File size: 23,954 Bytes
874173c |
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 |
2023-10-14 22:53:22,080 ----------------------------------------------------------------------------------------------------
2023-10-14 22:53:22,081 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=13, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-14 22:53:22,081 ----------------------------------------------------------------------------------------------------
2023-10-14 22:53:22,081 MultiCorpus: 14465 train + 1392 dev + 2432 test sentences
- NER_HIPE_2022 Corpus: 14465 train + 1392 dev + 2432 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/letemps/fr/with_doc_seperator
2023-10-14 22:53:22,081 ----------------------------------------------------------------------------------------------------
2023-10-14 22:53:22,081 Train: 14465 sentences
2023-10-14 22:53:22,081 (train_with_dev=False, train_with_test=False)
2023-10-14 22:53:22,081 ----------------------------------------------------------------------------------------------------
2023-10-14 22:53:22,081 Training Params:
2023-10-14 22:53:22,081 - learning_rate: "3e-05"
2023-10-14 22:53:22,081 - mini_batch_size: "8"
2023-10-14 22:53:22,081 - max_epochs: "10"
2023-10-14 22:53:22,081 - shuffle: "True"
2023-10-14 22:53:22,081 ----------------------------------------------------------------------------------------------------
2023-10-14 22:53:22,081 Plugins:
2023-10-14 22:53:22,081 - LinearScheduler | warmup_fraction: '0.1'
2023-10-14 22:53:22,082 ----------------------------------------------------------------------------------------------------
2023-10-14 22:53:22,082 Final evaluation on model from best epoch (best-model.pt)
2023-10-14 22:53:22,082 - metric: "('micro avg', 'f1-score')"
2023-10-14 22:53:22,082 ----------------------------------------------------------------------------------------------------
2023-10-14 22:53:22,082 Computation:
2023-10-14 22:53:22,082 - compute on device: cuda:0
2023-10-14 22:53:22,082 - embedding storage: none
2023-10-14 22:53:22,082 ----------------------------------------------------------------------------------------------------
2023-10-14 22:53:22,082 Model training base path: "hmbench-letemps/fr-dbmdz/bert-base-historic-multilingual-cased-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-3"
2023-10-14 22:53:22,082 ----------------------------------------------------------------------------------------------------
2023-10-14 22:53:22,082 ----------------------------------------------------------------------------------------------------
2023-10-14 22:53:33,152 epoch 1 - iter 180/1809 - loss 1.67027407 - time (sec): 11.07 - samples/sec: 3457.13 - lr: 0.000003 - momentum: 0.000000
2023-10-14 22:53:44,247 epoch 1 - iter 360/1809 - loss 0.96696948 - time (sec): 22.16 - samples/sec: 3425.79 - lr: 0.000006 - momentum: 0.000000
2023-10-14 22:53:55,940 epoch 1 - iter 540/1809 - loss 0.70638013 - time (sec): 33.86 - samples/sec: 3331.88 - lr: 0.000009 - momentum: 0.000000
2023-10-14 22:54:07,179 epoch 1 - iter 720/1809 - loss 0.55894456 - time (sec): 45.10 - samples/sec: 3377.37 - lr: 0.000012 - momentum: 0.000000
2023-10-14 22:54:18,106 epoch 1 - iter 900/1809 - loss 0.47409400 - time (sec): 56.02 - samples/sec: 3391.19 - lr: 0.000015 - momentum: 0.000000
2023-10-14 22:54:29,388 epoch 1 - iter 1080/1809 - loss 0.41428178 - time (sec): 67.31 - samples/sec: 3405.66 - lr: 0.000018 - momentum: 0.000000
2023-10-14 22:54:40,232 epoch 1 - iter 1260/1809 - loss 0.37349997 - time (sec): 78.15 - samples/sec: 3406.85 - lr: 0.000021 - momentum: 0.000000
2023-10-14 22:54:51,158 epoch 1 - iter 1440/1809 - loss 0.34066974 - time (sec): 89.08 - samples/sec: 3410.54 - lr: 0.000024 - momentum: 0.000000
2023-10-14 22:55:02,254 epoch 1 - iter 1620/1809 - loss 0.31439415 - time (sec): 100.17 - samples/sec: 3403.63 - lr: 0.000027 - momentum: 0.000000
2023-10-14 22:55:13,151 epoch 1 - iter 1800/1809 - loss 0.29381810 - time (sec): 111.07 - samples/sec: 3406.22 - lr: 0.000030 - momentum: 0.000000
2023-10-14 22:55:13,661 ----------------------------------------------------------------------------------------------------
2023-10-14 22:55:13,661 EPOCH 1 done: loss 0.2930 - lr: 0.000030
2023-10-14 22:55:19,299 DEV : loss 0.10940668731927872 - f1-score (micro avg) 0.604
2023-10-14 22:55:19,334 saving best model
2023-10-14 22:55:19,823 ----------------------------------------------------------------------------------------------------
2023-10-14 22:55:30,992 epoch 2 - iter 180/1809 - loss 0.08979228 - time (sec): 11.17 - samples/sec: 3317.64 - lr: 0.000030 - momentum: 0.000000
2023-10-14 22:55:42,224 epoch 2 - iter 360/1809 - loss 0.08989058 - time (sec): 22.40 - samples/sec: 3360.44 - lr: 0.000029 - momentum: 0.000000
2023-10-14 22:55:53,269 epoch 2 - iter 540/1809 - loss 0.08901176 - time (sec): 33.44 - samples/sec: 3395.78 - lr: 0.000029 - momentum: 0.000000
2023-10-14 22:56:04,648 epoch 2 - iter 720/1809 - loss 0.08769618 - time (sec): 44.82 - samples/sec: 3400.92 - lr: 0.000029 - momentum: 0.000000
2023-10-14 22:56:15,731 epoch 2 - iter 900/1809 - loss 0.08657795 - time (sec): 55.91 - samples/sec: 3423.49 - lr: 0.000028 - momentum: 0.000000
2023-10-14 22:56:26,940 epoch 2 - iter 1080/1809 - loss 0.08614963 - time (sec): 67.11 - samples/sec: 3412.98 - lr: 0.000028 - momentum: 0.000000
2023-10-14 22:56:37,915 epoch 2 - iter 1260/1809 - loss 0.08650169 - time (sec): 78.09 - samples/sec: 3409.66 - lr: 0.000028 - momentum: 0.000000
2023-10-14 22:56:48,753 epoch 2 - iter 1440/1809 - loss 0.08591524 - time (sec): 88.93 - samples/sec: 3397.15 - lr: 0.000027 - momentum: 0.000000
2023-10-14 22:57:00,131 epoch 2 - iter 1620/1809 - loss 0.08467002 - time (sec): 100.31 - samples/sec: 3398.25 - lr: 0.000027 - momentum: 0.000000
2023-10-14 22:57:10,944 epoch 2 - iter 1800/1809 - loss 0.08346842 - time (sec): 111.12 - samples/sec: 3402.53 - lr: 0.000027 - momentum: 0.000000
2023-10-14 22:57:11,479 ----------------------------------------------------------------------------------------------------
2023-10-14 22:57:11,479 EPOCH 2 done: loss 0.0833 - lr: 0.000027
2023-10-14 22:57:18,912 DEV : loss 0.12328627705574036 - f1-score (micro avg) 0.6436
2023-10-14 22:57:18,946 saving best model
2023-10-14 22:57:19,426 ----------------------------------------------------------------------------------------------------
2023-10-14 22:57:31,144 epoch 3 - iter 180/1809 - loss 0.05050900 - time (sec): 11.72 - samples/sec: 3299.37 - lr: 0.000026 - momentum: 0.000000
2023-10-14 22:57:42,023 epoch 3 - iter 360/1809 - loss 0.05640218 - time (sec): 22.59 - samples/sec: 3385.87 - lr: 0.000026 - momentum: 0.000000
2023-10-14 22:57:52,973 epoch 3 - iter 540/1809 - loss 0.05884620 - time (sec): 33.54 - samples/sec: 3385.83 - lr: 0.000026 - momentum: 0.000000
2023-10-14 22:58:03,993 epoch 3 - iter 720/1809 - loss 0.05715791 - time (sec): 44.57 - samples/sec: 3417.03 - lr: 0.000025 - momentum: 0.000000
2023-10-14 22:58:15,316 epoch 3 - iter 900/1809 - loss 0.05631954 - time (sec): 55.89 - samples/sec: 3394.29 - lr: 0.000025 - momentum: 0.000000
2023-10-14 22:58:26,071 epoch 3 - iter 1080/1809 - loss 0.05690696 - time (sec): 66.64 - samples/sec: 3409.30 - lr: 0.000025 - momentum: 0.000000
2023-10-14 22:58:37,282 epoch 3 - iter 1260/1809 - loss 0.05623779 - time (sec): 77.85 - samples/sec: 3404.87 - lr: 0.000024 - momentum: 0.000000
2023-10-14 22:58:48,305 epoch 3 - iter 1440/1809 - loss 0.05612202 - time (sec): 88.88 - samples/sec: 3399.30 - lr: 0.000024 - momentum: 0.000000
2023-10-14 22:58:59,489 epoch 3 - iter 1620/1809 - loss 0.05782234 - time (sec): 100.06 - samples/sec: 3403.20 - lr: 0.000024 - momentum: 0.000000
2023-10-14 22:59:10,384 epoch 3 - iter 1800/1809 - loss 0.05647561 - time (sec): 110.96 - samples/sec: 3408.09 - lr: 0.000023 - momentum: 0.000000
2023-10-14 22:59:11,005 ----------------------------------------------------------------------------------------------------
2023-10-14 22:59:11,005 EPOCH 3 done: loss 0.0563 - lr: 0.000023
2023-10-14 22:59:16,771 DEV : loss 0.1787535846233368 - f1-score (micro avg) 0.6304
2023-10-14 22:59:16,818 ----------------------------------------------------------------------------------------------------
2023-10-14 22:59:28,293 epoch 4 - iter 180/1809 - loss 0.03058125 - time (sec): 11.47 - samples/sec: 3393.40 - lr: 0.000023 - momentum: 0.000000
2023-10-14 22:59:40,346 epoch 4 - iter 360/1809 - loss 0.03335145 - time (sec): 23.53 - samples/sec: 3247.71 - lr: 0.000023 - momentum: 0.000000
2023-10-14 22:59:51,427 epoch 4 - iter 540/1809 - loss 0.03771304 - time (sec): 34.61 - samples/sec: 3305.71 - lr: 0.000022 - momentum: 0.000000
2023-10-14 23:00:02,403 epoch 4 - iter 720/1809 - loss 0.03909361 - time (sec): 45.58 - samples/sec: 3315.77 - lr: 0.000022 - momentum: 0.000000
2023-10-14 23:00:13,244 epoch 4 - iter 900/1809 - loss 0.03854045 - time (sec): 56.42 - samples/sec: 3339.92 - lr: 0.000022 - momentum: 0.000000
2023-10-14 23:00:24,493 epoch 4 - iter 1080/1809 - loss 0.03909550 - time (sec): 67.67 - samples/sec: 3349.27 - lr: 0.000021 - momentum: 0.000000
2023-10-14 23:00:35,558 epoch 4 - iter 1260/1809 - loss 0.03949444 - time (sec): 78.74 - samples/sec: 3361.35 - lr: 0.000021 - momentum: 0.000000
2023-10-14 23:00:46,652 epoch 4 - iter 1440/1809 - loss 0.03909637 - time (sec): 89.83 - samples/sec: 3374.67 - lr: 0.000021 - momentum: 0.000000
2023-10-14 23:00:57,583 epoch 4 - iter 1620/1809 - loss 0.03977330 - time (sec): 100.76 - samples/sec: 3381.39 - lr: 0.000020 - momentum: 0.000000
2023-10-14 23:01:08,579 epoch 4 - iter 1800/1809 - loss 0.04051474 - time (sec): 111.76 - samples/sec: 3382.87 - lr: 0.000020 - momentum: 0.000000
2023-10-14 23:01:09,100 ----------------------------------------------------------------------------------------------------
2023-10-14 23:01:09,100 EPOCH 4 done: loss 0.0407 - lr: 0.000020
2023-10-14 23:01:14,916 DEV : loss 0.212530717253685 - f1-score (micro avg) 0.6394
2023-10-14 23:01:14,965 ----------------------------------------------------------------------------------------------------
2023-10-14 23:01:26,608 epoch 5 - iter 180/1809 - loss 0.03110605 - time (sec): 11.64 - samples/sec: 3244.49 - lr: 0.000020 - momentum: 0.000000
2023-10-14 23:01:37,407 epoch 5 - iter 360/1809 - loss 0.02633661 - time (sec): 22.44 - samples/sec: 3367.39 - lr: 0.000019 - momentum: 0.000000
2023-10-14 23:01:48,476 epoch 5 - iter 540/1809 - loss 0.02512481 - time (sec): 33.51 - samples/sec: 3381.73 - lr: 0.000019 - momentum: 0.000000
2023-10-14 23:01:59,358 epoch 5 - iter 720/1809 - loss 0.02642180 - time (sec): 44.39 - samples/sec: 3378.28 - lr: 0.000019 - momentum: 0.000000
2023-10-14 23:02:10,472 epoch 5 - iter 900/1809 - loss 0.02681334 - time (sec): 55.51 - samples/sec: 3383.01 - lr: 0.000018 - momentum: 0.000000
2023-10-14 23:02:21,350 epoch 5 - iter 1080/1809 - loss 0.02711812 - time (sec): 66.38 - samples/sec: 3387.39 - lr: 0.000018 - momentum: 0.000000
2023-10-14 23:02:31,971 epoch 5 - iter 1260/1809 - loss 0.02724187 - time (sec): 77.00 - samples/sec: 3399.19 - lr: 0.000018 - momentum: 0.000000
2023-10-14 23:02:43,223 epoch 5 - iter 1440/1809 - loss 0.02806269 - time (sec): 88.26 - samples/sec: 3411.19 - lr: 0.000017 - momentum: 0.000000
2023-10-14 23:02:55,040 epoch 5 - iter 1620/1809 - loss 0.02837556 - time (sec): 100.07 - samples/sec: 3395.44 - lr: 0.000017 - momentum: 0.000000
2023-10-14 23:03:06,144 epoch 5 - iter 1800/1809 - loss 0.02959314 - time (sec): 111.18 - samples/sec: 3402.73 - lr: 0.000017 - momentum: 0.000000
2023-10-14 23:03:06,663 ----------------------------------------------------------------------------------------------------
2023-10-14 23:03:06,663 EPOCH 5 done: loss 0.0297 - lr: 0.000017
2023-10-14 23:03:12,415 DEV : loss 0.30899283289909363 - f1-score (micro avg) 0.6383
2023-10-14 23:03:12,460 ----------------------------------------------------------------------------------------------------
2023-10-14 23:03:23,454 epoch 6 - iter 180/1809 - loss 0.01975153 - time (sec): 10.99 - samples/sec: 3268.70 - lr: 0.000016 - momentum: 0.000000
2023-10-14 23:03:34,423 epoch 6 - iter 360/1809 - loss 0.02159189 - time (sec): 21.96 - samples/sec: 3373.45 - lr: 0.000016 - momentum: 0.000000
2023-10-14 23:03:45,156 epoch 6 - iter 540/1809 - loss 0.02205122 - time (sec): 32.69 - samples/sec: 3395.68 - lr: 0.000016 - momentum: 0.000000
2023-10-14 23:03:55,922 epoch 6 - iter 720/1809 - loss 0.02227128 - time (sec): 43.46 - samples/sec: 3422.32 - lr: 0.000015 - momentum: 0.000000
2023-10-14 23:04:06,988 epoch 6 - iter 900/1809 - loss 0.02215978 - time (sec): 54.53 - samples/sec: 3437.87 - lr: 0.000015 - momentum: 0.000000
2023-10-14 23:04:18,180 epoch 6 - iter 1080/1809 - loss 0.02229865 - time (sec): 65.72 - samples/sec: 3431.52 - lr: 0.000015 - momentum: 0.000000
2023-10-14 23:04:29,141 epoch 6 - iter 1260/1809 - loss 0.02131826 - time (sec): 76.68 - samples/sec: 3450.76 - lr: 0.000014 - momentum: 0.000000
2023-10-14 23:04:40,072 epoch 6 - iter 1440/1809 - loss 0.02100046 - time (sec): 87.61 - samples/sec: 3471.36 - lr: 0.000014 - momentum: 0.000000
2023-10-14 23:04:50,521 epoch 6 - iter 1620/1809 - loss 0.02171985 - time (sec): 98.06 - samples/sec: 3468.48 - lr: 0.000014 - momentum: 0.000000
2023-10-14 23:05:01,563 epoch 6 - iter 1800/1809 - loss 0.02103705 - time (sec): 109.10 - samples/sec: 3465.50 - lr: 0.000013 - momentum: 0.000000
2023-10-14 23:05:02,125 ----------------------------------------------------------------------------------------------------
2023-10-14 23:05:02,125 EPOCH 6 done: loss 0.0211 - lr: 0.000013
2023-10-14 23:05:09,525 DEV : loss 0.3297453224658966 - f1-score (micro avg) 0.6349
2023-10-14 23:05:09,564 ----------------------------------------------------------------------------------------------------
2023-10-14 23:05:21,406 epoch 7 - iter 180/1809 - loss 0.01552725 - time (sec): 11.84 - samples/sec: 3136.20 - lr: 0.000013 - momentum: 0.000000
2023-10-14 23:05:32,981 epoch 7 - iter 360/1809 - loss 0.01402831 - time (sec): 23.42 - samples/sec: 3263.96 - lr: 0.000013 - momentum: 0.000000
2023-10-14 23:05:44,453 epoch 7 - iter 540/1809 - loss 0.01360336 - time (sec): 34.89 - samples/sec: 3314.74 - lr: 0.000012 - momentum: 0.000000
2023-10-14 23:05:55,765 epoch 7 - iter 720/1809 - loss 0.01494343 - time (sec): 46.20 - samples/sec: 3294.49 - lr: 0.000012 - momentum: 0.000000
2023-10-14 23:06:06,988 epoch 7 - iter 900/1809 - loss 0.01605299 - time (sec): 57.42 - samples/sec: 3310.58 - lr: 0.000012 - momentum: 0.000000
2023-10-14 23:06:17,896 epoch 7 - iter 1080/1809 - loss 0.01580151 - time (sec): 68.33 - samples/sec: 3335.50 - lr: 0.000011 - momentum: 0.000000
2023-10-14 23:06:29,114 epoch 7 - iter 1260/1809 - loss 0.01590065 - time (sec): 79.55 - samples/sec: 3352.47 - lr: 0.000011 - momentum: 0.000000
2023-10-14 23:06:39,900 epoch 7 - iter 1440/1809 - loss 0.01510279 - time (sec): 90.33 - samples/sec: 3348.18 - lr: 0.000011 - momentum: 0.000000
2023-10-14 23:06:51,240 epoch 7 - iter 1620/1809 - loss 0.01461003 - time (sec): 101.67 - samples/sec: 3346.97 - lr: 0.000010 - momentum: 0.000000
2023-10-14 23:07:02,417 epoch 7 - iter 1800/1809 - loss 0.01467349 - time (sec): 112.85 - samples/sec: 3351.90 - lr: 0.000010 - momentum: 0.000000
2023-10-14 23:07:02,945 ----------------------------------------------------------------------------------------------------
2023-10-14 23:07:02,945 EPOCH 7 done: loss 0.0146 - lr: 0.000010
2023-10-14 23:07:10,634 DEV : loss 0.3628890812397003 - f1-score (micro avg) 0.6504
2023-10-14 23:07:10,675 saving best model
2023-10-14 23:07:11,201 ----------------------------------------------------------------------------------------------------
2023-10-14 23:07:22,863 epoch 8 - iter 180/1809 - loss 0.00662021 - time (sec): 11.66 - samples/sec: 3155.29 - lr: 0.000010 - momentum: 0.000000
2023-10-14 23:07:34,630 epoch 8 - iter 360/1809 - loss 0.00795343 - time (sec): 23.43 - samples/sec: 3215.56 - lr: 0.000009 - momentum: 0.000000
2023-10-14 23:07:46,178 epoch 8 - iter 540/1809 - loss 0.00814355 - time (sec): 34.97 - samples/sec: 3244.02 - lr: 0.000009 - momentum: 0.000000
2023-10-14 23:07:57,360 epoch 8 - iter 720/1809 - loss 0.00832031 - time (sec): 46.16 - samples/sec: 3271.35 - lr: 0.000009 - momentum: 0.000000
2023-10-14 23:08:08,704 epoch 8 - iter 900/1809 - loss 0.00816301 - time (sec): 57.50 - samples/sec: 3270.22 - lr: 0.000008 - momentum: 0.000000
2023-10-14 23:08:20,051 epoch 8 - iter 1080/1809 - loss 0.00860007 - time (sec): 68.85 - samples/sec: 3288.75 - lr: 0.000008 - momentum: 0.000000
2023-10-14 23:08:31,136 epoch 8 - iter 1260/1809 - loss 0.00904243 - time (sec): 79.93 - samples/sec: 3313.67 - lr: 0.000008 - momentum: 0.000000
2023-10-14 23:08:42,429 epoch 8 - iter 1440/1809 - loss 0.00953422 - time (sec): 91.23 - samples/sec: 3313.24 - lr: 0.000007 - momentum: 0.000000
2023-10-14 23:08:53,796 epoch 8 - iter 1620/1809 - loss 0.01000057 - time (sec): 102.59 - samples/sec: 3308.69 - lr: 0.000007 - momentum: 0.000000
2023-10-14 23:09:05,437 epoch 8 - iter 1800/1809 - loss 0.01005559 - time (sec): 114.23 - samples/sec: 3312.91 - lr: 0.000007 - momentum: 0.000000
2023-10-14 23:09:05,941 ----------------------------------------------------------------------------------------------------
2023-10-14 23:09:05,941 EPOCH 8 done: loss 0.0100 - lr: 0.000007
2023-10-14 23:09:13,141 DEV : loss 0.37255582213401794 - f1-score (micro avg) 0.6496
2023-10-14 23:09:13,188 ----------------------------------------------------------------------------------------------------
2023-10-14 23:09:24,752 epoch 9 - iter 180/1809 - loss 0.00494447 - time (sec): 11.56 - samples/sec: 3303.26 - lr: 0.000006 - momentum: 0.000000
2023-10-14 23:09:36,272 epoch 9 - iter 360/1809 - loss 0.00585614 - time (sec): 23.08 - samples/sec: 3278.56 - lr: 0.000006 - momentum: 0.000000
2023-10-14 23:09:47,503 epoch 9 - iter 540/1809 - loss 0.00614073 - time (sec): 34.31 - samples/sec: 3302.47 - lr: 0.000006 - momentum: 0.000000
2023-10-14 23:09:58,681 epoch 9 - iter 720/1809 - loss 0.00551232 - time (sec): 45.49 - samples/sec: 3340.65 - lr: 0.000005 - momentum: 0.000000
2023-10-14 23:10:09,760 epoch 9 - iter 900/1809 - loss 0.00625027 - time (sec): 56.57 - samples/sec: 3357.66 - lr: 0.000005 - momentum: 0.000000
2023-10-14 23:10:20,942 epoch 9 - iter 1080/1809 - loss 0.00619889 - time (sec): 67.75 - samples/sec: 3379.13 - lr: 0.000005 - momentum: 0.000000
2023-10-14 23:10:31,916 epoch 9 - iter 1260/1809 - loss 0.00604495 - time (sec): 78.73 - samples/sec: 3390.61 - lr: 0.000004 - momentum: 0.000000
2023-10-14 23:10:42,708 epoch 9 - iter 1440/1809 - loss 0.00630439 - time (sec): 89.52 - samples/sec: 3391.73 - lr: 0.000004 - momentum: 0.000000
2023-10-14 23:10:53,793 epoch 9 - iter 1620/1809 - loss 0.00656510 - time (sec): 100.60 - samples/sec: 3397.99 - lr: 0.000004 - momentum: 0.000000
2023-10-14 23:11:04,500 epoch 9 - iter 1800/1809 - loss 0.00635203 - time (sec): 111.31 - samples/sec: 3397.12 - lr: 0.000003 - momentum: 0.000000
2023-10-14 23:11:05,069 ----------------------------------------------------------------------------------------------------
2023-10-14 23:11:05,069 EPOCH 9 done: loss 0.0063 - lr: 0.000003
2023-10-14 23:11:10,696 DEV : loss 0.38722819089889526 - f1-score (micro avg) 0.6478
2023-10-14 23:11:10,731 ----------------------------------------------------------------------------------------------------
2023-10-14 23:11:23,051 epoch 10 - iter 180/1809 - loss 0.00392312 - time (sec): 12.32 - samples/sec: 3093.81 - lr: 0.000003 - momentum: 0.000000
2023-10-14 23:11:33,944 epoch 10 - iter 360/1809 - loss 0.00404755 - time (sec): 23.21 - samples/sec: 3274.69 - lr: 0.000003 - momentum: 0.000000
2023-10-14 23:11:44,860 epoch 10 - iter 540/1809 - loss 0.00406751 - time (sec): 34.13 - samples/sec: 3311.50 - lr: 0.000002 - momentum: 0.000000
2023-10-14 23:11:55,986 epoch 10 - iter 720/1809 - loss 0.00488523 - time (sec): 45.25 - samples/sec: 3346.71 - lr: 0.000002 - momentum: 0.000000
2023-10-14 23:12:07,005 epoch 10 - iter 900/1809 - loss 0.00447546 - time (sec): 56.27 - samples/sec: 3372.90 - lr: 0.000002 - momentum: 0.000000
2023-10-14 23:12:17,916 epoch 10 - iter 1080/1809 - loss 0.00409456 - time (sec): 67.18 - samples/sec: 3382.70 - lr: 0.000001 - momentum: 0.000000
2023-10-14 23:12:29,157 epoch 10 - iter 1260/1809 - loss 0.00427865 - time (sec): 78.42 - samples/sec: 3377.11 - lr: 0.000001 - momentum: 0.000000
2023-10-14 23:12:40,344 epoch 10 - iter 1440/1809 - loss 0.00393300 - time (sec): 89.61 - samples/sec: 3387.40 - lr: 0.000001 - momentum: 0.000000
2023-10-14 23:12:51,215 epoch 10 - iter 1620/1809 - loss 0.00442583 - time (sec): 100.48 - samples/sec: 3373.72 - lr: 0.000000 - momentum: 0.000000
2023-10-14 23:13:02,697 epoch 10 - iter 1800/1809 - loss 0.00442660 - time (sec): 111.96 - samples/sec: 3379.21 - lr: 0.000000 - momentum: 0.000000
2023-10-14 23:13:03,201 ----------------------------------------------------------------------------------------------------
2023-10-14 23:13:03,201 EPOCH 10 done: loss 0.0044 - lr: 0.000000
2023-10-14 23:13:08,887 DEV : loss 0.39746981859207153 - f1-score (micro avg) 0.6532
2023-10-14 23:13:08,933 saving best model
2023-10-14 23:13:09,734 ----------------------------------------------------------------------------------------------------
2023-10-14 23:13:09,736 Loading model from best epoch ...
2023-10-14 23:13:11,362 SequenceTagger predicts: Dictionary with 13 tags: O, S-loc, B-loc, E-loc, I-loc, S-pers, B-pers, E-pers, I-pers, S-org, B-org, E-org, I-org
2023-10-14 23:13:20,771
Results:
- F-score (micro) 0.6532
- F-score (macro) 0.5047
- Accuracy 0.5013
By class:
precision recall f1-score support
loc 0.6327 0.7986 0.7061 591
pers 0.5670 0.7703 0.6532 357
org 0.1746 0.1392 0.1549 79
micro avg 0.5858 0.7381 0.6532 1027
macro avg 0.4581 0.5694 0.5047 1027
weighted avg 0.5746 0.7381 0.6453 1027
2023-10-14 23:13:20,771 ----------------------------------------------------------------------------------------------------
|