File size: 23,938 Bytes
4bbb6d6 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 |
2023-10-25 21:22:49,010 ----------------------------------------------------------------------------------------------------
2023-10-25 21:22:49,011 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 21:22:49,011 ----------------------------------------------------------------------------------------------------
2023-10-25 21:22:49,012 MultiCorpus: 1166 train + 165 dev + 415 test sentences
- NER_HIPE_2022 Corpus: 1166 train + 165 dev + 415 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/newseye/fi/with_doc_seperator
2023-10-25 21:22:49,012 ----------------------------------------------------------------------------------------------------
2023-10-25 21:22:49,012 Train: 1166 sentences
2023-10-25 21:22:49,012 (train_with_dev=False, train_with_test=False)
2023-10-25 21:22:49,012 ----------------------------------------------------------------------------------------------------
2023-10-25 21:22:49,012 Training Params:
2023-10-25 21:22:49,012 - learning_rate: "5e-05"
2023-10-25 21:22:49,012 - mini_batch_size: "4"
2023-10-25 21:22:49,012 - max_epochs: "10"
2023-10-25 21:22:49,012 - shuffle: "True"
2023-10-25 21:22:49,012 ----------------------------------------------------------------------------------------------------
2023-10-25 21:22:49,012 Plugins:
2023-10-25 21:22:49,012 - TensorboardLogger
2023-10-25 21:22:49,012 - LinearScheduler | warmup_fraction: '0.1'
2023-10-25 21:22:49,012 ----------------------------------------------------------------------------------------------------
2023-10-25 21:22:49,012 Final evaluation on model from best epoch (best-model.pt)
2023-10-25 21:22:49,012 - metric: "('micro avg', 'f1-score')"
2023-10-25 21:22:49,012 ----------------------------------------------------------------------------------------------------
2023-10-25 21:22:49,012 Computation:
2023-10-25 21:22:49,012 - compute on device: cuda:0
2023-10-25 21:22:49,012 - embedding storage: none
2023-10-25 21:22:49,012 ----------------------------------------------------------------------------------------------------
2023-10-25 21:22:49,012 Model training base path: "hmbench-newseye/fi-dbmdz/bert-base-historic-multilingual-64k-td-cased-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-4"
2023-10-25 21:22:49,012 ----------------------------------------------------------------------------------------------------
2023-10-25 21:22:49,012 ----------------------------------------------------------------------------------------------------
2023-10-25 21:22:49,012 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-25 21:22:50,365 epoch 1 - iter 29/292 - loss 3.04468690 - time (sec): 1.35 - samples/sec: 3544.24 - lr: 0.000005 - momentum: 0.000000
2023-10-25 21:22:51,616 epoch 1 - iter 58/292 - loss 2.08228056 - time (sec): 2.60 - samples/sec: 3418.55 - lr: 0.000010 - momentum: 0.000000
2023-10-25 21:22:52,870 epoch 1 - iter 87/292 - loss 1.67391242 - time (sec): 3.86 - samples/sec: 3380.18 - lr: 0.000015 - momentum: 0.000000
2023-10-25 21:22:54,119 epoch 1 - iter 116/292 - loss 1.39271840 - time (sec): 5.11 - samples/sec: 3361.81 - lr: 0.000020 - momentum: 0.000000
2023-10-25 21:22:55,408 epoch 1 - iter 145/292 - loss 1.19734159 - time (sec): 6.39 - samples/sec: 3286.22 - lr: 0.000025 - momentum: 0.000000
2023-10-25 21:22:56,810 epoch 1 - iter 174/292 - loss 1.01730119 - time (sec): 7.80 - samples/sec: 3367.82 - lr: 0.000030 - momentum: 0.000000
2023-10-25 21:22:58,079 epoch 1 - iter 203/292 - loss 0.91202992 - time (sec): 9.07 - samples/sec: 3355.88 - lr: 0.000035 - momentum: 0.000000
2023-10-25 21:22:59,525 epoch 1 - iter 232/292 - loss 0.82288451 - time (sec): 10.51 - samples/sec: 3320.22 - lr: 0.000040 - momentum: 0.000000
2023-10-25 21:23:00,832 epoch 1 - iter 261/292 - loss 0.74308263 - time (sec): 11.82 - samples/sec: 3348.87 - lr: 0.000045 - momentum: 0.000000
2023-10-25 21:23:02,179 epoch 1 - iter 290/292 - loss 0.68034745 - time (sec): 13.17 - samples/sec: 3354.10 - lr: 0.000049 - momentum: 0.000000
2023-10-25 21:23:02,263 ----------------------------------------------------------------------------------------------------
2023-10-25 21:23:02,263 EPOCH 1 done: loss 0.6780 - lr: 0.000049
2023-10-25 21:23:02,769 DEV : loss 0.15992717444896698 - f1-score (micro avg) 0.5378
2023-10-25 21:23:02,773 saving best model
2023-10-25 21:23:03,304 ----------------------------------------------------------------------------------------------------
2023-10-25 21:23:04,601 epoch 2 - iter 29/292 - loss 0.21214502 - time (sec): 1.30 - samples/sec: 3350.43 - lr: 0.000049 - momentum: 0.000000
2023-10-25 21:23:05,969 epoch 2 - iter 58/292 - loss 0.16803234 - time (sec): 2.66 - samples/sec: 3478.53 - lr: 0.000049 - momentum: 0.000000
2023-10-25 21:23:07,229 epoch 2 - iter 87/292 - loss 0.16062619 - time (sec): 3.92 - samples/sec: 3417.74 - lr: 0.000048 - momentum: 0.000000
2023-10-25 21:23:08,555 epoch 2 - iter 116/292 - loss 0.17011069 - time (sec): 5.25 - samples/sec: 3419.64 - lr: 0.000048 - momentum: 0.000000
2023-10-25 21:23:09,852 epoch 2 - iter 145/292 - loss 0.16743668 - time (sec): 6.55 - samples/sec: 3382.00 - lr: 0.000047 - momentum: 0.000000
2023-10-25 21:23:11,093 epoch 2 - iter 174/292 - loss 0.16331818 - time (sec): 7.79 - samples/sec: 3318.18 - lr: 0.000047 - momentum: 0.000000
2023-10-25 21:23:12,390 epoch 2 - iter 203/292 - loss 0.16521386 - time (sec): 9.08 - samples/sec: 3293.04 - lr: 0.000046 - momentum: 0.000000
2023-10-25 21:23:13,727 epoch 2 - iter 232/292 - loss 0.16386177 - time (sec): 10.42 - samples/sec: 3308.08 - lr: 0.000046 - momentum: 0.000000
2023-10-25 21:23:15,077 epoch 2 - iter 261/292 - loss 0.15723361 - time (sec): 11.77 - samples/sec: 3350.27 - lr: 0.000045 - momentum: 0.000000
2023-10-25 21:23:16,379 epoch 2 - iter 290/292 - loss 0.15379255 - time (sec): 13.07 - samples/sec: 3390.10 - lr: 0.000045 - momentum: 0.000000
2023-10-25 21:23:16,458 ----------------------------------------------------------------------------------------------------
2023-10-25 21:23:16,458 EPOCH 2 done: loss 0.1536 - lr: 0.000045
2023-10-25 21:23:17,366 DEV : loss 0.12357047200202942 - f1-score (micro avg) 0.7236
2023-10-25 21:23:17,370 saving best model
2023-10-25 21:23:17,907 ----------------------------------------------------------------------------------------------------
2023-10-25 21:23:19,208 epoch 3 - iter 29/292 - loss 0.08209900 - time (sec): 1.30 - samples/sec: 3698.78 - lr: 0.000044 - momentum: 0.000000
2023-10-25 21:23:20,480 epoch 3 - iter 58/292 - loss 0.08954013 - time (sec): 2.57 - samples/sec: 3585.69 - lr: 0.000043 - momentum: 0.000000
2023-10-25 21:23:21,942 epoch 3 - iter 87/292 - loss 0.09024601 - time (sec): 4.03 - samples/sec: 3363.73 - lr: 0.000043 - momentum: 0.000000
2023-10-25 21:23:23,235 epoch 3 - iter 116/292 - loss 0.09815526 - time (sec): 5.32 - samples/sec: 3316.48 - lr: 0.000042 - momentum: 0.000000
2023-10-25 21:23:24,526 epoch 3 - iter 145/292 - loss 0.10736630 - time (sec): 6.61 - samples/sec: 3329.89 - lr: 0.000042 - momentum: 0.000000
2023-10-25 21:23:25,765 epoch 3 - iter 174/292 - loss 0.10295479 - time (sec): 7.85 - samples/sec: 3267.84 - lr: 0.000041 - momentum: 0.000000
2023-10-25 21:23:27,043 epoch 3 - iter 203/292 - loss 0.09578494 - time (sec): 9.13 - samples/sec: 3300.02 - lr: 0.000041 - momentum: 0.000000
2023-10-25 21:23:28,435 epoch 3 - iter 232/292 - loss 0.09515875 - time (sec): 10.52 - samples/sec: 3296.33 - lr: 0.000040 - momentum: 0.000000
2023-10-25 21:23:29,808 epoch 3 - iter 261/292 - loss 0.09502063 - time (sec): 11.90 - samples/sec: 3318.84 - lr: 0.000040 - momentum: 0.000000
2023-10-25 21:23:31,136 epoch 3 - iter 290/292 - loss 0.09090155 - time (sec): 13.22 - samples/sec: 3350.86 - lr: 0.000039 - momentum: 0.000000
2023-10-25 21:23:31,223 ----------------------------------------------------------------------------------------------------
2023-10-25 21:23:31,223 EPOCH 3 done: loss 0.0908 - lr: 0.000039
2023-10-25 21:23:32,135 DEV : loss 0.11771436035633087 - f1-score (micro avg) 0.7456
2023-10-25 21:23:32,140 saving best model
2023-10-25 21:23:32,814 ----------------------------------------------------------------------------------------------------
2023-10-25 21:23:34,120 epoch 4 - iter 29/292 - loss 0.04447974 - time (sec): 1.30 - samples/sec: 3415.97 - lr: 0.000038 - momentum: 0.000000
2023-10-25 21:23:35,447 epoch 4 - iter 58/292 - loss 0.05748043 - time (sec): 2.63 - samples/sec: 3591.72 - lr: 0.000038 - momentum: 0.000000
2023-10-25 21:23:36,803 epoch 4 - iter 87/292 - loss 0.05839321 - time (sec): 3.99 - samples/sec: 3601.72 - lr: 0.000037 - momentum: 0.000000
2023-10-25 21:23:38,071 epoch 4 - iter 116/292 - loss 0.05373818 - time (sec): 5.25 - samples/sec: 3569.23 - lr: 0.000037 - momentum: 0.000000
2023-10-25 21:23:39,367 epoch 4 - iter 145/292 - loss 0.06079832 - time (sec): 6.55 - samples/sec: 3529.73 - lr: 0.000036 - momentum: 0.000000
2023-10-25 21:23:40,548 epoch 4 - iter 174/292 - loss 0.05840615 - time (sec): 7.73 - samples/sec: 3460.91 - lr: 0.000036 - momentum: 0.000000
2023-10-25 21:23:41,735 epoch 4 - iter 203/292 - loss 0.05922118 - time (sec): 8.92 - samples/sec: 3457.73 - lr: 0.000035 - momentum: 0.000000
2023-10-25 21:23:43,026 epoch 4 - iter 232/292 - loss 0.05817814 - time (sec): 10.21 - samples/sec: 3490.62 - lr: 0.000035 - momentum: 0.000000
2023-10-25 21:23:44,270 epoch 4 - iter 261/292 - loss 0.05811754 - time (sec): 11.45 - samples/sec: 3463.49 - lr: 0.000034 - momentum: 0.000000
2023-10-25 21:23:45,618 epoch 4 - iter 290/292 - loss 0.05637850 - time (sec): 12.80 - samples/sec: 3456.42 - lr: 0.000033 - momentum: 0.000000
2023-10-25 21:23:45,698 ----------------------------------------------------------------------------------------------------
2023-10-25 21:23:45,698 EPOCH 4 done: loss 0.0563 - lr: 0.000033
2023-10-25 21:23:46,613 DEV : loss 0.16891229152679443 - f1-score (micro avg) 0.7172
2023-10-25 21:23:46,618 ----------------------------------------------------------------------------------------------------
2023-10-25 21:23:48,047 epoch 5 - iter 29/292 - loss 0.01793138 - time (sec): 1.43 - samples/sec: 3282.09 - lr: 0.000033 - momentum: 0.000000
2023-10-25 21:23:49,387 epoch 5 - iter 58/292 - loss 0.03776489 - time (sec): 2.77 - samples/sec: 3374.60 - lr: 0.000032 - momentum: 0.000000
2023-10-25 21:23:50,675 epoch 5 - iter 87/292 - loss 0.03871292 - time (sec): 4.06 - samples/sec: 3416.95 - lr: 0.000032 - momentum: 0.000000
2023-10-25 21:23:51,995 epoch 5 - iter 116/292 - loss 0.03671836 - time (sec): 5.38 - samples/sec: 3461.52 - lr: 0.000031 - momentum: 0.000000
2023-10-25 21:23:53,267 epoch 5 - iter 145/292 - loss 0.03345817 - time (sec): 6.65 - samples/sec: 3532.80 - lr: 0.000031 - momentum: 0.000000
2023-10-25 21:23:54,488 epoch 5 - iter 174/292 - loss 0.03604099 - time (sec): 7.87 - samples/sec: 3450.29 - lr: 0.000030 - momentum: 0.000000
2023-10-25 21:23:55,717 epoch 5 - iter 203/292 - loss 0.03563773 - time (sec): 9.10 - samples/sec: 3475.36 - lr: 0.000030 - momentum: 0.000000
2023-10-25 21:23:56,883 epoch 5 - iter 232/292 - loss 0.03978405 - time (sec): 10.26 - samples/sec: 3450.93 - lr: 0.000029 - momentum: 0.000000
2023-10-25 21:23:58,127 epoch 5 - iter 261/292 - loss 0.03922902 - time (sec): 11.51 - samples/sec: 3449.17 - lr: 0.000028 - momentum: 0.000000
2023-10-25 21:23:59,426 epoch 5 - iter 290/292 - loss 0.04000375 - time (sec): 12.81 - samples/sec: 3454.07 - lr: 0.000028 - momentum: 0.000000
2023-10-25 21:23:59,515 ----------------------------------------------------------------------------------------------------
2023-10-25 21:23:59,515 EPOCH 5 done: loss 0.0398 - lr: 0.000028
2023-10-25 21:24:00,428 DEV : loss 0.12579147517681122 - f1-score (micro avg) 0.7352
2023-10-25 21:24:00,433 ----------------------------------------------------------------------------------------------------
2023-10-25 21:24:01,711 epoch 6 - iter 29/292 - loss 0.02611874 - time (sec): 1.28 - samples/sec: 3400.46 - lr: 0.000027 - momentum: 0.000000
2023-10-25 21:24:02,992 epoch 6 - iter 58/292 - loss 0.03086291 - time (sec): 2.56 - samples/sec: 3281.23 - lr: 0.000027 - momentum: 0.000000
2023-10-25 21:24:04,290 epoch 6 - iter 87/292 - loss 0.02541001 - time (sec): 3.86 - samples/sec: 3282.22 - lr: 0.000026 - momentum: 0.000000
2023-10-25 21:24:05,568 epoch 6 - iter 116/292 - loss 0.02774603 - time (sec): 5.13 - samples/sec: 3237.20 - lr: 0.000026 - momentum: 0.000000
2023-10-25 21:24:06,856 epoch 6 - iter 145/292 - loss 0.02771761 - time (sec): 6.42 - samples/sec: 3299.67 - lr: 0.000025 - momentum: 0.000000
2023-10-25 21:24:08,149 epoch 6 - iter 174/292 - loss 0.02765889 - time (sec): 7.71 - samples/sec: 3353.44 - lr: 0.000025 - momentum: 0.000000
2023-10-25 21:24:09,408 epoch 6 - iter 203/292 - loss 0.02925281 - time (sec): 8.97 - samples/sec: 3370.19 - lr: 0.000024 - momentum: 0.000000
2023-10-25 21:24:10,826 epoch 6 - iter 232/292 - loss 0.02864905 - time (sec): 10.39 - samples/sec: 3431.12 - lr: 0.000023 - momentum: 0.000000
2023-10-25 21:24:12,058 epoch 6 - iter 261/292 - loss 0.02763788 - time (sec): 11.62 - samples/sec: 3413.92 - lr: 0.000023 - momentum: 0.000000
2023-10-25 21:24:13,367 epoch 6 - iter 290/292 - loss 0.02581966 - time (sec): 12.93 - samples/sec: 3424.87 - lr: 0.000022 - momentum: 0.000000
2023-10-25 21:24:13,446 ----------------------------------------------------------------------------------------------------
2023-10-25 21:24:13,446 EPOCH 6 done: loss 0.0257 - lr: 0.000022
2023-10-25 21:24:14,365 DEV : loss 0.1791224181652069 - f1-score (micro avg) 0.7289
2023-10-25 21:24:14,370 ----------------------------------------------------------------------------------------------------
2023-10-25 21:24:15,661 epoch 7 - iter 29/292 - loss 0.01525348 - time (sec): 1.29 - samples/sec: 2948.18 - lr: 0.000022 - momentum: 0.000000
2023-10-25 21:24:16,980 epoch 7 - iter 58/292 - loss 0.02610841 - time (sec): 2.61 - samples/sec: 3107.31 - lr: 0.000021 - momentum: 0.000000
2023-10-25 21:24:18,373 epoch 7 - iter 87/292 - loss 0.02233332 - time (sec): 4.00 - samples/sec: 3326.21 - lr: 0.000021 - momentum: 0.000000
2023-10-25 21:24:19,793 epoch 7 - iter 116/292 - loss 0.01975879 - time (sec): 5.42 - samples/sec: 3136.62 - lr: 0.000020 - momentum: 0.000000
2023-10-25 21:24:21,105 epoch 7 - iter 145/292 - loss 0.02079285 - time (sec): 6.73 - samples/sec: 3140.38 - lr: 0.000020 - momentum: 0.000000
2023-10-25 21:24:22,435 epoch 7 - iter 174/292 - loss 0.01893185 - time (sec): 8.06 - samples/sec: 3231.57 - lr: 0.000019 - momentum: 0.000000
2023-10-25 21:24:23,760 epoch 7 - iter 203/292 - loss 0.01874262 - time (sec): 9.39 - samples/sec: 3280.58 - lr: 0.000018 - momentum: 0.000000
2023-10-25 21:24:25,035 epoch 7 - iter 232/292 - loss 0.01913688 - time (sec): 10.66 - samples/sec: 3308.42 - lr: 0.000018 - momentum: 0.000000
2023-10-25 21:24:26,341 epoch 7 - iter 261/292 - loss 0.01811495 - time (sec): 11.97 - samples/sec: 3278.83 - lr: 0.000017 - momentum: 0.000000
2023-10-25 21:24:27,679 epoch 7 - iter 290/292 - loss 0.01871243 - time (sec): 13.31 - samples/sec: 3314.90 - lr: 0.000017 - momentum: 0.000000
2023-10-25 21:24:27,767 ----------------------------------------------------------------------------------------------------
2023-10-25 21:24:27,768 EPOCH 7 done: loss 0.0196 - lr: 0.000017
2023-10-25 21:24:28,684 DEV : loss 0.18422643840312958 - f1-score (micro avg) 0.6875
2023-10-25 21:24:28,688 ----------------------------------------------------------------------------------------------------
2023-10-25 21:24:29,929 epoch 8 - iter 29/292 - loss 0.00351023 - time (sec): 1.24 - samples/sec: 3412.68 - lr: 0.000016 - momentum: 0.000000
2023-10-25 21:24:31,209 epoch 8 - iter 58/292 - loss 0.00912950 - time (sec): 2.52 - samples/sec: 3786.95 - lr: 0.000016 - momentum: 0.000000
2023-10-25 21:24:32,409 epoch 8 - iter 87/292 - loss 0.00811109 - time (sec): 3.72 - samples/sec: 3672.28 - lr: 0.000015 - momentum: 0.000000
2023-10-25 21:24:33,594 epoch 8 - iter 116/292 - loss 0.00685205 - time (sec): 4.90 - samples/sec: 3593.20 - lr: 0.000015 - momentum: 0.000000
2023-10-25 21:24:34,855 epoch 8 - iter 145/292 - loss 0.00786036 - time (sec): 6.17 - samples/sec: 3596.15 - lr: 0.000014 - momentum: 0.000000
2023-10-25 21:24:36,191 epoch 8 - iter 174/292 - loss 0.01019612 - time (sec): 7.50 - samples/sec: 3593.87 - lr: 0.000013 - momentum: 0.000000
2023-10-25 21:24:37,508 epoch 8 - iter 203/292 - loss 0.00968103 - time (sec): 8.82 - samples/sec: 3519.18 - lr: 0.000013 - momentum: 0.000000
2023-10-25 21:24:38,796 epoch 8 - iter 232/292 - loss 0.00998358 - time (sec): 10.11 - samples/sec: 3443.22 - lr: 0.000012 - momentum: 0.000000
2023-10-25 21:24:40,122 epoch 8 - iter 261/292 - loss 0.00986264 - time (sec): 11.43 - samples/sec: 3449.32 - lr: 0.000012 - momentum: 0.000000
2023-10-25 21:24:41,512 epoch 8 - iter 290/292 - loss 0.01018475 - time (sec): 12.82 - samples/sec: 3452.18 - lr: 0.000011 - momentum: 0.000000
2023-10-25 21:24:41,591 ----------------------------------------------------------------------------------------------------
2023-10-25 21:24:41,591 EPOCH 8 done: loss 0.0104 - lr: 0.000011
2023-10-25 21:24:42,498 DEV : loss 0.19869066774845123 - f1-score (micro avg) 0.7532
2023-10-25 21:24:42,503 saving best model
2023-10-25 21:24:43,185 ----------------------------------------------------------------------------------------------------
2023-10-25 21:24:44,510 epoch 9 - iter 29/292 - loss 0.00300745 - time (sec): 1.32 - samples/sec: 3463.47 - lr: 0.000011 - momentum: 0.000000
2023-10-25 21:24:45,815 epoch 9 - iter 58/292 - loss 0.00560549 - time (sec): 2.63 - samples/sec: 3327.56 - lr: 0.000010 - momentum: 0.000000
2023-10-25 21:24:47,124 epoch 9 - iter 87/292 - loss 0.00597532 - time (sec): 3.94 - samples/sec: 3373.84 - lr: 0.000010 - momentum: 0.000000
2023-10-25 21:24:48,472 epoch 9 - iter 116/292 - loss 0.00660099 - time (sec): 5.28 - samples/sec: 3450.68 - lr: 0.000009 - momentum: 0.000000
2023-10-25 21:24:49,731 epoch 9 - iter 145/292 - loss 0.00583953 - time (sec): 6.54 - samples/sec: 3382.80 - lr: 0.000008 - momentum: 0.000000
2023-10-25 21:24:51,039 epoch 9 - iter 174/292 - loss 0.00673963 - time (sec): 7.85 - samples/sec: 3371.39 - lr: 0.000008 - momentum: 0.000000
2023-10-25 21:24:52,313 epoch 9 - iter 203/292 - loss 0.00632106 - time (sec): 9.13 - samples/sec: 3357.34 - lr: 0.000007 - momentum: 0.000000
2023-10-25 21:24:53,579 epoch 9 - iter 232/292 - loss 0.00585454 - time (sec): 10.39 - samples/sec: 3307.81 - lr: 0.000007 - momentum: 0.000000
2023-10-25 21:24:54,953 epoch 9 - iter 261/292 - loss 0.00520285 - time (sec): 11.77 - samples/sec: 3336.98 - lr: 0.000006 - momentum: 0.000000
2023-10-25 21:24:56,314 epoch 9 - iter 290/292 - loss 0.00729847 - time (sec): 13.13 - samples/sec: 3355.06 - lr: 0.000006 - momentum: 0.000000
2023-10-25 21:24:56,404 ----------------------------------------------------------------------------------------------------
2023-10-25 21:24:56,404 EPOCH 9 done: loss 0.0072 - lr: 0.000006
2023-10-25 21:24:57,331 DEV : loss 0.207131028175354 - f1-score (micro avg) 0.7122
2023-10-25 21:24:57,336 ----------------------------------------------------------------------------------------------------
2023-10-25 21:24:58,656 epoch 10 - iter 29/292 - loss 0.00724381 - time (sec): 1.32 - samples/sec: 3476.36 - lr: 0.000005 - momentum: 0.000000
2023-10-25 21:24:59,897 epoch 10 - iter 58/292 - loss 0.00640674 - time (sec): 2.56 - samples/sec: 3323.92 - lr: 0.000005 - momentum: 0.000000
2023-10-25 21:25:01,158 epoch 10 - iter 87/292 - loss 0.00503092 - time (sec): 3.82 - samples/sec: 3221.06 - lr: 0.000004 - momentum: 0.000000
2023-10-25 21:25:02,538 epoch 10 - iter 116/292 - loss 0.00775827 - time (sec): 5.20 - samples/sec: 3346.98 - lr: 0.000003 - momentum: 0.000000
2023-10-25 21:25:03,822 epoch 10 - iter 145/292 - loss 0.00633399 - time (sec): 6.48 - samples/sec: 3367.76 - lr: 0.000003 - momentum: 0.000000
2023-10-25 21:25:05,151 epoch 10 - iter 174/292 - loss 0.00539255 - time (sec): 7.81 - samples/sec: 3425.52 - lr: 0.000002 - momentum: 0.000000
2023-10-25 21:25:06,458 epoch 10 - iter 203/292 - loss 0.00516222 - time (sec): 9.12 - samples/sec: 3487.33 - lr: 0.000002 - momentum: 0.000000
2023-10-25 21:25:07,712 epoch 10 - iter 232/292 - loss 0.00511528 - time (sec): 10.38 - samples/sec: 3432.05 - lr: 0.000001 - momentum: 0.000000
2023-10-25 21:25:09,027 epoch 10 - iter 261/292 - loss 0.00467370 - time (sec): 11.69 - samples/sec: 3408.74 - lr: 0.000001 - momentum: 0.000000
2023-10-25 21:25:10,287 epoch 10 - iter 290/292 - loss 0.00471642 - time (sec): 12.95 - samples/sec: 3410.43 - lr: 0.000000 - momentum: 0.000000
2023-10-25 21:25:10,370 ----------------------------------------------------------------------------------------------------
2023-10-25 21:25:10,370 EPOCH 10 done: loss 0.0047 - lr: 0.000000
2023-10-25 21:25:11,373 DEV : loss 0.20934493839740753 - f1-score (micro avg) 0.7235
2023-10-25 21:25:11,898 ----------------------------------------------------------------------------------------------------
2023-10-25 21:25:11,899 Loading model from best epoch ...
2023-10-25 21:25:13,646 SequenceTagger predicts: Dictionary with 17 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, S-HumanProd, B-HumanProd, E-HumanProd, I-HumanProd
2023-10-25 21:25:15,364
Results:
- F-score (micro) 0.7502
- F-score (macro) 0.6687
- Accuracy 0.6236
By class:
precision recall f1-score support
PER 0.7936 0.8506 0.8211 348
LOC 0.6594 0.8084 0.7263 261
ORG 0.4038 0.4038 0.4038 52
HumanProd 0.6800 0.7727 0.7234 22
micro avg 0.7078 0.7980 0.7502 683
macro avg 0.6342 0.7089 0.6687 683
weighted avg 0.7090 0.7980 0.7500 683
2023-10-25 21:25:15,364 ----------------------------------------------------------------------------------------------------
|