File size: 24,055 Bytes
295fa69 |
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 |
2023-10-25 17:12:05,512 ----------------------------------------------------------------------------------------------------
2023-10-25 17:12:05,513 Model: "SequenceTagger(
(embeddings): TransformerWordEmbeddings(
(model): BertModel(
(embeddings): BertEmbeddings(
(word_embeddings): Embedding(64001, 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=17, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-25 17:12:05,514 ----------------------------------------------------------------------------------------------------
2023-10-25 17:12:05,514 MultiCorpus: 7142 train + 698 dev + 2570 test sentences
- NER_HIPE_2022 Corpus: 7142 train + 698 dev + 2570 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/newseye/fr/with_doc_seperator
2023-10-25 17:12:05,514 ----------------------------------------------------------------------------------------------------
2023-10-25 17:12:05,514 Train: 7142 sentences
2023-10-25 17:12:05,514 (train_with_dev=False, train_with_test=False)
2023-10-25 17:12:05,514 ----------------------------------------------------------------------------------------------------
2023-10-25 17:12:05,514 Training Params:
2023-10-25 17:12:05,514 - learning_rate: "5e-05"
2023-10-25 17:12:05,514 - mini_batch_size: "8"
2023-10-25 17:12:05,514 - max_epochs: "10"
2023-10-25 17:12:05,514 - shuffle: "True"
2023-10-25 17:12:05,514 ----------------------------------------------------------------------------------------------------
2023-10-25 17:12:05,514 Plugins:
2023-10-25 17:12:05,514 - TensorboardLogger
2023-10-25 17:12:05,514 - LinearScheduler | warmup_fraction: '0.1'
2023-10-25 17:12:05,515 ----------------------------------------------------------------------------------------------------
2023-10-25 17:12:05,515 Final evaluation on model from best epoch (best-model.pt)
2023-10-25 17:12:05,515 - metric: "('micro avg', 'f1-score')"
2023-10-25 17:12:05,515 ----------------------------------------------------------------------------------------------------
2023-10-25 17:12:05,515 Computation:
2023-10-25 17:12:05,515 - compute on device: cuda:0
2023-10-25 17:12:05,515 - embedding storage: none
2023-10-25 17:12:05,515 ----------------------------------------------------------------------------------------------------
2023-10-25 17:12:05,515 Model training base path: "hmbench-newseye/fr-dbmdz/bert-base-historic-multilingual-64k-td-cased-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-4"
2023-10-25 17:12:05,515 ----------------------------------------------------------------------------------------------------
2023-10-25 17:12:05,515 ----------------------------------------------------------------------------------------------------
2023-10-25 17:12:05,515 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-25 17:12:11,754 epoch 1 - iter 89/893 - loss 1.86966354 - time (sec): 6.24 - samples/sec: 4048.40 - lr: 0.000005 - momentum: 0.000000
2023-10-25 17:12:18,158 epoch 1 - iter 178/893 - loss 1.17037592 - time (sec): 12.64 - samples/sec: 4015.73 - lr: 0.000010 - momentum: 0.000000
2023-10-25 17:12:24,252 epoch 1 - iter 267/893 - loss 0.89779536 - time (sec): 18.74 - samples/sec: 3990.87 - lr: 0.000015 - momentum: 0.000000
2023-10-25 17:12:30,281 epoch 1 - iter 356/893 - loss 0.72799361 - time (sec): 24.77 - samples/sec: 4021.65 - lr: 0.000020 - momentum: 0.000000
2023-10-25 17:12:36,084 epoch 1 - iter 445/893 - loss 0.62359043 - time (sec): 30.57 - samples/sec: 4037.40 - lr: 0.000025 - momentum: 0.000000
2023-10-25 17:12:41,987 epoch 1 - iter 534/893 - loss 0.54978542 - time (sec): 36.47 - samples/sec: 4059.93 - lr: 0.000030 - momentum: 0.000000
2023-10-25 17:12:48,656 epoch 1 - iter 623/893 - loss 0.49086057 - time (sec): 43.14 - samples/sec: 4008.81 - lr: 0.000035 - momentum: 0.000000
2023-10-25 17:12:54,781 epoch 1 - iter 712/893 - loss 0.44720365 - time (sec): 49.27 - samples/sec: 4036.10 - lr: 0.000040 - momentum: 0.000000
2023-10-25 17:13:00,618 epoch 1 - iter 801/893 - loss 0.41503176 - time (sec): 55.10 - samples/sec: 4063.39 - lr: 0.000045 - momentum: 0.000000
2023-10-25 17:13:06,507 epoch 1 - iter 890/893 - loss 0.38844168 - time (sec): 60.99 - samples/sec: 4060.90 - lr: 0.000050 - momentum: 0.000000
2023-10-25 17:13:06,717 ----------------------------------------------------------------------------------------------------
2023-10-25 17:13:06,717 EPOCH 1 done: loss 0.3873 - lr: 0.000050
2023-10-25 17:13:09,813 DEV : loss 0.10254143178462982 - f1-score (micro avg) 0.7386
2023-10-25 17:13:09,836 saving best model
2023-10-25 17:13:10,384 ----------------------------------------------------------------------------------------------------
2023-10-25 17:13:16,410 epoch 2 - iter 89/893 - loss 0.11297432 - time (sec): 6.02 - samples/sec: 4100.95 - lr: 0.000049 - momentum: 0.000000
2023-10-25 17:13:22,388 epoch 2 - iter 178/893 - loss 0.10095221 - time (sec): 12.00 - samples/sec: 4091.37 - lr: 0.000049 - momentum: 0.000000
2023-10-25 17:13:28,578 epoch 2 - iter 267/893 - loss 0.09904465 - time (sec): 18.19 - samples/sec: 4123.74 - lr: 0.000048 - momentum: 0.000000
2023-10-25 17:13:34,706 epoch 2 - iter 356/893 - loss 0.10304399 - time (sec): 24.32 - samples/sec: 4161.23 - lr: 0.000048 - momentum: 0.000000
2023-10-25 17:13:40,891 epoch 2 - iter 445/893 - loss 0.10290863 - time (sec): 30.50 - samples/sec: 4163.32 - lr: 0.000047 - momentum: 0.000000
2023-10-25 17:13:46,799 epoch 2 - iter 534/893 - loss 0.10253728 - time (sec): 36.41 - samples/sec: 4138.20 - lr: 0.000047 - momentum: 0.000000
2023-10-25 17:13:52,856 epoch 2 - iter 623/893 - loss 0.10517292 - time (sec): 42.47 - samples/sec: 4127.97 - lr: 0.000046 - momentum: 0.000000
2023-10-25 17:13:59,058 epoch 2 - iter 712/893 - loss 0.10448830 - time (sec): 48.67 - samples/sec: 4121.70 - lr: 0.000046 - momentum: 0.000000
2023-10-25 17:14:05,695 epoch 2 - iter 801/893 - loss 0.10541126 - time (sec): 55.31 - samples/sec: 4030.63 - lr: 0.000045 - momentum: 0.000000
2023-10-25 17:14:11,612 epoch 2 - iter 890/893 - loss 0.10497282 - time (sec): 61.23 - samples/sec: 4053.12 - lr: 0.000044 - momentum: 0.000000
2023-10-25 17:14:11,796 ----------------------------------------------------------------------------------------------------
2023-10-25 17:14:11,797 EPOCH 2 done: loss 0.1048 - lr: 0.000044
2023-10-25 17:14:15,732 DEV : loss 0.09835401922464371 - f1-score (micro avg) 0.7484
2023-10-25 17:14:15,753 saving best model
2023-10-25 17:14:17,077 ----------------------------------------------------------------------------------------------------
2023-10-25 17:14:22,720 epoch 3 - iter 89/893 - loss 0.06670513 - time (sec): 5.64 - samples/sec: 4160.09 - lr: 0.000044 - momentum: 0.000000
2023-10-25 17:14:28,626 epoch 3 - iter 178/893 - loss 0.06908344 - time (sec): 11.55 - samples/sec: 4278.75 - lr: 0.000043 - momentum: 0.000000
2023-10-25 17:14:34,039 epoch 3 - iter 267/893 - loss 0.06482755 - time (sec): 16.96 - samples/sec: 4378.26 - lr: 0.000043 - momentum: 0.000000
2023-10-25 17:14:39,692 epoch 3 - iter 356/893 - loss 0.06663973 - time (sec): 22.61 - samples/sec: 4381.76 - lr: 0.000042 - momentum: 0.000000
2023-10-25 17:14:45,211 epoch 3 - iter 445/893 - loss 0.06653035 - time (sec): 28.13 - samples/sec: 4430.59 - lr: 0.000042 - momentum: 0.000000
2023-10-25 17:14:50,834 epoch 3 - iter 534/893 - loss 0.06616108 - time (sec): 33.75 - samples/sec: 4434.83 - lr: 0.000041 - momentum: 0.000000
2023-10-25 17:14:56,338 epoch 3 - iter 623/893 - loss 0.06517711 - time (sec): 39.26 - samples/sec: 4441.20 - lr: 0.000041 - momentum: 0.000000
2023-10-25 17:15:01,626 epoch 3 - iter 712/893 - loss 0.06547935 - time (sec): 44.55 - samples/sec: 4424.27 - lr: 0.000040 - momentum: 0.000000
2023-10-25 17:15:07,582 epoch 3 - iter 801/893 - loss 0.06531523 - time (sec): 50.50 - samples/sec: 4426.12 - lr: 0.000039 - momentum: 0.000000
2023-10-25 17:15:13,138 epoch 3 - iter 890/893 - loss 0.06609427 - time (sec): 56.06 - samples/sec: 4424.89 - lr: 0.000039 - momentum: 0.000000
2023-10-25 17:15:13,306 ----------------------------------------------------------------------------------------------------
2023-10-25 17:15:13,306 EPOCH 3 done: loss 0.0663 - lr: 0.000039
2023-10-25 17:15:18,728 DEV : loss 0.10818858444690704 - f1-score (micro avg) 0.78
2023-10-25 17:15:18,750 saving best model
2023-10-25 17:15:19,433 ----------------------------------------------------------------------------------------------------
2023-10-25 17:15:25,434 epoch 4 - iter 89/893 - loss 0.04360221 - time (sec): 6.00 - samples/sec: 4325.89 - lr: 0.000038 - momentum: 0.000000
2023-10-25 17:15:31,204 epoch 4 - iter 178/893 - loss 0.04758612 - time (sec): 11.77 - samples/sec: 4382.46 - lr: 0.000038 - momentum: 0.000000
2023-10-25 17:15:36,707 epoch 4 - iter 267/893 - loss 0.04849685 - time (sec): 17.27 - samples/sec: 4342.52 - lr: 0.000037 - momentum: 0.000000
2023-10-25 17:15:42,413 epoch 4 - iter 356/893 - loss 0.04765784 - time (sec): 22.98 - samples/sec: 4307.89 - lr: 0.000037 - momentum: 0.000000
2023-10-25 17:15:48,360 epoch 4 - iter 445/893 - loss 0.04585792 - time (sec): 28.92 - samples/sec: 4301.93 - lr: 0.000036 - momentum: 0.000000
2023-10-25 17:15:53,987 epoch 4 - iter 534/893 - loss 0.04678008 - time (sec): 34.55 - samples/sec: 4339.31 - lr: 0.000036 - momentum: 0.000000
2023-10-25 17:15:59,524 epoch 4 - iter 623/893 - loss 0.04623631 - time (sec): 40.09 - samples/sec: 4325.05 - lr: 0.000035 - momentum: 0.000000
2023-10-25 17:16:05,227 epoch 4 - iter 712/893 - loss 0.04616789 - time (sec): 45.79 - samples/sec: 4335.78 - lr: 0.000034 - momentum: 0.000000
2023-10-25 17:16:10,897 epoch 4 - iter 801/893 - loss 0.04654488 - time (sec): 51.46 - samples/sec: 4351.22 - lr: 0.000034 - momentum: 0.000000
2023-10-25 17:16:16,408 epoch 4 - iter 890/893 - loss 0.04626882 - time (sec): 56.97 - samples/sec: 4340.98 - lr: 0.000033 - momentum: 0.000000
2023-10-25 17:16:16,693 ----------------------------------------------------------------------------------------------------
2023-10-25 17:16:16,694 EPOCH 4 done: loss 0.0460 - lr: 0.000033
2023-10-25 17:16:21,128 DEV : loss 0.14430440962314606 - f1-score (micro avg) 0.7763
2023-10-25 17:16:21,148 ----------------------------------------------------------------------------------------------------
2023-10-25 17:16:26,833 epoch 5 - iter 89/893 - loss 0.03076065 - time (sec): 5.68 - samples/sec: 4079.73 - lr: 0.000033 - momentum: 0.000000
2023-10-25 17:16:32,613 epoch 5 - iter 178/893 - loss 0.03387457 - time (sec): 11.46 - samples/sec: 4189.45 - lr: 0.000032 - momentum: 0.000000
2023-10-25 17:16:38,412 epoch 5 - iter 267/893 - loss 0.03857465 - time (sec): 17.26 - samples/sec: 4219.15 - lr: 0.000032 - momentum: 0.000000
2023-10-25 17:16:44,270 epoch 5 - iter 356/893 - loss 0.03668496 - time (sec): 23.12 - samples/sec: 4226.11 - lr: 0.000031 - momentum: 0.000000
2023-10-25 17:16:50,915 epoch 5 - iter 445/893 - loss 0.03779554 - time (sec): 29.77 - samples/sec: 4131.62 - lr: 0.000031 - momentum: 0.000000
2023-10-25 17:16:56,745 epoch 5 - iter 534/893 - loss 0.03736583 - time (sec): 35.60 - samples/sec: 4160.67 - lr: 0.000030 - momentum: 0.000000
2023-10-25 17:17:02,359 epoch 5 - iter 623/893 - loss 0.03619574 - time (sec): 41.21 - samples/sec: 4183.66 - lr: 0.000029 - momentum: 0.000000
2023-10-25 17:17:08,388 epoch 5 - iter 712/893 - loss 0.03616854 - time (sec): 47.24 - samples/sec: 4165.28 - lr: 0.000029 - momentum: 0.000000
2023-10-25 17:17:14,180 epoch 5 - iter 801/893 - loss 0.03605795 - time (sec): 53.03 - samples/sec: 4207.84 - lr: 0.000028 - momentum: 0.000000
2023-10-25 17:17:19,715 epoch 5 - iter 890/893 - loss 0.03609116 - time (sec): 58.57 - samples/sec: 4231.73 - lr: 0.000028 - momentum: 0.000000
2023-10-25 17:17:19,915 ----------------------------------------------------------------------------------------------------
2023-10-25 17:17:19,915 EPOCH 5 done: loss 0.0361 - lr: 0.000028
2023-10-25 17:17:23,906 DEV : loss 0.1808791607618332 - f1-score (micro avg) 0.7874
2023-10-25 17:17:23,926 saving best model
2023-10-25 17:17:24,575 ----------------------------------------------------------------------------------------------------
2023-10-25 17:17:30,427 epoch 6 - iter 89/893 - loss 0.03174096 - time (sec): 5.85 - samples/sec: 4049.45 - lr: 0.000027 - momentum: 0.000000
2023-10-25 17:17:36,184 epoch 6 - iter 178/893 - loss 0.02710066 - time (sec): 11.61 - samples/sec: 4011.60 - lr: 0.000027 - momentum: 0.000000
2023-10-25 17:17:42,186 epoch 6 - iter 267/893 - loss 0.02807563 - time (sec): 17.61 - samples/sec: 4098.40 - lr: 0.000026 - momentum: 0.000000
2023-10-25 17:17:48,150 epoch 6 - iter 356/893 - loss 0.02692728 - time (sec): 23.57 - samples/sec: 4121.36 - lr: 0.000026 - momentum: 0.000000
2023-10-25 17:17:54,081 epoch 6 - iter 445/893 - loss 0.02735570 - time (sec): 29.50 - samples/sec: 4159.09 - lr: 0.000025 - momentum: 0.000000
2023-10-25 17:18:00,130 epoch 6 - iter 534/893 - loss 0.02805044 - time (sec): 35.55 - samples/sec: 4171.09 - lr: 0.000024 - momentum: 0.000000
2023-10-25 17:18:06,195 epoch 6 - iter 623/893 - loss 0.02725646 - time (sec): 41.62 - samples/sec: 4157.30 - lr: 0.000024 - momentum: 0.000000
2023-10-25 17:18:12,254 epoch 6 - iter 712/893 - loss 0.02769298 - time (sec): 47.68 - samples/sec: 4162.52 - lr: 0.000023 - momentum: 0.000000
2023-10-25 17:18:18,094 epoch 6 - iter 801/893 - loss 0.02746510 - time (sec): 53.52 - samples/sec: 4161.49 - lr: 0.000023 - momentum: 0.000000
2023-10-25 17:18:24,001 epoch 6 - iter 890/893 - loss 0.02725677 - time (sec): 59.42 - samples/sec: 4178.69 - lr: 0.000022 - momentum: 0.000000
2023-10-25 17:18:24,190 ----------------------------------------------------------------------------------------------------
2023-10-25 17:18:24,190 EPOCH 6 done: loss 0.0274 - lr: 0.000022
2023-10-25 17:18:29,212 DEV : loss 0.18829816579818726 - f1-score (micro avg) 0.8008
2023-10-25 17:18:29,234 saving best model
2023-10-25 17:18:29,906 ----------------------------------------------------------------------------------------------------
2023-10-25 17:18:35,941 epoch 7 - iter 89/893 - loss 0.01515087 - time (sec): 6.03 - samples/sec: 3972.69 - lr: 0.000022 - momentum: 0.000000
2023-10-25 17:18:41,986 epoch 7 - iter 178/893 - loss 0.02051512 - time (sec): 12.08 - samples/sec: 4024.69 - lr: 0.000021 - momentum: 0.000000
2023-10-25 17:18:48,051 epoch 7 - iter 267/893 - loss 0.01998553 - time (sec): 18.14 - samples/sec: 4115.71 - lr: 0.000021 - momentum: 0.000000
2023-10-25 17:18:54,021 epoch 7 - iter 356/893 - loss 0.01986934 - time (sec): 24.11 - samples/sec: 4124.11 - lr: 0.000020 - momentum: 0.000000
2023-10-25 17:18:59,887 epoch 7 - iter 445/893 - loss 0.02129780 - time (sec): 29.98 - samples/sec: 4178.30 - lr: 0.000019 - momentum: 0.000000
2023-10-25 17:19:05,771 epoch 7 - iter 534/893 - loss 0.02089146 - time (sec): 35.86 - samples/sec: 4208.17 - lr: 0.000019 - momentum: 0.000000
2023-10-25 17:19:11,666 epoch 7 - iter 623/893 - loss 0.02180007 - time (sec): 41.76 - samples/sec: 4201.24 - lr: 0.000018 - momentum: 0.000000
2023-10-25 17:19:17,316 epoch 7 - iter 712/893 - loss 0.02130021 - time (sec): 47.41 - samples/sec: 4179.67 - lr: 0.000018 - momentum: 0.000000
2023-10-25 17:19:23,114 epoch 7 - iter 801/893 - loss 0.02092378 - time (sec): 53.21 - samples/sec: 4187.33 - lr: 0.000017 - momentum: 0.000000
2023-10-25 17:19:29,000 epoch 7 - iter 890/893 - loss 0.02059414 - time (sec): 59.09 - samples/sec: 4200.88 - lr: 0.000017 - momentum: 0.000000
2023-10-25 17:19:29,168 ----------------------------------------------------------------------------------------------------
2023-10-25 17:19:29,168 EPOCH 7 done: loss 0.0206 - lr: 0.000017
2023-10-25 17:19:33,135 DEV : loss 0.20971202850341797 - f1-score (micro avg) 0.7835
2023-10-25 17:19:33,158 ----------------------------------------------------------------------------------------------------
2023-10-25 17:19:39,046 epoch 8 - iter 89/893 - loss 0.01764633 - time (sec): 5.89 - samples/sec: 4379.59 - lr: 0.000016 - momentum: 0.000000
2023-10-25 17:19:45,066 epoch 8 - iter 178/893 - loss 0.01628569 - time (sec): 11.91 - samples/sec: 4235.59 - lr: 0.000016 - momentum: 0.000000
2023-10-25 17:19:51,714 epoch 8 - iter 267/893 - loss 0.01625886 - time (sec): 18.55 - samples/sec: 4031.94 - lr: 0.000015 - momentum: 0.000000
2023-10-25 17:19:57,412 epoch 8 - iter 356/893 - loss 0.01628296 - time (sec): 24.25 - samples/sec: 4045.87 - lr: 0.000014 - momentum: 0.000000
2023-10-25 17:20:03,072 epoch 8 - iter 445/893 - loss 0.01501426 - time (sec): 29.91 - samples/sec: 4086.07 - lr: 0.000014 - momentum: 0.000000
2023-10-25 17:20:08,856 epoch 8 - iter 534/893 - loss 0.01465639 - time (sec): 35.70 - samples/sec: 4134.41 - lr: 0.000013 - momentum: 0.000000
2023-10-25 17:20:14,511 epoch 8 - iter 623/893 - loss 0.01504650 - time (sec): 41.35 - samples/sec: 4158.41 - lr: 0.000013 - momentum: 0.000000
2023-10-25 17:20:20,232 epoch 8 - iter 712/893 - loss 0.01458875 - time (sec): 47.07 - samples/sec: 4166.37 - lr: 0.000012 - momentum: 0.000000
2023-10-25 17:20:26,137 epoch 8 - iter 801/893 - loss 0.01570324 - time (sec): 52.98 - samples/sec: 4186.22 - lr: 0.000012 - momentum: 0.000000
2023-10-25 17:20:32,101 epoch 8 - iter 890/893 - loss 0.01581706 - time (sec): 58.94 - samples/sec: 4207.71 - lr: 0.000011 - momentum: 0.000000
2023-10-25 17:20:32,280 ----------------------------------------------------------------------------------------------------
2023-10-25 17:20:32,281 EPOCH 8 done: loss 0.0158 - lr: 0.000011
2023-10-25 17:20:36,350 DEV : loss 0.21289943158626556 - f1-score (micro avg) 0.8
2023-10-25 17:20:36,375 ----------------------------------------------------------------------------------------------------
2023-10-25 17:20:42,196 epoch 9 - iter 89/893 - loss 0.00765889 - time (sec): 5.82 - samples/sec: 4349.70 - lr: 0.000011 - momentum: 0.000000
2023-10-25 17:20:48,053 epoch 9 - iter 178/893 - loss 0.00998425 - time (sec): 11.68 - samples/sec: 4306.03 - lr: 0.000010 - momentum: 0.000000
2023-10-25 17:20:54,019 epoch 9 - iter 267/893 - loss 0.01244333 - time (sec): 17.64 - samples/sec: 4202.39 - lr: 0.000009 - momentum: 0.000000
2023-10-25 17:20:59,745 epoch 9 - iter 356/893 - loss 0.01212745 - time (sec): 23.37 - samples/sec: 4282.98 - lr: 0.000009 - momentum: 0.000000
2023-10-25 17:21:05,354 epoch 9 - iter 445/893 - loss 0.01178149 - time (sec): 28.98 - samples/sec: 4326.46 - lr: 0.000008 - momentum: 0.000000
2023-10-25 17:21:10,965 epoch 9 - iter 534/893 - loss 0.01170529 - time (sec): 34.59 - samples/sec: 4302.74 - lr: 0.000008 - momentum: 0.000000
2023-10-25 17:21:16,851 epoch 9 - iter 623/893 - loss 0.01136873 - time (sec): 40.47 - samples/sec: 4324.76 - lr: 0.000007 - momentum: 0.000000
2023-10-25 17:21:22,488 epoch 9 - iter 712/893 - loss 0.01142673 - time (sec): 46.11 - samples/sec: 4298.22 - lr: 0.000007 - momentum: 0.000000
2023-10-25 17:21:28,182 epoch 9 - iter 801/893 - loss 0.01136731 - time (sec): 51.81 - samples/sec: 4291.71 - lr: 0.000006 - momentum: 0.000000
2023-10-25 17:21:34,076 epoch 9 - iter 890/893 - loss 0.01120456 - time (sec): 57.70 - samples/sec: 4294.92 - lr: 0.000006 - momentum: 0.000000
2023-10-25 17:21:34,259 ----------------------------------------------------------------------------------------------------
2023-10-25 17:21:34,259 EPOCH 9 done: loss 0.0112 - lr: 0.000006
2023-10-25 17:21:39,355 DEV : loss 0.22147664427757263 - f1-score (micro avg) 0.7981
2023-10-25 17:21:39,378 ----------------------------------------------------------------------------------------------------
2023-10-25 17:21:44,817 epoch 10 - iter 89/893 - loss 0.00613679 - time (sec): 5.44 - samples/sec: 4461.62 - lr: 0.000005 - momentum: 0.000000
2023-10-25 17:21:50,598 epoch 10 - iter 178/893 - loss 0.00627718 - time (sec): 11.22 - samples/sec: 4222.26 - lr: 0.000004 - momentum: 0.000000
2023-10-25 17:21:56,574 epoch 10 - iter 267/893 - loss 0.00759554 - time (sec): 17.19 - samples/sec: 4268.40 - lr: 0.000004 - momentum: 0.000000
2023-10-25 17:22:02,673 epoch 10 - iter 356/893 - loss 0.00807461 - time (sec): 23.29 - samples/sec: 4232.88 - lr: 0.000003 - momentum: 0.000000
2023-10-25 17:22:08,727 epoch 10 - iter 445/893 - loss 0.00789473 - time (sec): 29.35 - samples/sec: 4155.31 - lr: 0.000003 - momentum: 0.000000
2023-10-25 17:22:14,934 epoch 10 - iter 534/893 - loss 0.00783961 - time (sec): 35.55 - samples/sec: 4162.38 - lr: 0.000002 - momentum: 0.000000
2023-10-25 17:22:20,987 epoch 10 - iter 623/893 - loss 0.00764283 - time (sec): 41.61 - samples/sec: 4158.54 - lr: 0.000002 - momentum: 0.000000
2023-10-25 17:22:26,725 epoch 10 - iter 712/893 - loss 0.00703458 - time (sec): 47.35 - samples/sec: 4144.99 - lr: 0.000001 - momentum: 0.000000
2023-10-25 17:22:32,754 epoch 10 - iter 801/893 - loss 0.00689931 - time (sec): 53.37 - samples/sec: 4157.08 - lr: 0.000001 - momentum: 0.000000
2023-10-25 17:22:38,960 epoch 10 - iter 890/893 - loss 0.00666490 - time (sec): 59.58 - samples/sec: 4161.77 - lr: 0.000000 - momentum: 0.000000
2023-10-25 17:22:39,141 ----------------------------------------------------------------------------------------------------
2023-10-25 17:22:39,141 EPOCH 10 done: loss 0.0066 - lr: 0.000000
2023-10-25 17:22:43,875 DEV : loss 0.23105905950069427 - f1-score (micro avg) 0.8
2023-10-25 17:22:44,387 ----------------------------------------------------------------------------------------------------
2023-10-25 17:22:44,388 Loading model from best epoch ...
2023-10-25 17:22:46,202 SequenceTagger predicts: Dictionary with 17 tags: O, S-PER, B-PER, E-PER, I-PER, S-LOC, B-LOC, E-LOC, I-LOC, S-ORG, B-ORG, E-ORG, I-ORG, S-HumanProd, B-HumanProd, E-HumanProd, I-HumanProd
2023-10-25 17:22:58,838
Results:
- F-score (micro) 0.6773
- F-score (macro) 0.588
- Accuracy 0.5304
By class:
precision recall f1-score support
LOC 0.6839 0.6877 0.6858 1095
PER 0.7644 0.7500 0.7571 1012
ORG 0.4379 0.5434 0.4850 357
HumanProd 0.3182 0.6364 0.4242 33
micro avg 0.6635 0.6916 0.6773 2497
macro avg 0.5511 0.6544 0.5880 2497
weighted avg 0.6765 0.6916 0.6825 2497
2023-10-25 17:22:58,839 ----------------------------------------------------------------------------------------------------
|