File size: 23,944 Bytes
e3f791b |
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 |
2023-10-13 11:07:50,255 ----------------------------------------------------------------------------------------------------
2023-10-13 11:07:50,256 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 11:07:50,256 ----------------------------------------------------------------------------------------------------
2023-10-13 11:07:50,256 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 11:07:50,257 ----------------------------------------------------------------------------------------------------
2023-10-13 11:07:50,257 Train: 966 sentences
2023-10-13 11:07:50,257 (train_with_dev=False, train_with_test=False)
2023-10-13 11:07:50,257 ----------------------------------------------------------------------------------------------------
2023-10-13 11:07:50,257 Training Params:
2023-10-13 11:07:50,257 - learning_rate: "5e-05"
2023-10-13 11:07:50,257 - mini_batch_size: "4"
2023-10-13 11:07:50,257 - max_epochs: "10"
2023-10-13 11:07:50,257 - shuffle: "True"
2023-10-13 11:07:50,257 ----------------------------------------------------------------------------------------------------
2023-10-13 11:07:50,257 Plugins:
2023-10-13 11:07:50,257 - LinearScheduler | warmup_fraction: '0.1'
2023-10-13 11:07:50,257 ----------------------------------------------------------------------------------------------------
2023-10-13 11:07:50,257 Final evaluation on model from best epoch (best-model.pt)
2023-10-13 11:07:50,257 - metric: "('micro avg', 'f1-score')"
2023-10-13 11:07:50,257 ----------------------------------------------------------------------------------------------------
2023-10-13 11:07:50,257 Computation:
2023-10-13 11:07:50,257 - compute on device: cuda:0
2023-10-13 11:07:50,257 - embedding storage: none
2023-10-13 11:07:50,257 ----------------------------------------------------------------------------------------------------
2023-10-13 11:07:50,257 Model training base path: "hmbench-ajmc/fr-dbmdz/bert-base-historic-multilingual-cased-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-5"
2023-10-13 11:07:50,257 ----------------------------------------------------------------------------------------------------
2023-10-13 11:07:50,257 ----------------------------------------------------------------------------------------------------
2023-10-13 11:07:51,349 epoch 1 - iter 24/242 - loss 3.18371892 - time (sec): 1.09 - samples/sec: 2014.36 - lr: 0.000005 - momentum: 0.000000
2023-10-13 11:07:52,451 epoch 1 - iter 48/242 - loss 2.57596024 - time (sec): 2.19 - samples/sec: 2197.83 - lr: 0.000010 - momentum: 0.000000
2023-10-13 11:07:53,520 epoch 1 - iter 72/242 - loss 1.95454570 - time (sec): 3.26 - samples/sec: 2210.52 - lr: 0.000015 - momentum: 0.000000
2023-10-13 11:07:54,630 epoch 1 - iter 96/242 - loss 1.58160402 - time (sec): 4.37 - samples/sec: 2274.47 - lr: 0.000020 - momentum: 0.000000
2023-10-13 11:07:55,735 epoch 1 - iter 120/242 - loss 1.37388781 - time (sec): 5.48 - samples/sec: 2274.02 - lr: 0.000025 - momentum: 0.000000
2023-10-13 11:07:56,807 epoch 1 - iter 144/242 - loss 1.22479270 - time (sec): 6.55 - samples/sec: 2225.17 - lr: 0.000030 - momentum: 0.000000
2023-10-13 11:07:57,917 epoch 1 - iter 168/242 - loss 1.10074637 - time (sec): 7.66 - samples/sec: 2240.57 - lr: 0.000035 - momentum: 0.000000
2023-10-13 11:07:58,997 epoch 1 - iter 192/242 - loss 1.00418203 - time (sec): 8.74 - samples/sec: 2238.04 - lr: 0.000039 - momentum: 0.000000
2023-10-13 11:08:00,089 epoch 1 - iter 216/242 - loss 0.92938882 - time (sec): 9.83 - samples/sec: 2223.02 - lr: 0.000044 - momentum: 0.000000
2023-10-13 11:08:01,205 epoch 1 - iter 240/242 - loss 0.85286162 - time (sec): 10.95 - samples/sec: 2236.10 - lr: 0.000049 - momentum: 0.000000
2023-10-13 11:08:01,298 ----------------------------------------------------------------------------------------------------
2023-10-13 11:08:01,299 EPOCH 1 done: loss 0.8448 - lr: 0.000049
2023-10-13 11:08:02,174 DEV : loss 0.20342281460762024 - f1-score (micro avg) 0.6
2023-10-13 11:08:02,179 saving best model
2023-10-13 11:08:02,540 ----------------------------------------------------------------------------------------------------
2023-10-13 11:08:03,647 epoch 2 - iter 24/242 - loss 0.22758628 - time (sec): 1.11 - samples/sec: 2334.16 - lr: 0.000049 - momentum: 0.000000
2023-10-13 11:08:04,716 epoch 2 - iter 48/242 - loss 0.21528058 - time (sec): 2.17 - samples/sec: 2390.69 - lr: 0.000049 - momentum: 0.000000
2023-10-13 11:08:05,784 epoch 2 - iter 72/242 - loss 0.22048385 - time (sec): 3.24 - samples/sec: 2270.24 - lr: 0.000048 - momentum: 0.000000
2023-10-13 11:08:06,890 epoch 2 - iter 96/242 - loss 0.20683587 - time (sec): 4.35 - samples/sec: 2282.48 - lr: 0.000048 - momentum: 0.000000
2023-10-13 11:08:07,953 epoch 2 - iter 120/242 - loss 0.19005376 - time (sec): 5.41 - samples/sec: 2263.80 - lr: 0.000047 - momentum: 0.000000
2023-10-13 11:08:09,036 epoch 2 - iter 144/242 - loss 0.18792079 - time (sec): 6.49 - samples/sec: 2270.81 - lr: 0.000047 - momentum: 0.000000
2023-10-13 11:08:10,154 epoch 2 - iter 168/242 - loss 0.18222929 - time (sec): 7.61 - samples/sec: 2270.44 - lr: 0.000046 - momentum: 0.000000
2023-10-13 11:08:11,258 epoch 2 - iter 192/242 - loss 0.17949989 - time (sec): 8.72 - samples/sec: 2269.25 - lr: 0.000046 - momentum: 0.000000
2023-10-13 11:08:12,348 epoch 2 - iter 216/242 - loss 0.17136972 - time (sec): 9.81 - samples/sec: 2271.24 - lr: 0.000045 - momentum: 0.000000
2023-10-13 11:08:13,402 epoch 2 - iter 240/242 - loss 0.16899960 - time (sec): 10.86 - samples/sec: 2268.54 - lr: 0.000045 - momentum: 0.000000
2023-10-13 11:08:13,487 ----------------------------------------------------------------------------------------------------
2023-10-13 11:08:13,487 EPOCH 2 done: loss 0.1698 - lr: 0.000045
2023-10-13 11:08:14,272 DEV : loss 0.14844480156898499 - f1-score (micro avg) 0.7626
2023-10-13 11:08:14,277 saving best model
2023-10-13 11:08:14,744 ----------------------------------------------------------------------------------------------------
2023-10-13 11:08:15,898 epoch 3 - iter 24/242 - loss 0.10777340 - time (sec): 1.15 - samples/sec: 2145.43 - lr: 0.000044 - momentum: 0.000000
2023-10-13 11:08:17,068 epoch 3 - iter 48/242 - loss 0.12615430 - time (sec): 2.32 - samples/sec: 2127.10 - lr: 0.000043 - momentum: 0.000000
2023-10-13 11:08:18,176 epoch 3 - iter 72/242 - loss 0.11486814 - time (sec): 3.43 - samples/sec: 2038.06 - lr: 0.000043 - momentum: 0.000000
2023-10-13 11:08:19,257 epoch 3 - iter 96/242 - loss 0.10526047 - time (sec): 4.51 - samples/sec: 2116.82 - lr: 0.000042 - momentum: 0.000000
2023-10-13 11:08:20,358 epoch 3 - iter 120/242 - loss 0.12266542 - time (sec): 5.61 - samples/sec: 2158.18 - lr: 0.000042 - momentum: 0.000000
2023-10-13 11:08:21,436 epoch 3 - iter 144/242 - loss 0.12778498 - time (sec): 6.69 - samples/sec: 2200.90 - lr: 0.000041 - momentum: 0.000000
2023-10-13 11:08:22,539 epoch 3 - iter 168/242 - loss 0.11820235 - time (sec): 7.79 - samples/sec: 2227.12 - lr: 0.000041 - momentum: 0.000000
2023-10-13 11:08:23,616 epoch 3 - iter 192/242 - loss 0.12499658 - time (sec): 8.87 - samples/sec: 2225.14 - lr: 0.000040 - momentum: 0.000000
2023-10-13 11:08:24,672 epoch 3 - iter 216/242 - loss 0.11906164 - time (sec): 9.92 - samples/sec: 2208.90 - lr: 0.000040 - momentum: 0.000000
2023-10-13 11:08:25,750 epoch 3 - iter 240/242 - loss 0.11394204 - time (sec): 11.00 - samples/sec: 2227.03 - lr: 0.000039 - momentum: 0.000000
2023-10-13 11:08:25,836 ----------------------------------------------------------------------------------------------------
2023-10-13 11:08:25,837 EPOCH 3 done: loss 0.1131 - lr: 0.000039
2023-10-13 11:08:26,656 DEV : loss 0.1403370201587677 - f1-score (micro avg) 0.8243
2023-10-13 11:08:26,662 saving best model
2023-10-13 11:08:27,161 ----------------------------------------------------------------------------------------------------
2023-10-13 11:08:28,331 epoch 4 - iter 24/242 - loss 0.06596586 - time (sec): 1.17 - samples/sec: 2068.96 - lr: 0.000038 - momentum: 0.000000
2023-10-13 11:08:29,552 epoch 4 - iter 48/242 - loss 0.06831484 - time (sec): 2.39 - samples/sec: 2106.37 - lr: 0.000038 - momentum: 0.000000
2023-10-13 11:08:30,743 epoch 4 - iter 72/242 - loss 0.06654065 - time (sec): 3.58 - samples/sec: 2055.36 - lr: 0.000037 - momentum: 0.000000
2023-10-13 11:08:31,906 epoch 4 - iter 96/242 - loss 0.06398005 - time (sec): 4.74 - samples/sec: 2067.72 - lr: 0.000037 - momentum: 0.000000
2023-10-13 11:08:33,021 epoch 4 - iter 120/242 - loss 0.06641754 - time (sec): 5.86 - samples/sec: 2114.27 - lr: 0.000036 - momentum: 0.000000
2023-10-13 11:08:34,089 epoch 4 - iter 144/242 - loss 0.06951926 - time (sec): 6.93 - samples/sec: 2158.36 - lr: 0.000036 - momentum: 0.000000
2023-10-13 11:08:35,176 epoch 4 - iter 168/242 - loss 0.06799826 - time (sec): 8.01 - samples/sec: 2156.32 - lr: 0.000035 - momentum: 0.000000
2023-10-13 11:08:36,247 epoch 4 - iter 192/242 - loss 0.07037930 - time (sec): 9.09 - samples/sec: 2148.14 - lr: 0.000035 - momentum: 0.000000
2023-10-13 11:08:37,328 epoch 4 - iter 216/242 - loss 0.07292490 - time (sec): 10.17 - samples/sec: 2155.33 - lr: 0.000034 - momentum: 0.000000
2023-10-13 11:08:38,464 epoch 4 - iter 240/242 - loss 0.07036982 - time (sec): 11.30 - samples/sec: 2177.46 - lr: 0.000033 - momentum: 0.000000
2023-10-13 11:08:38,556 ----------------------------------------------------------------------------------------------------
2023-10-13 11:08:38,556 EPOCH 4 done: loss 0.0700 - lr: 0.000033
2023-10-13 11:08:39,398 DEV : loss 0.1509368121623993 - f1-score (micro avg) 0.8365
2023-10-13 11:08:39,404 saving best model
2023-10-13 11:08:39,914 ----------------------------------------------------------------------------------------------------
2023-10-13 11:08:41,015 epoch 5 - iter 24/242 - loss 0.07172988 - time (sec): 1.10 - samples/sec: 2410.95 - lr: 0.000033 - momentum: 0.000000
2023-10-13 11:08:42,079 epoch 5 - iter 48/242 - loss 0.07010272 - time (sec): 2.16 - samples/sec: 2314.26 - lr: 0.000032 - momentum: 0.000000
2023-10-13 11:08:43,161 epoch 5 - iter 72/242 - loss 0.05357899 - time (sec): 3.24 - samples/sec: 2241.95 - lr: 0.000032 - momentum: 0.000000
2023-10-13 11:08:44,258 epoch 5 - iter 96/242 - loss 0.06053123 - time (sec): 4.34 - samples/sec: 2252.25 - lr: 0.000031 - momentum: 0.000000
2023-10-13 11:08:45,328 epoch 5 - iter 120/242 - loss 0.06273107 - time (sec): 5.41 - samples/sec: 2268.25 - lr: 0.000031 - momentum: 0.000000
2023-10-13 11:08:46,377 epoch 5 - iter 144/242 - loss 0.05890511 - time (sec): 6.46 - samples/sec: 2288.21 - lr: 0.000030 - momentum: 0.000000
2023-10-13 11:08:47,452 epoch 5 - iter 168/242 - loss 0.06151768 - time (sec): 7.53 - samples/sec: 2314.99 - lr: 0.000030 - momentum: 0.000000
2023-10-13 11:08:48,608 epoch 5 - iter 192/242 - loss 0.05962828 - time (sec): 8.69 - samples/sec: 2285.31 - lr: 0.000029 - momentum: 0.000000
2023-10-13 11:08:49,735 epoch 5 - iter 216/242 - loss 0.05957175 - time (sec): 9.82 - samples/sec: 2286.62 - lr: 0.000028 - momentum: 0.000000
2023-10-13 11:08:50,786 epoch 5 - iter 240/242 - loss 0.05693543 - time (sec): 10.87 - samples/sec: 2264.38 - lr: 0.000028 - momentum: 0.000000
2023-10-13 11:08:50,868 ----------------------------------------------------------------------------------------------------
2023-10-13 11:08:50,868 EPOCH 5 done: loss 0.0572 - lr: 0.000028
2023-10-13 11:08:51,759 DEV : loss 0.18690580129623413 - f1-score (micro avg) 0.8143
2023-10-13 11:08:51,766 ----------------------------------------------------------------------------------------------------
2023-10-13 11:08:53,082 epoch 6 - iter 24/242 - loss 0.05722285 - time (sec): 1.31 - samples/sec: 1970.01 - lr: 0.000027 - momentum: 0.000000
2023-10-13 11:08:54,456 epoch 6 - iter 48/242 - loss 0.05244200 - time (sec): 2.69 - samples/sec: 1893.11 - lr: 0.000027 - momentum: 0.000000
2023-10-13 11:08:55,814 epoch 6 - iter 72/242 - loss 0.04904005 - time (sec): 4.05 - samples/sec: 1902.82 - lr: 0.000026 - momentum: 0.000000
2023-10-13 11:08:57,057 epoch 6 - iter 96/242 - loss 0.04467185 - time (sec): 5.29 - samples/sec: 1854.33 - lr: 0.000026 - momentum: 0.000000
2023-10-13 11:08:58,206 epoch 6 - iter 120/242 - loss 0.04420946 - time (sec): 6.44 - samples/sec: 1940.66 - lr: 0.000025 - momentum: 0.000000
2023-10-13 11:08:59,280 epoch 6 - iter 144/242 - loss 0.03970356 - time (sec): 7.51 - samples/sec: 1985.99 - lr: 0.000025 - momentum: 0.000000
2023-10-13 11:09:00,328 epoch 6 - iter 168/242 - loss 0.04408623 - time (sec): 8.56 - samples/sec: 1993.20 - lr: 0.000024 - momentum: 0.000000
2023-10-13 11:09:01,398 epoch 6 - iter 192/242 - loss 0.04034396 - time (sec): 9.63 - samples/sec: 2014.68 - lr: 0.000023 - momentum: 0.000000
2023-10-13 11:09:02,476 epoch 6 - iter 216/242 - loss 0.03986324 - time (sec): 10.71 - samples/sec: 2039.88 - lr: 0.000023 - momentum: 0.000000
2023-10-13 11:09:03,576 epoch 6 - iter 240/242 - loss 0.03959649 - time (sec): 11.81 - samples/sec: 2083.43 - lr: 0.000022 - momentum: 0.000000
2023-10-13 11:09:03,664 ----------------------------------------------------------------------------------------------------
2023-10-13 11:09:03,664 EPOCH 6 done: loss 0.0399 - lr: 0.000022
2023-10-13 11:09:04,432 DEV : loss 0.20703579485416412 - f1-score (micro avg) 0.8067
2023-10-13 11:09:04,437 ----------------------------------------------------------------------------------------------------
2023-10-13 11:09:05,523 epoch 7 - iter 24/242 - loss 0.01304402 - time (sec): 1.08 - samples/sec: 2354.20 - lr: 0.000022 - momentum: 0.000000
2023-10-13 11:09:06,615 epoch 7 - iter 48/242 - loss 0.01891244 - time (sec): 2.18 - samples/sec: 2320.61 - lr: 0.000021 - momentum: 0.000000
2023-10-13 11:09:07,679 epoch 7 - iter 72/242 - loss 0.02175415 - time (sec): 3.24 - samples/sec: 2271.91 - lr: 0.000021 - momentum: 0.000000
2023-10-13 11:09:08,749 epoch 7 - iter 96/242 - loss 0.02731527 - time (sec): 4.31 - samples/sec: 2262.01 - lr: 0.000020 - momentum: 0.000000
2023-10-13 11:09:09,812 epoch 7 - iter 120/242 - loss 0.03234705 - time (sec): 5.37 - samples/sec: 2276.02 - lr: 0.000020 - momentum: 0.000000
2023-10-13 11:09:10,923 epoch 7 - iter 144/242 - loss 0.02891570 - time (sec): 6.48 - samples/sec: 2285.96 - lr: 0.000019 - momentum: 0.000000
2023-10-13 11:09:11,986 epoch 7 - iter 168/242 - loss 0.02815719 - time (sec): 7.55 - samples/sec: 2295.21 - lr: 0.000018 - momentum: 0.000000
2023-10-13 11:09:13,061 epoch 7 - iter 192/242 - loss 0.02865721 - time (sec): 8.62 - samples/sec: 2270.16 - lr: 0.000018 - momentum: 0.000000
2023-10-13 11:09:14,140 epoch 7 - iter 216/242 - loss 0.02953733 - time (sec): 9.70 - samples/sec: 2277.81 - lr: 0.000017 - momentum: 0.000000
2023-10-13 11:09:15,223 epoch 7 - iter 240/242 - loss 0.02922147 - time (sec): 10.78 - samples/sec: 2274.96 - lr: 0.000017 - momentum: 0.000000
2023-10-13 11:09:15,311 ----------------------------------------------------------------------------------------------------
2023-10-13 11:09:15,311 EPOCH 7 done: loss 0.0290 - lr: 0.000017
2023-10-13 11:09:16,067 DEV : loss 0.203294038772583 - f1-score (micro avg) 0.8335
2023-10-13 11:09:16,072 ----------------------------------------------------------------------------------------------------
2023-10-13 11:09:17,175 epoch 8 - iter 24/242 - loss 0.02027913 - time (sec): 1.10 - samples/sec: 2439.39 - lr: 0.000016 - momentum: 0.000000
2023-10-13 11:09:18,248 epoch 8 - iter 48/242 - loss 0.02740789 - time (sec): 2.18 - samples/sec: 2388.95 - lr: 0.000016 - momentum: 0.000000
2023-10-13 11:09:19,334 epoch 8 - iter 72/242 - loss 0.02091675 - time (sec): 3.26 - samples/sec: 2361.24 - lr: 0.000015 - momentum: 0.000000
2023-10-13 11:09:20,420 epoch 8 - iter 96/242 - loss 0.02309290 - time (sec): 4.35 - samples/sec: 2349.70 - lr: 0.000015 - momentum: 0.000000
2023-10-13 11:09:21,490 epoch 8 - iter 120/242 - loss 0.02237206 - time (sec): 5.42 - samples/sec: 2367.97 - lr: 0.000014 - momentum: 0.000000
2023-10-13 11:09:22,599 epoch 8 - iter 144/242 - loss 0.02123187 - time (sec): 6.53 - samples/sec: 2329.24 - lr: 0.000013 - momentum: 0.000000
2023-10-13 11:09:23,709 epoch 8 - iter 168/242 - loss 0.02196109 - time (sec): 7.64 - samples/sec: 2288.74 - lr: 0.000013 - momentum: 0.000000
2023-10-13 11:09:24,804 epoch 8 - iter 192/242 - loss 0.02266697 - time (sec): 8.73 - samples/sec: 2269.13 - lr: 0.000012 - momentum: 0.000000
2023-10-13 11:09:25,914 epoch 8 - iter 216/242 - loss 0.02158182 - time (sec): 9.84 - samples/sec: 2271.83 - lr: 0.000012 - momentum: 0.000000
2023-10-13 11:09:26,980 epoch 8 - iter 240/242 - loss 0.02105113 - time (sec): 10.91 - samples/sec: 2260.99 - lr: 0.000011 - momentum: 0.000000
2023-10-13 11:09:27,061 ----------------------------------------------------------------------------------------------------
2023-10-13 11:09:27,061 EPOCH 8 done: loss 0.0210 - lr: 0.000011
2023-10-13 11:09:28,017 DEV : loss 0.19338418543338776 - f1-score (micro avg) 0.8507
2023-10-13 11:09:28,023 saving best model
2023-10-13 11:09:28,540 ----------------------------------------------------------------------------------------------------
2023-10-13 11:09:29,691 epoch 9 - iter 24/242 - loss 0.01795266 - time (sec): 1.15 - samples/sec: 2122.41 - lr: 0.000011 - momentum: 0.000000
2023-10-13 11:09:30,875 epoch 9 - iter 48/242 - loss 0.01915869 - time (sec): 2.33 - samples/sec: 2182.73 - lr: 0.000010 - momentum: 0.000000
2023-10-13 11:09:32,038 epoch 9 - iter 72/242 - loss 0.02158611 - time (sec): 3.49 - samples/sec: 2223.27 - lr: 0.000010 - momentum: 0.000000
2023-10-13 11:09:33,189 epoch 9 - iter 96/242 - loss 0.01719087 - time (sec): 4.65 - samples/sec: 2120.24 - lr: 0.000009 - momentum: 0.000000
2023-10-13 11:09:34,331 epoch 9 - iter 120/242 - loss 0.01594553 - time (sec): 5.79 - samples/sec: 2076.69 - lr: 0.000008 - momentum: 0.000000
2023-10-13 11:09:35,409 epoch 9 - iter 144/242 - loss 0.01441884 - time (sec): 6.87 - samples/sec: 2112.07 - lr: 0.000008 - momentum: 0.000000
2023-10-13 11:09:36,489 epoch 9 - iter 168/242 - loss 0.01395817 - time (sec): 7.95 - samples/sec: 2166.27 - lr: 0.000007 - momentum: 0.000000
2023-10-13 11:09:37,582 epoch 9 - iter 192/242 - loss 0.01254251 - time (sec): 9.04 - samples/sec: 2170.68 - lr: 0.000007 - momentum: 0.000000
2023-10-13 11:09:38,636 epoch 9 - iter 216/242 - loss 0.01437384 - time (sec): 10.09 - samples/sec: 2159.19 - lr: 0.000006 - momentum: 0.000000
2023-10-13 11:09:39,741 epoch 9 - iter 240/242 - loss 0.01382778 - time (sec): 11.20 - samples/sec: 2197.18 - lr: 0.000006 - momentum: 0.000000
2023-10-13 11:09:39,827 ----------------------------------------------------------------------------------------------------
2023-10-13 11:09:39,827 EPOCH 9 done: loss 0.0140 - lr: 0.000006
2023-10-13 11:09:40,667 DEV : loss 0.1840001940727234 - f1-score (micro avg) 0.8553
2023-10-13 11:09:40,673 saving best model
2023-10-13 11:09:41,174 ----------------------------------------------------------------------------------------------------
2023-10-13 11:09:42,272 epoch 10 - iter 24/242 - loss 0.00815395 - time (sec): 1.10 - samples/sec: 2217.13 - lr: 0.000005 - momentum: 0.000000
2023-10-13 11:09:43,373 epoch 10 - iter 48/242 - loss 0.01267221 - time (sec): 2.20 - samples/sec: 2360.19 - lr: 0.000005 - momentum: 0.000000
2023-10-13 11:09:44,445 epoch 10 - iter 72/242 - loss 0.00988114 - time (sec): 3.27 - samples/sec: 2277.47 - lr: 0.000004 - momentum: 0.000000
2023-10-13 11:09:45,514 epoch 10 - iter 96/242 - loss 0.01014757 - time (sec): 4.34 - samples/sec: 2212.92 - lr: 0.000003 - momentum: 0.000000
2023-10-13 11:09:46,614 epoch 10 - iter 120/242 - loss 0.00919350 - time (sec): 5.44 - samples/sec: 2298.46 - lr: 0.000003 - momentum: 0.000000
2023-10-13 11:09:47,695 epoch 10 - iter 144/242 - loss 0.00878841 - time (sec): 6.52 - samples/sec: 2280.14 - lr: 0.000002 - momentum: 0.000000
2023-10-13 11:09:48,784 epoch 10 - iter 168/242 - loss 0.01029727 - time (sec): 7.61 - samples/sec: 2258.55 - lr: 0.000002 - momentum: 0.000000
2023-10-13 11:09:49,851 epoch 10 - iter 192/242 - loss 0.01213754 - time (sec): 8.67 - samples/sec: 2239.26 - lr: 0.000001 - momentum: 0.000000
2023-10-13 11:09:50,933 epoch 10 - iter 216/242 - loss 0.01112444 - time (sec): 9.76 - samples/sec: 2245.84 - lr: 0.000001 - momentum: 0.000000
2023-10-13 11:09:52,058 epoch 10 - iter 240/242 - loss 0.01028077 - time (sec): 10.88 - samples/sec: 2257.48 - lr: 0.000000 - momentum: 0.000000
2023-10-13 11:09:52,141 ----------------------------------------------------------------------------------------------------
2023-10-13 11:09:52,142 EPOCH 10 done: loss 0.0102 - lr: 0.000000
2023-10-13 11:09:52,913 DEV : loss 0.19032004475593567 - f1-score (micro avg) 0.8468
2023-10-13 11:09:53,275 ----------------------------------------------------------------------------------------------------
2023-10-13 11:09:53,276 Loading model from best epoch ...
2023-10-13 11:09:54,656 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 11:09:55,491
Results:
- F-score (micro) 0.8235
- F-score (macro) 0.5974
- Accuracy 0.7146
By class:
precision recall f1-score support
pers 0.8581 0.9137 0.8850 139
scope 0.8156 0.8915 0.8519 129
work 0.6667 0.7750 0.7168 80
loc 0.6667 0.4444 0.5333 9
date 0.0000 0.0000 0.0000 3
micro avg 0.7938 0.8556 0.8235 360
macro avg 0.6014 0.6049 0.5974 360
weighted avg 0.7884 0.8556 0.8196 360
2023-10-13 11:09:55,491 ----------------------------------------------------------------------------------------------------
|