File size: 24,165 Bytes
d8c6207 |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 |
2023-10-18 22:51:24,698 ----------------------------------------------------------------------------------------------------
2023-10-18 22:51:24,698 Model: "SequenceTagger(
(embeddings): TransformerWordEmbeddings(
(model): BertModel(
(embeddings): BertEmbeddings(
(word_embeddings): Embedding(32001, 128)
(position_embeddings): Embedding(512, 128)
(token_type_embeddings): Embedding(2, 128)
(LayerNorm): LayerNorm((128,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(encoder): BertEncoder(
(layer): ModuleList(
(0-1): 2 x BertLayer(
(attention): BertAttention(
(self): BertSelfAttention(
(query): Linear(in_features=128, out_features=128, bias=True)
(key): Linear(in_features=128, out_features=128, bias=True)
(value): Linear(in_features=128, out_features=128, bias=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(output): BertSelfOutput(
(dense): Linear(in_features=128, out_features=128, bias=True)
(LayerNorm): LayerNorm((128,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
(intermediate): BertIntermediate(
(dense): Linear(in_features=128, out_features=512, bias=True)
(intermediate_act_fn): GELUActivation()
)
(output): BertOutput(
(dense): Linear(in_features=512, out_features=128, bias=True)
(LayerNorm): LayerNorm((128,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
)
(pooler): BertPooler(
(dense): Linear(in_features=128, out_features=128, bias=True)
(activation): Tanh()
)
)
)
(locked_dropout): LockedDropout(p=0.5)
(linear): Linear(in_features=128, out_features=13, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-18 22:51:24,698 ----------------------------------------------------------------------------------------------------
2023-10-18 22:51:24,699 MultiCorpus: 5777 train + 722 dev + 723 test sentences
- NER_ICDAR_EUROPEANA Corpus: 5777 train + 722 dev + 723 test sentences - /root/.flair/datasets/ner_icdar_europeana/nl
2023-10-18 22:51:24,699 ----------------------------------------------------------------------------------------------------
2023-10-18 22:51:24,699 Train: 5777 sentences
2023-10-18 22:51:24,699 (train_with_dev=False, train_with_test=False)
2023-10-18 22:51:24,699 ----------------------------------------------------------------------------------------------------
2023-10-18 22:51:24,699 Training Params:
2023-10-18 22:51:24,699 - learning_rate: "5e-05"
2023-10-18 22:51:24,699 - mini_batch_size: "4"
2023-10-18 22:51:24,699 - max_epochs: "10"
2023-10-18 22:51:24,699 - shuffle: "True"
2023-10-18 22:51:24,699 ----------------------------------------------------------------------------------------------------
2023-10-18 22:51:24,699 Plugins:
2023-10-18 22:51:24,699 - TensorboardLogger
2023-10-18 22:51:24,699 - LinearScheduler | warmup_fraction: '0.1'
2023-10-18 22:51:24,699 ----------------------------------------------------------------------------------------------------
2023-10-18 22:51:24,699 Final evaluation on model from best epoch (best-model.pt)
2023-10-18 22:51:24,699 - metric: "('micro avg', 'f1-score')"
2023-10-18 22:51:24,699 ----------------------------------------------------------------------------------------------------
2023-10-18 22:51:24,699 Computation:
2023-10-18 22:51:24,699 - compute on device: cuda:0
2023-10-18 22:51:24,699 - embedding storage: none
2023-10-18 22:51:24,699 ----------------------------------------------------------------------------------------------------
2023-10-18 22:51:24,699 Model training base path: "hmbench-icdar/nl-dbmdz/bert-tiny-historic-multilingual-cased-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-4"
2023-10-18 22:51:24,699 ----------------------------------------------------------------------------------------------------
2023-10-18 22:51:24,699 ----------------------------------------------------------------------------------------------------
2023-10-18 22:51:24,699 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-18 22:51:27,059 epoch 1 - iter 144/1445 - loss 3.29539325 - time (sec): 2.36 - samples/sec: 6871.85 - lr: 0.000005 - momentum: 0.000000
2023-10-18 22:51:29,599 epoch 1 - iter 288/1445 - loss 2.80667047 - time (sec): 4.90 - samples/sec: 6851.93 - lr: 0.000010 - momentum: 0.000000
2023-10-18 22:51:32,037 epoch 1 - iter 432/1445 - loss 2.17298777 - time (sec): 7.34 - samples/sec: 7006.80 - lr: 0.000015 - momentum: 0.000000
2023-10-18 22:51:34,467 epoch 1 - iter 576/1445 - loss 1.69993643 - time (sec): 9.77 - samples/sec: 7136.03 - lr: 0.000020 - momentum: 0.000000
2023-10-18 22:51:36,827 epoch 1 - iter 720/1445 - loss 1.42956310 - time (sec): 12.13 - samples/sec: 7137.99 - lr: 0.000025 - momentum: 0.000000
2023-10-18 22:51:39,163 epoch 1 - iter 864/1445 - loss 1.24708491 - time (sec): 14.46 - samples/sec: 7154.15 - lr: 0.000030 - momentum: 0.000000
2023-10-18 22:51:41,607 epoch 1 - iter 1008/1445 - loss 1.11449457 - time (sec): 16.91 - samples/sec: 7121.10 - lr: 0.000035 - momentum: 0.000000
2023-10-18 22:51:44,080 epoch 1 - iter 1152/1445 - loss 1.00767761 - time (sec): 19.38 - samples/sec: 7180.56 - lr: 0.000040 - momentum: 0.000000
2023-10-18 22:51:46,593 epoch 1 - iter 1296/1445 - loss 0.91487367 - time (sec): 21.89 - samples/sec: 7218.53 - lr: 0.000045 - momentum: 0.000000
2023-10-18 22:51:49,002 epoch 1 - iter 1440/1445 - loss 0.84936645 - time (sec): 24.30 - samples/sec: 7229.17 - lr: 0.000050 - momentum: 0.000000
2023-10-18 22:51:49,078 ----------------------------------------------------------------------------------------------------
2023-10-18 22:51:49,079 EPOCH 1 done: loss 0.8477 - lr: 0.000050
2023-10-18 22:51:50,368 DEV : loss 0.28886058926582336 - f1-score (micro avg) 0.0363
2023-10-18 22:51:50,382 saving best model
2023-10-18 22:51:50,413 ----------------------------------------------------------------------------------------------------
2023-10-18 22:51:52,534 epoch 2 - iter 144/1445 - loss 0.20771123 - time (sec): 2.12 - samples/sec: 8378.68 - lr: 0.000049 - momentum: 0.000000
2023-10-18 22:51:54,705 epoch 2 - iter 288/1445 - loss 0.21302193 - time (sec): 4.29 - samples/sec: 8059.18 - lr: 0.000049 - momentum: 0.000000
2023-10-18 22:51:57,097 epoch 2 - iter 432/1445 - loss 0.21308278 - time (sec): 6.68 - samples/sec: 7668.37 - lr: 0.000048 - momentum: 0.000000
2023-10-18 22:51:59,685 epoch 2 - iter 576/1445 - loss 0.20687075 - time (sec): 9.27 - samples/sec: 7600.92 - lr: 0.000048 - momentum: 0.000000
2023-10-18 22:52:02,218 epoch 2 - iter 720/1445 - loss 0.20244105 - time (sec): 11.80 - samples/sec: 7393.23 - lr: 0.000047 - momentum: 0.000000
2023-10-18 22:52:04,652 epoch 2 - iter 864/1445 - loss 0.19886336 - time (sec): 14.24 - samples/sec: 7394.02 - lr: 0.000047 - momentum: 0.000000
2023-10-18 22:52:06,878 epoch 2 - iter 1008/1445 - loss 0.19645336 - time (sec): 16.46 - samples/sec: 7481.26 - lr: 0.000046 - momentum: 0.000000
2023-10-18 22:52:09,110 epoch 2 - iter 1152/1445 - loss 0.20134357 - time (sec): 18.70 - samples/sec: 7488.90 - lr: 0.000046 - momentum: 0.000000
2023-10-18 22:52:11,497 epoch 2 - iter 1296/1445 - loss 0.19916074 - time (sec): 21.08 - samples/sec: 7494.59 - lr: 0.000045 - momentum: 0.000000
2023-10-18 22:52:13,935 epoch 2 - iter 1440/1445 - loss 0.19916757 - time (sec): 23.52 - samples/sec: 7459.46 - lr: 0.000044 - momentum: 0.000000
2023-10-18 22:52:14,018 ----------------------------------------------------------------------------------------------------
2023-10-18 22:52:14,019 EPOCH 2 done: loss 0.1991 - lr: 0.000044
2023-10-18 22:52:15,801 DEV : loss 0.2434382140636444 - f1-score (micro avg) 0.3074
2023-10-18 22:52:15,815 saving best model
2023-10-18 22:52:15,849 ----------------------------------------------------------------------------------------------------
2023-10-18 22:52:18,303 epoch 3 - iter 144/1445 - loss 0.16824685 - time (sec): 2.45 - samples/sec: 7641.85 - lr: 0.000044 - momentum: 0.000000
2023-10-18 22:52:20,724 epoch 3 - iter 288/1445 - loss 0.17832135 - time (sec): 4.87 - samples/sec: 7722.80 - lr: 0.000043 - momentum: 0.000000
2023-10-18 22:52:23,482 epoch 3 - iter 432/1445 - loss 0.17388072 - time (sec): 7.63 - samples/sec: 7228.05 - lr: 0.000043 - momentum: 0.000000
2023-10-18 22:52:25,913 epoch 3 - iter 576/1445 - loss 0.16949600 - time (sec): 10.06 - samples/sec: 7262.71 - lr: 0.000042 - momentum: 0.000000
2023-10-18 22:52:28,376 epoch 3 - iter 720/1445 - loss 0.17010711 - time (sec): 12.53 - samples/sec: 7214.37 - lr: 0.000042 - momentum: 0.000000
2023-10-18 22:52:30,724 epoch 3 - iter 864/1445 - loss 0.16944462 - time (sec): 14.87 - samples/sec: 7161.25 - lr: 0.000041 - momentum: 0.000000
2023-10-18 22:52:33,038 epoch 3 - iter 1008/1445 - loss 0.17031912 - time (sec): 17.19 - samples/sec: 7241.37 - lr: 0.000041 - momentum: 0.000000
2023-10-18 22:52:35,207 epoch 3 - iter 1152/1445 - loss 0.16944422 - time (sec): 19.36 - samples/sec: 7297.31 - lr: 0.000040 - momentum: 0.000000
2023-10-18 22:52:37,602 epoch 3 - iter 1296/1445 - loss 0.16674107 - time (sec): 21.75 - samples/sec: 7316.34 - lr: 0.000039 - momentum: 0.000000
2023-10-18 22:52:39,958 epoch 3 - iter 1440/1445 - loss 0.16647381 - time (sec): 24.11 - samples/sec: 7278.75 - lr: 0.000039 - momentum: 0.000000
2023-10-18 22:52:40,039 ----------------------------------------------------------------------------------------------------
2023-10-18 22:52:40,039 EPOCH 3 done: loss 0.1663 - lr: 0.000039
2023-10-18 22:52:41,821 DEV : loss 0.21055997908115387 - f1-score (micro avg) 0.4219
2023-10-18 22:52:41,835 saving best model
2023-10-18 22:52:41,870 ----------------------------------------------------------------------------------------------------
2023-10-18 22:52:44,200 epoch 4 - iter 144/1445 - loss 0.15283862 - time (sec): 2.33 - samples/sec: 7307.47 - lr: 0.000038 - momentum: 0.000000
2023-10-18 22:52:46,578 epoch 4 - iter 288/1445 - loss 0.14799565 - time (sec): 4.71 - samples/sec: 7015.01 - lr: 0.000038 - momentum: 0.000000
2023-10-18 22:52:48,944 epoch 4 - iter 432/1445 - loss 0.15105169 - time (sec): 7.07 - samples/sec: 7106.38 - lr: 0.000037 - momentum: 0.000000
2023-10-18 22:52:51,342 epoch 4 - iter 576/1445 - loss 0.15217093 - time (sec): 9.47 - samples/sec: 7174.33 - lr: 0.000037 - momentum: 0.000000
2023-10-18 22:52:53,783 epoch 4 - iter 720/1445 - loss 0.15166536 - time (sec): 11.91 - samples/sec: 7162.93 - lr: 0.000036 - momentum: 0.000000
2023-10-18 22:52:56,210 epoch 4 - iter 864/1445 - loss 0.15237011 - time (sec): 14.34 - samples/sec: 7158.14 - lr: 0.000036 - momentum: 0.000000
2023-10-18 22:52:58,867 epoch 4 - iter 1008/1445 - loss 0.15079524 - time (sec): 17.00 - samples/sec: 7087.84 - lr: 0.000035 - momentum: 0.000000
2023-10-18 22:53:01,592 epoch 4 - iter 1152/1445 - loss 0.15204078 - time (sec): 19.72 - samples/sec: 7121.06 - lr: 0.000034 - momentum: 0.000000
2023-10-18 22:53:03,966 epoch 4 - iter 1296/1445 - loss 0.15103163 - time (sec): 22.10 - samples/sec: 7125.45 - lr: 0.000034 - momentum: 0.000000
2023-10-18 22:53:06,380 epoch 4 - iter 1440/1445 - loss 0.15143506 - time (sec): 24.51 - samples/sec: 7168.62 - lr: 0.000033 - momentum: 0.000000
2023-10-18 22:53:06,461 ----------------------------------------------------------------------------------------------------
2023-10-18 22:53:06,461 EPOCH 4 done: loss 0.1515 - lr: 0.000033
2023-10-18 22:53:08,244 DEV : loss 0.19626548886299133 - f1-score (micro avg) 0.4869
2023-10-18 22:53:08,258 saving best model
2023-10-18 22:53:08,293 ----------------------------------------------------------------------------------------------------
2023-10-18 22:53:10,756 epoch 5 - iter 144/1445 - loss 0.14791715 - time (sec): 2.46 - samples/sec: 7246.53 - lr: 0.000033 - momentum: 0.000000
2023-10-18 22:53:13,168 epoch 5 - iter 288/1445 - loss 0.14595192 - time (sec): 4.87 - samples/sec: 7399.69 - lr: 0.000032 - momentum: 0.000000
2023-10-18 22:53:15,540 epoch 5 - iter 432/1445 - loss 0.14209987 - time (sec): 7.25 - samples/sec: 7414.07 - lr: 0.000032 - momentum: 0.000000
2023-10-18 22:53:17,946 epoch 5 - iter 576/1445 - loss 0.14142116 - time (sec): 9.65 - samples/sec: 7301.39 - lr: 0.000031 - momentum: 0.000000
2023-10-18 22:53:20,291 epoch 5 - iter 720/1445 - loss 0.13858369 - time (sec): 12.00 - samples/sec: 7226.89 - lr: 0.000031 - momentum: 0.000000
2023-10-18 22:53:22,707 epoch 5 - iter 864/1445 - loss 0.13944449 - time (sec): 14.41 - samples/sec: 7219.53 - lr: 0.000030 - momentum: 0.000000
2023-10-18 22:53:25,161 epoch 5 - iter 1008/1445 - loss 0.13904087 - time (sec): 16.87 - samples/sec: 7204.05 - lr: 0.000029 - momentum: 0.000000
2023-10-18 22:53:27,753 epoch 5 - iter 1152/1445 - loss 0.13936345 - time (sec): 19.46 - samples/sec: 7198.86 - lr: 0.000029 - momentum: 0.000000
2023-10-18 22:53:30,106 epoch 5 - iter 1296/1445 - loss 0.13877879 - time (sec): 21.81 - samples/sec: 7220.31 - lr: 0.000028 - momentum: 0.000000
2023-10-18 22:53:32,573 epoch 5 - iter 1440/1445 - loss 0.13778394 - time (sec): 24.28 - samples/sec: 7245.29 - lr: 0.000028 - momentum: 0.000000
2023-10-18 22:53:32,646 ----------------------------------------------------------------------------------------------------
2023-10-18 22:53:32,646 EPOCH 5 done: loss 0.1378 - lr: 0.000028
2023-10-18 22:53:34,776 DEV : loss 0.186900332570076 - f1-score (micro avg) 0.5261
2023-10-18 22:53:34,792 saving best model
2023-10-18 22:53:34,828 ----------------------------------------------------------------------------------------------------
2023-10-18 22:53:37,198 epoch 6 - iter 144/1445 - loss 0.12941650 - time (sec): 2.37 - samples/sec: 6979.43 - lr: 0.000027 - momentum: 0.000000
2023-10-18 22:53:39,571 epoch 6 - iter 288/1445 - loss 0.12211079 - time (sec): 4.74 - samples/sec: 7141.82 - lr: 0.000027 - momentum: 0.000000
2023-10-18 22:53:41,959 epoch 6 - iter 432/1445 - loss 0.12637372 - time (sec): 7.13 - samples/sec: 7336.97 - lr: 0.000026 - momentum: 0.000000
2023-10-18 22:53:44,418 epoch 6 - iter 576/1445 - loss 0.12653560 - time (sec): 9.59 - samples/sec: 7368.21 - lr: 0.000026 - momentum: 0.000000
2023-10-18 22:53:46,769 epoch 6 - iter 720/1445 - loss 0.12839698 - time (sec): 11.94 - samples/sec: 7326.25 - lr: 0.000025 - momentum: 0.000000
2023-10-18 22:53:49,154 epoch 6 - iter 864/1445 - loss 0.12535563 - time (sec): 14.32 - samples/sec: 7325.47 - lr: 0.000024 - momentum: 0.000000
2023-10-18 22:53:51,561 epoch 6 - iter 1008/1445 - loss 0.12897091 - time (sec): 16.73 - samples/sec: 7261.16 - lr: 0.000024 - momentum: 0.000000
2023-10-18 22:53:53,981 epoch 6 - iter 1152/1445 - loss 0.12610695 - time (sec): 19.15 - samples/sec: 7266.88 - lr: 0.000023 - momentum: 0.000000
2023-10-18 22:53:56,366 epoch 6 - iter 1296/1445 - loss 0.12806517 - time (sec): 21.54 - samples/sec: 7280.58 - lr: 0.000023 - momentum: 0.000000
2023-10-18 22:53:58,957 epoch 6 - iter 1440/1445 - loss 0.12929665 - time (sec): 24.13 - samples/sec: 7283.06 - lr: 0.000022 - momentum: 0.000000
2023-10-18 22:53:59,044 ----------------------------------------------------------------------------------------------------
2023-10-18 22:53:59,044 EPOCH 6 done: loss 0.1292 - lr: 0.000022
2023-10-18 22:54:00,808 DEV : loss 0.1867137998342514 - f1-score (micro avg) 0.5326
2023-10-18 22:54:00,822 saving best model
2023-10-18 22:54:00,858 ----------------------------------------------------------------------------------------------------
2023-10-18 22:54:03,301 epoch 7 - iter 144/1445 - loss 0.11727356 - time (sec): 2.44 - samples/sec: 6746.22 - lr: 0.000022 - momentum: 0.000000
2023-10-18 22:54:05,687 epoch 7 - iter 288/1445 - loss 0.11920637 - time (sec): 4.83 - samples/sec: 7110.68 - lr: 0.000021 - momentum: 0.000000
2023-10-18 22:54:08,078 epoch 7 - iter 432/1445 - loss 0.12142107 - time (sec): 7.22 - samples/sec: 7086.33 - lr: 0.000021 - momentum: 0.000000
2023-10-18 22:54:10,549 epoch 7 - iter 576/1445 - loss 0.12054887 - time (sec): 9.69 - samples/sec: 7176.58 - lr: 0.000020 - momentum: 0.000000
2023-10-18 22:54:12,850 epoch 7 - iter 720/1445 - loss 0.12121286 - time (sec): 11.99 - samples/sec: 7219.02 - lr: 0.000019 - momentum: 0.000000
2023-10-18 22:54:15,406 epoch 7 - iter 864/1445 - loss 0.12151057 - time (sec): 14.55 - samples/sec: 7155.32 - lr: 0.000019 - momentum: 0.000000
2023-10-18 22:54:17,870 epoch 7 - iter 1008/1445 - loss 0.12126598 - time (sec): 17.01 - samples/sec: 7193.07 - lr: 0.000018 - momentum: 0.000000
2023-10-18 22:54:20,281 epoch 7 - iter 1152/1445 - loss 0.12219795 - time (sec): 19.42 - samples/sec: 7208.87 - lr: 0.000018 - momentum: 0.000000
2023-10-18 22:54:22,749 epoch 7 - iter 1296/1445 - loss 0.12311442 - time (sec): 21.89 - samples/sec: 7217.81 - lr: 0.000017 - momentum: 0.000000
2023-10-18 22:54:25,139 epoch 7 - iter 1440/1445 - loss 0.12171338 - time (sec): 24.28 - samples/sec: 7233.22 - lr: 0.000017 - momentum: 0.000000
2023-10-18 22:54:25,219 ----------------------------------------------------------------------------------------------------
2023-10-18 22:54:25,220 EPOCH 7 done: loss 0.1216 - lr: 0.000017
2023-10-18 22:54:26,988 DEV : loss 0.19099119305610657 - f1-score (micro avg) 0.5487
2023-10-18 22:54:27,003 saving best model
2023-10-18 22:54:27,040 ----------------------------------------------------------------------------------------------------
2023-10-18 22:54:29,347 epoch 8 - iter 144/1445 - loss 0.13909632 - time (sec): 2.31 - samples/sec: 7149.79 - lr: 0.000016 - momentum: 0.000000
2023-10-18 22:54:31,777 epoch 8 - iter 288/1445 - loss 0.12875856 - time (sec): 4.74 - samples/sec: 7333.26 - lr: 0.000016 - momentum: 0.000000
2023-10-18 22:54:34,165 epoch 8 - iter 432/1445 - loss 0.12665044 - time (sec): 7.13 - samples/sec: 7443.17 - lr: 0.000015 - momentum: 0.000000
2023-10-18 22:54:36,574 epoch 8 - iter 576/1445 - loss 0.12517964 - time (sec): 9.53 - samples/sec: 7321.07 - lr: 0.000014 - momentum: 0.000000
2023-10-18 22:54:38,887 epoch 8 - iter 720/1445 - loss 0.12049045 - time (sec): 11.85 - samples/sec: 7417.55 - lr: 0.000014 - momentum: 0.000000
2023-10-18 22:54:41,339 epoch 8 - iter 864/1445 - loss 0.11769624 - time (sec): 14.30 - samples/sec: 7392.12 - lr: 0.000013 - momentum: 0.000000
2023-10-18 22:54:43,795 epoch 8 - iter 1008/1445 - loss 0.11773878 - time (sec): 16.76 - samples/sec: 7377.68 - lr: 0.000013 - momentum: 0.000000
2023-10-18 22:54:46,163 epoch 8 - iter 1152/1445 - loss 0.11579635 - time (sec): 19.12 - samples/sec: 7331.13 - lr: 0.000012 - momentum: 0.000000
2023-10-18 22:54:48,494 epoch 8 - iter 1296/1445 - loss 0.11660615 - time (sec): 21.45 - samples/sec: 7351.02 - lr: 0.000012 - momentum: 0.000000
2023-10-18 22:54:50,905 epoch 8 - iter 1440/1445 - loss 0.11664721 - time (sec): 23.86 - samples/sec: 7365.31 - lr: 0.000011 - momentum: 0.000000
2023-10-18 22:54:50,980 ----------------------------------------------------------------------------------------------------
2023-10-18 22:54:50,980 EPOCH 8 done: loss 0.1166 - lr: 0.000011
2023-10-18 22:54:53,066 DEV : loss 0.19956204295158386 - f1-score (micro avg) 0.5434
2023-10-18 22:54:53,080 ----------------------------------------------------------------------------------------------------
2023-10-18 22:54:55,471 epoch 9 - iter 144/1445 - loss 0.11520321 - time (sec): 2.39 - samples/sec: 7521.67 - lr: 0.000011 - momentum: 0.000000
2023-10-18 22:54:57,898 epoch 9 - iter 288/1445 - loss 0.12001827 - time (sec): 4.82 - samples/sec: 7468.77 - lr: 0.000010 - momentum: 0.000000
2023-10-18 22:55:00,238 epoch 9 - iter 432/1445 - loss 0.11044915 - time (sec): 7.16 - samples/sec: 7365.68 - lr: 0.000009 - momentum: 0.000000
2023-10-18 22:55:02,679 epoch 9 - iter 576/1445 - loss 0.10805370 - time (sec): 9.60 - samples/sec: 7325.82 - lr: 0.000009 - momentum: 0.000000
2023-10-18 22:55:05,091 epoch 9 - iter 720/1445 - loss 0.11126004 - time (sec): 12.01 - samples/sec: 7351.91 - lr: 0.000008 - momentum: 0.000000
2023-10-18 22:55:07,481 epoch 9 - iter 864/1445 - loss 0.11107034 - time (sec): 14.40 - samples/sec: 7406.58 - lr: 0.000008 - momentum: 0.000000
2023-10-18 22:55:09,864 epoch 9 - iter 1008/1445 - loss 0.11224647 - time (sec): 16.78 - samples/sec: 7416.01 - lr: 0.000007 - momentum: 0.000000
2023-10-18 22:55:12,182 epoch 9 - iter 1152/1445 - loss 0.11351795 - time (sec): 19.10 - samples/sec: 7395.91 - lr: 0.000007 - momentum: 0.000000
2023-10-18 22:55:14,679 epoch 9 - iter 1296/1445 - loss 0.11382066 - time (sec): 21.60 - samples/sec: 7353.71 - lr: 0.000006 - momentum: 0.000000
2023-10-18 22:55:17,000 epoch 9 - iter 1440/1445 - loss 0.11288922 - time (sec): 23.92 - samples/sec: 7345.54 - lr: 0.000006 - momentum: 0.000000
2023-10-18 22:55:17,074 ----------------------------------------------------------------------------------------------------
2023-10-18 22:55:17,075 EPOCH 9 done: loss 0.1130 - lr: 0.000006
2023-10-18 22:55:18,856 DEV : loss 0.19409048557281494 - f1-score (micro avg) 0.5636
2023-10-18 22:55:18,870 saving best model
2023-10-18 22:55:18,907 ----------------------------------------------------------------------------------------------------
2023-10-18 22:55:21,246 epoch 10 - iter 144/1445 - loss 0.11919407 - time (sec): 2.34 - samples/sec: 7369.04 - lr: 0.000005 - momentum: 0.000000
2023-10-18 22:55:23,624 epoch 10 - iter 288/1445 - loss 0.12311079 - time (sec): 4.72 - samples/sec: 7239.49 - lr: 0.000004 - momentum: 0.000000
2023-10-18 22:55:26,056 epoch 10 - iter 432/1445 - loss 0.12140389 - time (sec): 7.15 - samples/sec: 7266.09 - lr: 0.000004 - momentum: 0.000000
2023-10-18 22:55:28,507 epoch 10 - iter 576/1445 - loss 0.11657262 - time (sec): 9.60 - samples/sec: 7407.10 - lr: 0.000003 - momentum: 0.000000
2023-10-18 22:55:30,961 epoch 10 - iter 720/1445 - loss 0.11243952 - time (sec): 12.05 - samples/sec: 7385.04 - lr: 0.000003 - momentum: 0.000000
2023-10-18 22:55:33,205 epoch 10 - iter 864/1445 - loss 0.11354347 - time (sec): 14.30 - samples/sec: 7387.47 - lr: 0.000002 - momentum: 0.000000
2023-10-18 22:55:35,316 epoch 10 - iter 1008/1445 - loss 0.11085507 - time (sec): 16.41 - samples/sec: 7513.69 - lr: 0.000002 - momentum: 0.000000
2023-10-18 22:55:37,805 epoch 10 - iter 1152/1445 - loss 0.10959603 - time (sec): 18.90 - samples/sec: 7503.00 - lr: 0.000001 - momentum: 0.000000
2023-10-18 22:55:40,166 epoch 10 - iter 1296/1445 - loss 0.11097144 - time (sec): 21.26 - samples/sec: 7447.61 - lr: 0.000001 - momentum: 0.000000
2023-10-18 22:55:42,632 epoch 10 - iter 1440/1445 - loss 0.11183179 - time (sec): 23.72 - samples/sec: 7402.41 - lr: 0.000000 - momentum: 0.000000
2023-10-18 22:55:42,711 ----------------------------------------------------------------------------------------------------
2023-10-18 22:55:42,712 EPOCH 10 done: loss 0.1119 - lr: 0.000000
2023-10-18 22:55:44,486 DEV : loss 0.19939179718494415 - f1-score (micro avg) 0.5604
2023-10-18 22:55:44,530 ----------------------------------------------------------------------------------------------------
2023-10-18 22:55:44,530 Loading model from best epoch ...
2023-10-18 22:55:44,613 SequenceTagger predicts: Dictionary with 13 tags: O, S-LOC, B-LOC, E-LOC, I-LOC, S-PER, B-PER, E-PER, I-PER, S-ORG, B-ORG, E-ORG, I-ORG
2023-10-18 22:55:45,906
Results:
- F-score (micro) 0.5542
- F-score (macro) 0.392
- Accuracy 0.393
By class:
precision recall f1-score support
LOC 0.6227 0.6594 0.6405 458
PER 0.5196 0.4959 0.5074 482
ORG 0.5000 0.0145 0.0282 69
micro avg 0.5723 0.5372 0.5542 1009
macro avg 0.5474 0.3899 0.3920 1009
weighted avg 0.5650 0.5372 0.5351 1009
2023-10-18 22:55:45,907 ----------------------------------------------------------------------------------------------------
|