-17 10:37:10,835 epoch 1 - iter 144/242 - loss 1.36870228 - time (sec): 6.94 - samples/sec: 2106.44 - lr: 0.000030 - momentum: 0.000000
2023-10-17 10:37:11,903 epoch 1 - iter 168/242 - loss 1.19860174 - time (sec): 8.01 - samples/sec: 2160.94 - lr: 0.000035 - momentum: 0.000000
2023-10-17 10:37:12,972 epoch 1 - iter 192/242 - loss 1.10814173 - time (sec): 9.08 - samples/sec: 2145.29 - lr: 0.000039 - momentum: 0.000000
2023-10-17 10:37:14,094 epoch 1 - iter 216/242 - loss 1.00786775 - time (sec): 10.20 - samples/sec: 2169.34 - lr: 0.000044 - momentum: 0.000000
2023-10-17 10:37:15,202 epoch 1 - iter 240/242 - loss 0.92435467 - time (sec): 11.31 - samples/sec: 2177.19 - lr: 0.000049 - momentum: 0.000000
2023-10-17 10:37:15,302 ----------------------------------------------------------------------------------------------------
2023-10-17 10:37:15,302 EPOCH 1 done: loss 0.9204 - lr: 0.000049
2023-10-17 10:37:15,925 DEV : loss 0.15383611619472504 - f1-score (micro avg) 0.7509
2023-10-17 10:37:15,930 saving best model
2023-10-17 10:37:16,417 ----------------------------------------------------------------------------------------------------
2023-10-17 10:37:17,563 epoch 2 - iter 24/242 - loss 0.22040646 - time (sec): 1.14 - samples/sec: 2030.45 - lr: 0.000049 - momentum: 0.000000
2023-10-17 10:37:18,712 epoch 2 - iter 48/242 - loss 0.22115689 - time (sec): 2.29 - samples/sec: 1951.21 - lr: 0.000049 - momentum: 0.000000
2023-10-17 10:37:19,781 epoch 2 - iter 72/242 - loss 0.21889785 - time (sec): 3.36 - samples/sec: 2071.37 - lr: 0.000048 - momentum: 0.000000
2023-10-17 10:37:20,813 epoch 2 - iter 96/242 - loss 0.19591235 - time (sec): 4.39 - samples/sec: 2069.58 - lr: 0.000048 - momentum: 0.000000
2023-10-17 10:37:21,879 epoch 2 - iter 120/242 - loss 0.19387319 - time (sec): 5.46 - samples/sec: 2115.04 - lr: 0.000047 - momentum: 0.000000
2023-10-17 10:37:22,974 epoch 2 - iter 144/242 - loss 0.18492007 - time (sec): 6.56 - samples/sec: 2160.77 - lr: 0.000047 - momentum: 0.000000
2023-10-17 10:37:24,027 epoch 2 - iter 168/242 - loss 0.18355651 - time (sec): 7.61 - samples/sec: 2191.45 - lr: 0.000046 - momentum: 0.000000
2023-10-17 10:37:25,087 epoch 2 - iter 192/242 - loss 0.17936022 - time (sec): 8.67 - samples/sec: 2224.48 - lr: 0.000046 - momentum: 0.000000
2023-10-17 10:37:26,178 epoch 2 - iter 216/242 - loss 0.17205080 - time (sec): 9.76 - samples/sec: 2233.69 - lr: 0.000045 - momentum: 0.000000
2023-10-17 10:37:27,277 epoch 2 - iter 240/242 - loss 0.16709350 - time (sec): 10.86 - samples/sec: 2254.65 - lr: 0.000045 - momentum: 0.000000
2023-10-17 10:37:27,382 ----------------------------------------------------------------------------------------------------
2023-10-17 10:37:27,382 EPOCH 2 done: loss 0.1659 - lr: 0.000045
2023-10-17 10:37:28,340 DEV : loss 0.16854800283908844 - f1-score (micro avg) 0.8193
2023-10-17 10:37:28,346 saving best model
2023-10-17 10:37:28,899 ----------------------------------------------------------------------------------------------------
2023-10-17 10:37:30,078 epoch 3 - iter 24/242 - loss 0.08547562 - time (sec): 1.18 - samples/sec: 2200.96 - lr: 0.000044 - momentum: 0.000000
2023-10-17 10:37:31,219 epoch 3 - iter 48/242 - loss 0.09930094 - time (sec): 2.32 - samples/sec: 2256.21 - lr: 0.000043 - momentum: 0.000000
2023-10-17 10:37:32,346 epoch 3 - iter 72/242 - loss 0.09997574 - time (sec): 3.44 - samples/sec: 2271.77 - lr: 0.000043 - momentum: 0.000000
2023-10-17 10:37:33,441 epoch 3 - iter 96/242 - loss 0.09649896 - time (sec): 4.54 - samples/sec: 2195.98 - lr: 0.000042 - momentum: 0.000000
2023-10-17 10:37:34,531 epoch 3 - iter 120/242 - loss 0.09690468 - time (sec): 5.63 - samples/sec: 2195.49 - lr: 0.000042 - momentum: 0.000000
2023-10-17 10:37:35,660 epoch 3 - iter 144/242 - loss 0.10360914 - time (sec): 6.76 - samples/sec: 2190.53 - lr: 0.000041 - momentum: 0.000000
2023-10-17 10:37:36,760 epoch 3 - iter 168/242 - loss 0.10467477 - time (sec): 7.86 - samples/sec: 2211.36 - lr: 0.000041 - momentum: 0.000000
2023-10-17 10:37:37,869 epoch 3 - iter 192/242 - loss 0.10251829 - time (sec): 8.97 - samples/sec: 2173.48 - lr: 0.000040 - momentum: 0.000000
2023-10-17 10:37:38,926 epoch 3 - iter 216/242 - loss 0.10081840 - time (sec): 10.02 - samples/sec: 2191.27 - lr: 0.000040 - momentum: 0.000000
2023-10-17 10:37:40,016 epoch 3 - iter 240/242 - loss 0.09893098 - time (sec): 11.11 - samples/sec: 2213.87 - lr: 0.000039 - momentum: 0.000000
2023-10-17 10:37:40,101 ----------------------------------------------------------------------------------------------------
2023-10-17 10:37:40,102 EPOCH 3 done: loss 0.0984 - lr: 0.000039
2023-10-17 10:37:40,878 DEV : loss 0.16904763877391815 - f1-score (micro avg) 0.8005
2023-10-17 10:37:40,883 ----------------------------------------------------------------------------------------------------
2023-10-17 10:37:41,991 epoch 4 - iter 24/242 - loss 0.06905514 - time (sec): 1.11 - samples/sec: 2155.26 - lr: 0.000038 - momentum: 0.000000
2023-10-17 10:37:43,099 epoch 4 - iter 48/242 - loss 0.07863967 - time (sec): 2.21 - samples/sec: 2147.25 - lr: 0.000038 - momentum: 0.000000
2023-10-17 10:37:44,197 epoch 4 - iter 72/242 - loss 0.07484439 - time (sec): 3.31 - samples/sec: 2064.99 - lr: 0.000037 - momentum: 0.000000
2023-10-17 10:37:45,293 epoch 4 - iter 96/242 - loss 0.07701269 - time (sec): 4.41 - samples/sec: 2066.82 - lr: 0.000037 - momentum: 0.000000
2023-10-17 10:37:46,405 epoch 4 - iter 120/242 - loss 0.07718673 - time (sec): 5.52 - samples/sec: 2118.94 - lr: 0.000036 - momentum: 0.000000
2023-10-17 10:37:47,558 epoch 4 - iter 144/242 - loss 0.07126967 - time (sec): 6.67 - samples/sec: 2189.42 - lr: 0.000036 - momentum: 0.000000
2023-10-17 10:37:48,733 epoch 4 - iter 168/242 - loss 0.07234945 - time (sec): 7.85 - samples/sec: 2181.02 - lr: 0.000035 - momentum: 0.000000
2023-10-17 10:37:49,932 epoch 4 - iter 192/242 - loss 0.07556880 - time (sec): 9.05 - samples/sec: 2191.41 - lr: 0.000035 - momentum: 0.000000
2023-10-17 10:37:51,087 epoch 4 - iter 216/242 - loss 0.07433665 - time (sec): 10.20 - samples/sec: 2183.17 - lr: 0.000034 - momentum: 0.000000
2023-10-17 10:37:52,225 epoch 4 - iter 240/242 - loss 0.07607920 - time (sec): 11.34 - samples/sec: 2168.57 - lr: 0.000033 - momentum: 0.000000
2023-10-17 10:37:52,316 ----------------------------------------------------------------------------------------------------
2023-10-17 10:37:52,317 EPOCH 4 done: loss 0.0764 - lr: 0.000033
2023-10-17 10:37:53,112 DEV : loss 0.18921326100826263 - f1-score (micro avg) 0.8275
2023-10-17 10:37:53,118 saving best model
2023-10-17 10:37:53,641 ----------------------------------------------------------------------------------------------------
2023-10-17 10:37:54,788 epoch 5 - iter 24/242 - loss 0.04703834 - time (sec): 1.14 - samples/sec: 2179.92 - lr: 0.000033 - momentum: 0.000000
2023-10-17 10:37:55,915 epoch 5 - iter 48/242 - loss 0.05396331 - time (sec): 2.27 - samples/sec: 2272.55 - lr: 0.000032 - momentum: 0.000000
2023-10-17 10:37:57,003 epoch 5 - iter 72/242 - loss 0.04618997 - time (sec): 3.36 - samples/sec: 2310.46 - lr: 0.000032 - momentum: 0.000000
2023-10-17 10:37:58,089 epoch 5 - iter 96/242 - loss 0.04325612 - time (sec): 4.44 - samples/sec: 2344.72 - lr: 0.000031 - momentum: 0.000000
2023-10-17 10:37:59,230 epoch 5 - iter 120/242 - loss 0.04315048 - time (sec): 5.58 - samples/sec: 2293.40 - lr: 0.000031 - momentum: 0.000000
2023-10-17 10:38:00,379 epoch 5 - iter 144/242 - loss 0.04749375 - time (sec): 6.73 - samples/sec: 2282.81 - lr: 0.000030 - momentum: 0.000000
2023-10-17 10:38:01,521 epoch 5 - iter 168/242 - loss 0.04617460 - time (sec): 7.87 - samples/sec: 2239.02 - lr: 0.000030 - momentum: 0.000000
2023-10-17 10:38:02,645 epoch 5 - iter 192/242 - loss 0.04659794 - time (sec): 9.00 - samples/sec: 2218.50 - lr: 0.000029 - momentum: 0.000000
2023-10-17 10:38:03,752 epoch 5 - iter 216/242 - loss 0.04654054 - time (sec): 10.11 - samples/sec: 2212.46 - lr: 0.000028 - momentum: 0.000000
2023-10-17 10:38:04,898 epoch 5 - iter 240/242 - loss 0.04692163 - time (sec): 11.25 - samples/sec: 2191.72 - lr: 0.000028 - momentum: 0.000000
2023-10-17 10:38:04,987 ----------------------------------------------------------------------------------------------------
2023-10-17 10:38:04,988 EPOCH 5 done: loss 0.0468 - lr: 0.000028
2023-10-17 10:38:05,786 DEV : loss 0.22835828363895416 - f1-score (micro avg) 0.8076
2023-10-17 10:38:05,793 ----------------------------------------------------------------------------------------------------
2023-10-17 10:38:06,901 epoch 6 - iter 24/242 - loss 0.04790088 - time (sec): 1.11 - samples/sec: 2161.79 - lr: 0.000027 - momentum: 0.000000
2023-10-17 10:38:07,986 epoch 6 - iter 48/242 - loss 0.04316168 - time (sec): 2.19 - samples/sec: 2078.71 - lr: 0.000027 - momentum: 0.000000
2023-10-17 10:38:09,081 epoch 6 - iter 72/242 - loss 0.03759371 - time (sec): 3.29 - samples/sec: 2177.47 - lr: 0.000026 - momentum: 0.000000
2023-10-17 10:38:10,177 epoch 6 - iter 96/242 - loss 0.04423285 - time (sec): 4.38 - samples/sec: 2219.50 - lr: 0.000026 - momentum: 0.000000
2023-10-17 10:38:11,296 epoch 6 - iter 120/242 - loss 0.04092062 - time (sec): 5.50 - samples/sec: 2235.26 - lr: 0.000025 - momentum: 0.000000
2023-10-17 10:38:12,403 epoch 6 - iter 144/242 - loss 0.03873800 - time (sec): 6.61 - samples/sec: 2233.80 - lr: 0.000025 - momentum: 0.000000
2023-10-17 10:38:13,481 epoch 6 - iter 168/242 - loss 0.03907001 - time (sec): 7.69 - samples/sec: 2221.49 - lr: 0.000024 - momentum: 0.000000
2023-10-17 10:38:14,577 epoch 6 - iter 192/242 - loss 0.03692244 - time (sec): 8.78 - samples/sec: 2230.85 - lr: 0.000023 - momentum: 0.000000
2023-10-17 10:38:15,691 epoch 6 - iter 216/242 - loss 0.03678498 - time (sec): 9.90 - samples/sec: 2229.90 - lr: 0.000023 - momentum: 0.000000
2023-10-17 10:38:16,804 epoch 6 - iter 240/242 - loss 0.03565390 - time (sec): 11.01 - samples/sec: 2238.28 - lr: 0.000022 - momentum: 0.000000
2023-10-17 10:38:16,885 ----------------------------------------------------------------------------------------------------
2023-10-17 10:38:16,885 EPOCH 6 done: loss 0.0355 - lr: 0.000022
2023-10-17 10:38:17,717 DEV : loss 0.2321670949459076 - f1-score (micro avg) 0.8399
2023-10-17 10:38:17,723 saving best model
2023-10-17 10:38:18,240 ----------------------------------------------------------------------------------------------------
2023-10-17 10:38:19,366 epoch 7 - iter 24/242 - loss 0.01134798 - time (sec): 1.12 - samples/sec: 2068.64 - lr: 0.000022 - momentum: 0.000000
2023-10-17 10:38:20,535 epoch 7 - iter 48/242 - loss 0.00754801 - time (sec): 2.29 - samples/sec: 2181.77 - lr: 0.000021 - momentum: 0.000000
2023-10-17 10:38:21,653 epoch 7 - iter 72/242 - loss 0.01300960 - time (sec): 3.41 - samples/sec: 2102.00 - lr: 0.000021 - momentum: 0.000000
2023-10-17 10:38:22,795 epoch 7 - iter 96/242 - loss 0.01854904 - time (sec): 4.55 - samples/sec: 2125.98 - lr: 0.000020 - momentum: 0.000000
2023-10-17 10:38:23,917 epoch 7 - iter 120/242 - loss 0.01962791 - time (sec): 5.67 - samples/sec: 2167.16 - lr: 0.000020 - momentum: 0.000000
2023-10-17 10:38:25,020 epoch 7 - iter 144/242 - loss 0.01789323 - time (sec): 6.78 - samples/sec: 2190.67 - lr: 0.000019 - momentum: 0.000000
2023-10-17 10:38:26,159 epoch 7 - iter 168/242 - loss 0.01651922 - time (sec): 7.92 - samples/sec: 2192.11 - lr: 0.000018 - momentum: 0.000000
2023-10-17 10:38:27,244 epoch 7 - iter 192/242 - loss 0.01724750 - time (sec): 9.00 - samples/sec: 2185.00 - lr: 0.000018 - momentum: 0.000000
2023-10-17 10:38:28,365 epoch 7 - iter 216/242 - loss 0.01826242 - time (sec): 10.12 - samples/sec: 2200.14 - lr: 0.000017 - momentum: 0.000000
2023-10-17 10:38:29,481 epoch 7 - iter 240/242 - loss 0.01823347 - time (sec): 11.24 - samples/sec: 2189.17 - lr: 0.000017 - momentum: 0.000000
2023-10-17 10:38:29,567 ----------------------------------------------------------------------------------------------------
2023-10-17 10:38:29,567 EPOCH 7 done: loss 0.0181 - lr: 0.000017
2023-10-17 10:38:30,339 DEV : loss 0.22075609862804413 - f1-score (micro avg) 0.8407
2023-10-17 10:38:30,345 saving best model
2023-10-17 10:38:30,911 ----------------------------------------------------------------------------------------------------
2023-10-17 10:38:32,025 epoch 8 - iter 24/242 - loss 0.00594396 - time (sec): 1.11 - samples/sec: 2332.45 - lr: 0.000016 - momentum: 0.000000
2023-10-17 10:38:33,141 epoch 8 - iter 48/242 - loss 0.00863514 - time (sec): 2.23 - samples/sec: 2228.01 - lr: 0.000016 - momentum: 0.000000
2023-10-17 10:38:34,207 epoch 8 - iter 72/242 - loss 0.00735051 - time (sec): 3.29 - samples/sec: 2262.44 - lr: 0.000015 - momentum: 0.000000
2023-10-17 10:38:35,316 epoch 8 - iter 96/242 - loss 0.00590879 - time (sec): 4.40 - samples/sec: 2282.64 - lr: 0.000015 - momentum: 0.000000
2023-10-17 10:38:36,404 epoch 8 - iter 120/242 - loss 0.01160036 - time (sec): 5.49 - samples/sec: 2272.80 - lr: 0.000014 - momentum: 0.000000
2023-10-17 10:38:37,558 epoch 8 - iter 144/242 - loss 0.01229182 - time (sec): 6.64 - samples/sec: 2261.48 - lr: 0.000013 - momentum: 0.000000
2023-10-17 10:38:38,654 epoch 8 - iter 168/242 - loss 0.01536004 - time (sec): 7.74 - samples/sec: 2256.47 - lr: 0.000013 - momentum: 0.000000
2023-10-17 10:38:39,755 epoch 8 - iter 192/242 - loss 0.01414688 - time (sec): 8.84 - samples/sec: 2245.15 - lr: 0.000012 - momentum: 0.000000
2023-10-17 10:38:40,972 epoch 8 - iter 216/242 - loss 0.01338241 - time (sec): 10.06 - samples/sec: 2210.19 - lr: 0.000012 - momentum: 0.000000
2023-10-17 10:38:42,099 epoch 8 - iter 240/242 - loss 0.01417627 - time (sec): 11.18 - samples/sec: 2199.76 - lr: 0.000011 - momentum: 0.000000
2023-10-17 10:38:42,181 ----------------------------------------------------------------------------------------------------
2023-10-17 10:38:42,181 EPOCH 8 done: loss 0.0141 - lr: 0.000011
2023-10-17 10:38:42,998 DEV : loss 0.2549717128276825 - f1-score (micro avg) 0.8338
2023-10-17 10:38:43,005 ----------------------------------------------------------------------------------------------------
2023-10-17 10:38:44,164 epoch 9 - iter 24/242 - loss 0.01687544 - time (sec): 1.16 - samples/sec: 2245.85 - lr: 0.000011 - momentum: 0.000000
2023-10-17 10:38:45,246 epoch 9 - iter 48/242 - loss 0.01053392 - time (sec): 2.24 - samples/sec: 2210.65 - lr: 0.000010 - momentum: 0.000000
2023-10-17 10:38:46,346 epoch 9 - iter 72/242 - loss 0.00966376 - time (sec): 3.34 - samples/sec: 2136.57 - lr: 0.000010 - momentum: 0.000000
2023-10-17 10:38:47,519 epoch 9 - iter 96/242 - loss 0.01454358 - time (sec): 4.51 - samples/sec: 2128.66 - lr: 0.000009 - momentum: 0.000000
2023-10-17 10:38:48,733 epoch 9 - iter 120/242 - loss 0.01171230 - time (sec): 5.73 - samples/sec: 2112.89 - lr: 0.000008 - momentum: 0.000000
2023-10-17 10:38:49,926 epoch 9 - iter 144/242 - loss 0.01031532 - time (sec): 6.92 - samples/sec: 2126.99 - lr: 0.000008 - momentum: 0.000000
2023-10-17 10:38:51,116 epoch 9 - iter 168/242 - loss 0.00962467 - time (sec): 8.11 - samples/sec: 2099.93 - lr: 0.000007 - momentum: 0.000000
2023-10-17 10:38:52,307 epoch 9 - iter 192/242 - loss 0.01038682 - time (sec): 9.30 - samples/sec: 2099.75 - lr: 0.000007 - momentum: 0.000000
2023-10-17 10:38:53,449 epoch 9 - iter 216/242 - loss 0.00980807 - time (sec): 10.44 - samples/sec: 2109.84 - lr: 0.000006 - momentum: 0.000000
2023-10-17 10:38:54,583 epoch 9 - iter 240/242 - loss 0.00882063 - time (sec): 11.58 - samples/sec: 2125.53 - lr: 0.000006 - momentum: 0.000000
2023-10-17 10:38:54,687 ----------------------------------------------------------------------------------------------------
2023-10-17 10:38:54,687 EPOCH 9 done: loss 0.0088 - lr: 0.000006
2023-10-17 10:38:55,451 DEV : loss 0.24903777241706848 - f1-score (micro avg) 0.8489
2023-10-17 10:38:55,456 saving best model
2023-10-17 10:38:55,997 ----------------------------------------------------------------------------------------------------
2023-10-17 10:38:57,204 epoch 10 - iter 24/242 - loss 0.02172065 - time (sec): 1.20 - samples/sec: 2019.71 - lr: 0.000005 - momentum: 0.000000
2023-10-17 10:38:58,442 epoch 10 - iter 48/242 - loss 0.01225404 - time (sec): 2.44 - samples/sec: 1994.99 - lr: 0.000005 - momentum: 0.000000
2023-10-17 10:38:59,531 epoch 10 - iter 72/242 - loss 0.00903661 - time (sec): 3.53 - samples/sec: 2027.85 - lr: 0.000004 - momentum: 0.000000
2023-10-17 10:39:00,637 epoch 10 - iter 96/242 - loss 0.00663820 - time (sec): 4.63 - samples/sec: 2116.32 - lr: 0.000003 - momentum: 0.000000
2023-10-17 10:39:01,758 epoch 10 - iter 120/242 - loss 0.00654181 - time (sec): 5.76 - samples/sec: 2151.73 - lr: 0.000003 - momentum: 0.000000
2023-10-17 10:39:02,894 epoch 10 - iter 144/242 - loss 0.00618297 - time (sec): 6.89 - samples/sec: 2131.46 - lr: 0.000002 - momentum: 0.000000
2023-10-17 10:39:04,036 epoch 10 - iter 168/242 - loss 0.00694130 - time (sec): 8.03 - samples/sec: 2139.22 - lr: 0.000002 - momentum: 0.000000
2023-10-17 10:39:05,171 epoch 10 - iter 192/242 - loss 0.00610222 - time (sec): 9.17 - samples/sec: 2136.28 - lr: 0.000001 - momentum: 0.000000
2023-10-17 10:39:06,368 epoch 10 - iter 216/242 - loss 0.00558065 - time (sec): 10.37 - samples/sec: 2126.89 - lr: 0.000001 - momentum: 0.000000
2023-10-17 10:39:07,516 epoch 10 - iter 240/242 - loss 0.00568743 - time (sec): 11.51 - samples/sec: 2136.26 - lr: 0.000000 - momentum: 0.000000
2023-10-17 10:39:07,602 ----------------------------------------------------------------------------------------------------
2023-10-17 10:39:07,603 EPOCH 10 done: loss 0.0057 - lr: 0.000000
2023-10-17 10:39:08,413 DEV : loss 0.25036904215812683 - f1-score (micro avg) 0.8564
2023-10-17 10:39:08,419 saving best model
2023-10-17 10:39:09,372 ----------------------------------------------------------------------------------------------------
2023-10-17 10:39:09,374 Loading model from best epoch ...
2023-10-17 10:39:10,807 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-object, B-object, E-object, I-object, S-date, B-date, E-date, I-date
2023-10-17 10:39:11,766
Results:
- F-score (micro) 0.8338
- F-score (macro) 0.5576
- Accuracy 0.7356
By class:
precision recall f1-score support
pers 0.8944 0.9137 0.9039 139
scope 0.8321 0.8837 0.8571 129
work 0.7000 0.7875 0.7412 80
loc 0.4000 0.2222 0.2857 9
date 0.0000 0.0000 0.0000 3
micro avg 0.8182 0.8500 0.8338 360
macro avg 0.5653 0.5614 0.5576 360
weighted avg 0.8091 0.8500 0.8280 360
2023-10-17 10:39:11,766 ----------------------------------------------------------------------------------------------------
|