File size: 23,827 Bytes
128a210 |
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 |
2023-10-13 10:30:09,192 ----------------------------------------------------------------------------------------------------
2023-10-13 10:30:09,193 Model: "SequenceTagger(
(embeddings): TransformerWordEmbeddings(
(model): BertModel(
(embeddings): BertEmbeddings(
(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): BertEncoder(
(layer): ModuleList(
(0-11): 12 x BertLayer(
(attention): BertAttention(
(self): BertSelfAttention(
(query): Linear(in_features=768, out_features=768, bias=True)
(key): Linear(in_features=768, out_features=768, bias=True)
(value): Linear(in_features=768, out_features=768, bias=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(output): BertSelfOutput(
(dense): Linear(in_features=768, out_features=768, bias=True)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
(intermediate): BertIntermediate(
(dense): Linear(in_features=768, out_features=3072, bias=True)
(intermediate_act_fn): GELUActivation()
)
(output): BertOutput(
(dense): Linear(in_features=3072, out_features=768, bias=True)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
)
(pooler): BertPooler(
(dense): Linear(in_features=768, out_features=768, bias=True)
(activation): Tanh()
)
)
)
(locked_dropout): LockedDropout(p=0.5)
(linear): Linear(in_features=768, out_features=25, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-13 10:30:09,193 ----------------------------------------------------------------------------------------------------
2023-10-13 10:30:09,193 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
2023-10-13 10:30:09,193 ----------------------------------------------------------------------------------------------------
2023-10-13 10:30:09,193 Train: 966 sentences
2023-10-13 10:30:09,193 (train_with_dev=False, train_with_test=False)
2023-10-13 10:30:09,193 ----------------------------------------------------------------------------------------------------
2023-10-13 10:30:09,193 Training Params:
2023-10-13 10:30:09,193 - learning_rate: "3e-05"
2023-10-13 10:30:09,193 - mini_batch_size: "4"
2023-10-13 10:30:09,193 - max_epochs: "10"
2023-10-13 10:30:09,193 - shuffle: "True"
2023-10-13 10:30:09,193 ----------------------------------------------------------------------------------------------------
2023-10-13 10:30:09,193 Plugins:
2023-10-13 10:30:09,193 - LinearScheduler | warmup_fraction: '0.1'
2023-10-13 10:30:09,193 ----------------------------------------------------------------------------------------------------
2023-10-13 10:30:09,193 Final evaluation on model from best epoch (best-model.pt)
2023-10-13 10:30:09,193 - metric: "('micro avg', 'f1-score')"
2023-10-13 10:30:09,193 ----------------------------------------------------------------------------------------------------
2023-10-13 10:30:09,194 Computation:
2023-10-13 10:30:09,194 - compute on device: cuda:0
2023-10-13 10:30:09,194 - embedding storage: none
2023-10-13 10:30:09,194 ----------------------------------------------------------------------------------------------------
2023-10-13 10:30:09,194 Model training base path: "hmbench-ajmc/fr-dbmdz/bert-base-historic-multilingual-cased-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-1"
2023-10-13 10:30:09,194 ----------------------------------------------------------------------------------------------------
2023-10-13 10:30:09,194 ----------------------------------------------------------------------------------------------------
2023-10-13 10:30:11,316 epoch 1 - iter 24/242 - loss 3.31899833 - time (sec): 2.12 - samples/sec: 1099.73 - lr: 0.000003 - momentum: 0.000000
2023-10-13 10:30:12,625 epoch 1 - iter 48/242 - loss 2.96790424 - time (sec): 3.43 - samples/sec: 1462.00 - lr: 0.000006 - momentum: 0.000000
2023-10-13 10:30:13,935 epoch 1 - iter 72/242 - loss 2.37819050 - time (sec): 4.74 - samples/sec: 1614.82 - lr: 0.000009 - momentum: 0.000000
2023-10-13 10:30:15,194 epoch 1 - iter 96/242 - loss 1.96624830 - time (sec): 6.00 - samples/sec: 1677.13 - lr: 0.000012 - momentum: 0.000000
2023-10-13 10:30:16,452 epoch 1 - iter 120/242 - loss 1.73643668 - time (sec): 7.26 - samples/sec: 1683.88 - lr: 0.000015 - momentum: 0.000000
2023-10-13 10:30:17,743 epoch 1 - iter 144/242 - loss 1.51974639 - time (sec): 8.55 - samples/sec: 1739.01 - lr: 0.000018 - momentum: 0.000000
2023-10-13 10:30:19,013 epoch 1 - iter 168/242 - loss 1.36954910 - time (sec): 9.82 - samples/sec: 1761.71 - lr: 0.000021 - momentum: 0.000000
2023-10-13 10:30:20,238 epoch 1 - iter 192/242 - loss 1.24200364 - time (sec): 11.04 - samples/sec: 1801.53 - lr: 0.000024 - momentum: 0.000000
2023-10-13 10:30:21,443 epoch 1 - iter 216/242 - loss 1.14147406 - time (sec): 12.25 - samples/sec: 1811.60 - lr: 0.000027 - momentum: 0.000000
2023-10-13 10:30:22,713 epoch 1 - iter 240/242 - loss 1.05944884 - time (sec): 13.52 - samples/sec: 1816.46 - lr: 0.000030 - momentum: 0.000000
2023-10-13 10:30:22,834 ----------------------------------------------------------------------------------------------------
2023-10-13 10:30:22,834 EPOCH 1 done: loss 1.0537 - lr: 0.000030
2023-10-13 10:30:23,525 DEV : loss 0.27631089091300964 - f1-score (micro avg) 0.457
2023-10-13 10:30:23,530 saving best model
2023-10-13 10:30:23,924 ----------------------------------------------------------------------------------------------------
2023-10-13 10:30:25,134 epoch 2 - iter 24/242 - loss 0.25291019 - time (sec): 1.21 - samples/sec: 2019.76 - lr: 0.000030 - momentum: 0.000000
2023-10-13 10:30:26,315 epoch 2 - iter 48/242 - loss 0.28868353 - time (sec): 2.39 - samples/sec: 1996.82 - lr: 0.000029 - momentum: 0.000000
2023-10-13 10:30:27,555 epoch 2 - iter 72/242 - loss 0.26832173 - time (sec): 3.63 - samples/sec: 2035.03 - lr: 0.000029 - momentum: 0.000000
2023-10-13 10:30:28,799 epoch 2 - iter 96/242 - loss 0.25056380 - time (sec): 4.87 - samples/sec: 2040.48 - lr: 0.000029 - momentum: 0.000000
2023-10-13 10:30:30,034 epoch 2 - iter 120/242 - loss 0.24861522 - time (sec): 6.11 - samples/sec: 2019.03 - lr: 0.000028 - momentum: 0.000000
2023-10-13 10:30:31,264 epoch 2 - iter 144/242 - loss 0.24350868 - time (sec): 7.34 - samples/sec: 1981.02 - lr: 0.000028 - momentum: 0.000000
2023-10-13 10:30:32,502 epoch 2 - iter 168/242 - loss 0.23234542 - time (sec): 8.58 - samples/sec: 1988.14 - lr: 0.000028 - momentum: 0.000000
2023-10-13 10:30:33,737 epoch 2 - iter 192/242 - loss 0.23022383 - time (sec): 9.81 - samples/sec: 1999.91 - lr: 0.000027 - momentum: 0.000000
2023-10-13 10:30:34,936 epoch 2 - iter 216/242 - loss 0.21995150 - time (sec): 11.01 - samples/sec: 2043.60 - lr: 0.000027 - momentum: 0.000000
2023-10-13 10:30:36,034 epoch 2 - iter 240/242 - loss 0.21735290 - time (sec): 12.11 - samples/sec: 2034.20 - lr: 0.000027 - momentum: 0.000000
2023-10-13 10:30:36,117 ----------------------------------------------------------------------------------------------------
2023-10-13 10:30:36,117 EPOCH 2 done: loss 0.2168 - lr: 0.000027
2023-10-13 10:30:36,917 DEV : loss 0.14946463704109192 - f1-score (micro avg) 0.7621
2023-10-13 10:30:36,924 saving best model
2023-10-13 10:30:37,446 ----------------------------------------------------------------------------------------------------
2023-10-13 10:30:38,618 epoch 3 - iter 24/242 - loss 0.18291948 - time (sec): 1.17 - samples/sec: 2092.17 - lr: 0.000026 - momentum: 0.000000
2023-10-13 10:30:39,983 epoch 3 - iter 48/242 - loss 0.15886607 - time (sec): 2.53 - samples/sec: 1968.78 - lr: 0.000026 - momentum: 0.000000
2023-10-13 10:30:41,202 epoch 3 - iter 72/242 - loss 0.13699532 - time (sec): 3.75 - samples/sec: 1971.01 - lr: 0.000026 - momentum: 0.000000
2023-10-13 10:30:42,261 epoch 3 - iter 96/242 - loss 0.13323718 - time (sec): 4.81 - samples/sec: 2019.19 - lr: 0.000025 - momentum: 0.000000
2023-10-13 10:30:43,378 epoch 3 - iter 120/242 - loss 0.12809419 - time (sec): 5.93 - samples/sec: 2056.47 - lr: 0.000025 - momentum: 0.000000
2023-10-13 10:30:44,545 epoch 3 - iter 144/242 - loss 0.11962736 - time (sec): 7.10 - samples/sec: 2068.28 - lr: 0.000025 - momentum: 0.000000
2023-10-13 10:30:45,597 epoch 3 - iter 168/242 - loss 0.12663136 - time (sec): 8.15 - samples/sec: 2098.46 - lr: 0.000024 - momentum: 0.000000
2023-10-13 10:30:46,739 epoch 3 - iter 192/242 - loss 0.12128219 - time (sec): 9.29 - samples/sec: 2118.70 - lr: 0.000024 - momentum: 0.000000
2023-10-13 10:30:47,876 epoch 3 - iter 216/242 - loss 0.12074200 - time (sec): 10.43 - samples/sec: 2126.21 - lr: 0.000024 - momentum: 0.000000
2023-10-13 10:30:49,007 epoch 3 - iter 240/242 - loss 0.12123157 - time (sec): 11.56 - samples/sec: 2127.78 - lr: 0.000023 - momentum: 0.000000
2023-10-13 10:30:49,096 ----------------------------------------------------------------------------------------------------
2023-10-13 10:30:49,097 EPOCH 3 done: loss 0.1219 - lr: 0.000023
2023-10-13 10:30:49,922 DEV : loss 0.1250181347131729 - f1-score (micro avg) 0.8202
2023-10-13 10:30:49,927 saving best model
2023-10-13 10:30:50,429 ----------------------------------------------------------------------------------------------------
2023-10-13 10:30:51,583 epoch 4 - iter 24/242 - loss 0.12817993 - time (sec): 1.15 - samples/sec: 2190.49 - lr: 0.000023 - momentum: 0.000000
2023-10-13 10:30:52,701 epoch 4 - iter 48/242 - loss 0.09658017 - time (sec): 2.27 - samples/sec: 2274.08 - lr: 0.000023 - momentum: 0.000000
2023-10-13 10:30:53,861 epoch 4 - iter 72/242 - loss 0.08645078 - time (sec): 3.43 - samples/sec: 2234.05 - lr: 0.000022 - momentum: 0.000000
2023-10-13 10:30:55,032 epoch 4 - iter 96/242 - loss 0.08823514 - time (sec): 4.60 - samples/sec: 2122.16 - lr: 0.000022 - momentum: 0.000000
2023-10-13 10:30:56,153 epoch 4 - iter 120/242 - loss 0.08030354 - time (sec): 5.72 - samples/sec: 2190.18 - lr: 0.000022 - momentum: 0.000000
2023-10-13 10:30:57,271 epoch 4 - iter 144/242 - loss 0.08186884 - time (sec): 6.84 - samples/sec: 2181.87 - lr: 0.000021 - momentum: 0.000000
2023-10-13 10:30:58,487 epoch 4 - iter 168/242 - loss 0.08565366 - time (sec): 8.06 - samples/sec: 2156.28 - lr: 0.000021 - momentum: 0.000000
2023-10-13 10:30:59,707 epoch 4 - iter 192/242 - loss 0.08516789 - time (sec): 9.28 - samples/sec: 2127.35 - lr: 0.000021 - momentum: 0.000000
2023-10-13 10:31:00,937 epoch 4 - iter 216/242 - loss 0.08260317 - time (sec): 10.51 - samples/sec: 2094.32 - lr: 0.000020 - momentum: 0.000000
2023-10-13 10:31:02,125 epoch 4 - iter 240/242 - loss 0.08329194 - time (sec): 11.69 - samples/sec: 2099.31 - lr: 0.000020 - momentum: 0.000000
2023-10-13 10:31:02,212 ----------------------------------------------------------------------------------------------------
2023-10-13 10:31:02,213 EPOCH 4 done: loss 0.0832 - lr: 0.000020
2023-10-13 10:31:03,064 DEV : loss 0.12788808345794678 - f1-score (micro avg) 0.8173
2023-10-13 10:31:03,069 ----------------------------------------------------------------------------------------------------
2023-10-13 10:31:04,301 epoch 5 - iter 24/242 - loss 0.05323062 - time (sec): 1.23 - samples/sec: 2054.39 - lr: 0.000020 - momentum: 0.000000
2023-10-13 10:31:05,453 epoch 5 - iter 48/242 - loss 0.06788494 - time (sec): 2.38 - samples/sec: 2019.33 - lr: 0.000019 - momentum: 0.000000
2023-10-13 10:31:06,610 epoch 5 - iter 72/242 - loss 0.06078541 - time (sec): 3.54 - samples/sec: 2062.05 - lr: 0.000019 - momentum: 0.000000
2023-10-13 10:31:07,755 epoch 5 - iter 96/242 - loss 0.05537332 - time (sec): 4.68 - samples/sec: 2055.60 - lr: 0.000019 - momentum: 0.000000
2023-10-13 10:31:08,869 epoch 5 - iter 120/242 - loss 0.05728335 - time (sec): 5.80 - samples/sec: 2130.10 - lr: 0.000018 - momentum: 0.000000
2023-10-13 10:31:09,944 epoch 5 - iter 144/242 - loss 0.05842797 - time (sec): 6.87 - samples/sec: 2195.35 - lr: 0.000018 - momentum: 0.000000
2023-10-13 10:31:11,013 epoch 5 - iter 168/242 - loss 0.05932681 - time (sec): 7.94 - samples/sec: 2222.93 - lr: 0.000018 - momentum: 0.000000
2023-10-13 10:31:12,102 epoch 5 - iter 192/242 - loss 0.05776849 - time (sec): 9.03 - samples/sec: 2211.23 - lr: 0.000017 - momentum: 0.000000
2023-10-13 10:31:13,144 epoch 5 - iter 216/242 - loss 0.05734916 - time (sec): 10.07 - samples/sec: 2184.78 - lr: 0.000017 - momentum: 0.000000
2023-10-13 10:31:14,209 epoch 5 - iter 240/242 - loss 0.06319448 - time (sec): 11.14 - samples/sec: 2200.20 - lr: 0.000017 - momentum: 0.000000
2023-10-13 10:31:14,297 ----------------------------------------------------------------------------------------------------
2023-10-13 10:31:14,298 EPOCH 5 done: loss 0.0632 - lr: 0.000017
2023-10-13 10:31:15,123 DEV : loss 0.14355506002902985 - f1-score (micro avg) 0.8088
2023-10-13 10:31:15,128 ----------------------------------------------------------------------------------------------------
2023-10-13 10:31:16,350 epoch 6 - iter 24/242 - loss 0.03620606 - time (sec): 1.22 - samples/sec: 2088.71 - lr: 0.000016 - momentum: 0.000000
2023-10-13 10:31:17,549 epoch 6 - iter 48/242 - loss 0.05031164 - time (sec): 2.42 - samples/sec: 1929.09 - lr: 0.000016 - momentum: 0.000000
2023-10-13 10:31:18,745 epoch 6 - iter 72/242 - loss 0.04763385 - time (sec): 3.62 - samples/sec: 1944.95 - lr: 0.000016 - momentum: 0.000000
2023-10-13 10:31:19,911 epoch 6 - iter 96/242 - loss 0.04680506 - time (sec): 4.78 - samples/sec: 1955.54 - lr: 0.000015 - momentum: 0.000000
2023-10-13 10:31:21,079 epoch 6 - iter 120/242 - loss 0.04553476 - time (sec): 5.95 - samples/sec: 2025.34 - lr: 0.000015 - momentum: 0.000000
2023-10-13 10:31:22,243 epoch 6 - iter 144/242 - loss 0.04492284 - time (sec): 7.11 - samples/sec: 2042.94 - lr: 0.000015 - momentum: 0.000000
2023-10-13 10:31:23,419 epoch 6 - iter 168/242 - loss 0.04377389 - time (sec): 8.29 - samples/sec: 2076.37 - lr: 0.000014 - momentum: 0.000000
2023-10-13 10:31:24,567 epoch 6 - iter 192/242 - loss 0.04183081 - time (sec): 9.44 - samples/sec: 2113.37 - lr: 0.000014 - momentum: 0.000000
2023-10-13 10:31:25,718 epoch 6 - iter 216/242 - loss 0.04363399 - time (sec): 10.59 - samples/sec: 2121.87 - lr: 0.000014 - momentum: 0.000000
2023-10-13 10:31:26,813 epoch 6 - iter 240/242 - loss 0.04419028 - time (sec): 11.68 - samples/sec: 2107.39 - lr: 0.000013 - momentum: 0.000000
2023-10-13 10:31:26,904 ----------------------------------------------------------------------------------------------------
2023-10-13 10:31:26,904 EPOCH 6 done: loss 0.0441 - lr: 0.000013
2023-10-13 10:31:27,724 DEV : loss 0.1649819314479828 - f1-score (micro avg) 0.8183
2023-10-13 10:31:27,729 ----------------------------------------------------------------------------------------------------
2023-10-13 10:31:28,813 epoch 7 - iter 24/242 - loss 0.03165528 - time (sec): 1.08 - samples/sec: 2042.55 - lr: 0.000013 - momentum: 0.000000
2023-10-13 10:31:29,923 epoch 7 - iter 48/242 - loss 0.02546364 - time (sec): 2.19 - samples/sec: 2105.70 - lr: 0.000013 - momentum: 0.000000
2023-10-13 10:31:31,053 epoch 7 - iter 72/242 - loss 0.02380691 - time (sec): 3.32 - samples/sec: 2162.89 - lr: 0.000012 - momentum: 0.000000
2023-10-13 10:31:32,166 epoch 7 - iter 96/242 - loss 0.02865025 - time (sec): 4.44 - samples/sec: 2162.77 - lr: 0.000012 - momentum: 0.000000
2023-10-13 10:31:33,249 epoch 7 - iter 120/242 - loss 0.03286107 - time (sec): 5.52 - samples/sec: 2184.29 - lr: 0.000012 - momentum: 0.000000
2023-10-13 10:31:34,348 epoch 7 - iter 144/242 - loss 0.03808170 - time (sec): 6.62 - samples/sec: 2216.88 - lr: 0.000011 - momentum: 0.000000
2023-10-13 10:31:35,438 epoch 7 - iter 168/242 - loss 0.03745498 - time (sec): 7.71 - samples/sec: 2247.71 - lr: 0.000011 - momentum: 0.000000
2023-10-13 10:31:36,523 epoch 7 - iter 192/242 - loss 0.03486179 - time (sec): 8.79 - samples/sec: 2252.25 - lr: 0.000011 - momentum: 0.000000
2023-10-13 10:31:37,601 epoch 7 - iter 216/242 - loss 0.03330677 - time (sec): 9.87 - samples/sec: 2232.51 - lr: 0.000010 - momentum: 0.000000
2023-10-13 10:31:38,735 epoch 7 - iter 240/242 - loss 0.03221765 - time (sec): 11.00 - samples/sec: 2242.25 - lr: 0.000010 - momentum: 0.000000
2023-10-13 10:31:38,822 ----------------------------------------------------------------------------------------------------
2023-10-13 10:31:38,822 EPOCH 7 done: loss 0.0321 - lr: 0.000010
2023-10-13 10:31:39,671 DEV : loss 0.1871682107448578 - f1-score (micro avg) 0.8069
2023-10-13 10:31:39,676 ----------------------------------------------------------------------------------------------------
2023-10-13 10:31:40,777 epoch 8 - iter 24/242 - loss 0.01700191 - time (sec): 1.10 - samples/sec: 2214.72 - lr: 0.000010 - momentum: 0.000000
2023-10-13 10:31:41,955 epoch 8 - iter 48/242 - loss 0.01712285 - time (sec): 2.28 - samples/sec: 2021.92 - lr: 0.000009 - momentum: 0.000000
2023-10-13 10:31:43,087 epoch 8 - iter 72/242 - loss 0.02033325 - time (sec): 3.41 - samples/sec: 2071.63 - lr: 0.000009 - momentum: 0.000000
2023-10-13 10:31:44,180 epoch 8 - iter 96/242 - loss 0.01811933 - time (sec): 4.50 - samples/sec: 2065.26 - lr: 0.000009 - momentum: 0.000000
2023-10-13 10:31:45,378 epoch 8 - iter 120/242 - loss 0.01662740 - time (sec): 5.70 - samples/sec: 2134.06 - lr: 0.000008 - momentum: 0.000000
2023-10-13 10:31:46,543 epoch 8 - iter 144/242 - loss 0.01520657 - time (sec): 6.86 - samples/sec: 2131.70 - lr: 0.000008 - momentum: 0.000000
2023-10-13 10:31:47,677 epoch 8 - iter 168/242 - loss 0.01531022 - time (sec): 8.00 - samples/sec: 2118.00 - lr: 0.000008 - momentum: 0.000000
2023-10-13 10:31:48,782 epoch 8 - iter 192/242 - loss 0.01806298 - time (sec): 9.10 - samples/sec: 2148.20 - lr: 0.000007 - momentum: 0.000000
2023-10-13 10:31:49,864 epoch 8 - iter 216/242 - loss 0.02135875 - time (sec): 10.19 - samples/sec: 2170.09 - lr: 0.000007 - momentum: 0.000000
2023-10-13 10:31:50,948 epoch 8 - iter 240/242 - loss 0.02328127 - time (sec): 11.27 - samples/sec: 2183.11 - lr: 0.000007 - momentum: 0.000000
2023-10-13 10:31:51,034 ----------------------------------------------------------------------------------------------------
2023-10-13 10:31:51,034 EPOCH 8 done: loss 0.0232 - lr: 0.000007
2023-10-13 10:31:51,824 DEV : loss 0.19017556309700012 - f1-score (micro avg) 0.8074
2023-10-13 10:31:51,829 ----------------------------------------------------------------------------------------------------
2023-10-13 10:31:52,940 epoch 9 - iter 24/242 - loss 0.01171644 - time (sec): 1.11 - samples/sec: 2446.12 - lr: 0.000006 - momentum: 0.000000
2023-10-13 10:31:54,081 epoch 9 - iter 48/242 - loss 0.01109408 - time (sec): 2.25 - samples/sec: 2346.81 - lr: 0.000006 - momentum: 0.000000
2023-10-13 10:31:55,183 epoch 9 - iter 72/242 - loss 0.01061616 - time (sec): 3.35 - samples/sec: 2216.51 - lr: 0.000006 - momentum: 0.000000
2023-10-13 10:31:56,296 epoch 9 - iter 96/242 - loss 0.01554713 - time (sec): 4.47 - samples/sec: 2233.37 - lr: 0.000005 - momentum: 0.000000
2023-10-13 10:31:57,371 epoch 9 - iter 120/242 - loss 0.01353043 - time (sec): 5.54 - samples/sec: 2255.15 - lr: 0.000005 - momentum: 0.000000
2023-10-13 10:31:58,482 epoch 9 - iter 144/242 - loss 0.01882666 - time (sec): 6.65 - samples/sec: 2293.73 - lr: 0.000005 - momentum: 0.000000
2023-10-13 10:31:59,560 epoch 9 - iter 168/242 - loss 0.01755098 - time (sec): 7.73 - samples/sec: 2282.37 - lr: 0.000004 - momentum: 0.000000
2023-10-13 10:32:00,720 epoch 9 - iter 192/242 - loss 0.01714456 - time (sec): 8.89 - samples/sec: 2235.67 - lr: 0.000004 - momentum: 0.000000
2023-10-13 10:32:01,808 epoch 9 - iter 216/242 - loss 0.01700689 - time (sec): 9.98 - samples/sec: 2256.60 - lr: 0.000004 - momentum: 0.000000
2023-10-13 10:32:02,879 epoch 9 - iter 240/242 - loss 0.01626621 - time (sec): 11.05 - samples/sec: 2229.19 - lr: 0.000003 - momentum: 0.000000
2023-10-13 10:32:02,965 ----------------------------------------------------------------------------------------------------
2023-10-13 10:32:02,966 EPOCH 9 done: loss 0.0162 - lr: 0.000003
2023-10-13 10:32:03,716 DEV : loss 0.2009798139333725 - f1-score (micro avg) 0.8169
2023-10-13 10:32:03,721 ----------------------------------------------------------------------------------------------------
2023-10-13 10:32:04,784 epoch 10 - iter 24/242 - loss 0.02317398 - time (sec): 1.06 - samples/sec: 2128.45 - lr: 0.000003 - momentum: 0.000000
2023-10-13 10:32:05,865 epoch 10 - iter 48/242 - loss 0.02491819 - time (sec): 2.14 - samples/sec: 2189.82 - lr: 0.000003 - momentum: 0.000000
2023-10-13 10:32:06,936 epoch 10 - iter 72/242 - loss 0.02514844 - time (sec): 3.21 - samples/sec: 2264.57 - lr: 0.000002 - momentum: 0.000000
2023-10-13 10:32:08,134 epoch 10 - iter 96/242 - loss 0.02131346 - time (sec): 4.41 - samples/sec: 2244.67 - lr: 0.000002 - momentum: 0.000000
2023-10-13 10:32:09,333 epoch 10 - iter 120/242 - loss 0.01880207 - time (sec): 5.61 - samples/sec: 2212.12 - lr: 0.000002 - momentum: 0.000000
2023-10-13 10:32:10,487 epoch 10 - iter 144/242 - loss 0.01830613 - time (sec): 6.77 - samples/sec: 2156.40 - lr: 0.000001 - momentum: 0.000000
2023-10-13 10:32:11,608 epoch 10 - iter 168/242 - loss 0.01761614 - time (sec): 7.89 - samples/sec: 2137.04 - lr: 0.000001 - momentum: 0.000000
2023-10-13 10:32:12,752 epoch 10 - iter 192/242 - loss 0.01679154 - time (sec): 9.03 - samples/sec: 2117.70 - lr: 0.000001 - momentum: 0.000000
2023-10-13 10:32:13,867 epoch 10 - iter 216/242 - loss 0.01529364 - time (sec): 10.14 - samples/sec: 2145.05 - lr: 0.000000 - momentum: 0.000000
2023-10-13 10:32:15,026 epoch 10 - iter 240/242 - loss 0.01407741 - time (sec): 11.30 - samples/sec: 2170.83 - lr: 0.000000 - momentum: 0.000000
2023-10-13 10:32:15,114 ----------------------------------------------------------------------------------------------------
2023-10-13 10:32:15,115 EPOCH 10 done: loss 0.0140 - lr: 0.000000
2023-10-13 10:32:15,938 DEV : loss 0.20342250168323517 - f1-score (micro avg) 0.8144
2023-10-13 10:32:16,329 ----------------------------------------------------------------------------------------------------
2023-10-13 10:32:16,330 Loading model from best epoch ...
2023-10-13 10:32:18,153 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
2023-10-13 10:32:19,052
Results:
- F-score (micro) 0.7666
- F-score (macro) 0.4582
- Accuracy 0.6465
By class:
precision recall f1-score support
pers 0.7468 0.8489 0.7946 139
scope 0.8028 0.8837 0.8413 129
work 0.6064 0.7125 0.6552 80
loc 0.0000 0.0000 0.0000 9
date 0.0000 0.0000 0.0000 3
micro avg 0.7335 0.8028 0.7666 360
macro avg 0.4312 0.4890 0.4582 360
weighted avg 0.7108 0.8028 0.7539 360
2023-10-13 10:32:19,052 ----------------------------------------------------------------------------------------------------
|