File size: 23,783 Bytes
9403c51 |
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 |
2023-10-16 20:18:22,608 ----------------------------------------------------------------------------------------------------
2023-10-16 20:18:22,609 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=17, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-16 20:18:22,609 ----------------------------------------------------------------------------------------------------
2023-10-16 20:18:22,609 MultiCorpus: 1085 train + 148 dev + 364 test sentences
- NER_HIPE_2022 Corpus: 1085 train + 148 dev + 364 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/newseye/sv/with_doc_seperator
2023-10-16 20:18:22,609 ----------------------------------------------------------------------------------------------------
2023-10-16 20:18:22,609 Train: 1085 sentences
2023-10-16 20:18:22,609 (train_with_dev=False, train_with_test=False)
2023-10-16 20:18:22,609 ----------------------------------------------------------------------------------------------------
2023-10-16 20:18:22,610 Training Params:
2023-10-16 20:18:22,610 - learning_rate: "5e-05"
2023-10-16 20:18:22,610 - mini_batch_size: "8"
2023-10-16 20:18:22,610 - max_epochs: "10"
2023-10-16 20:18:22,610 - shuffle: "True"
2023-10-16 20:18:22,610 ----------------------------------------------------------------------------------------------------
2023-10-16 20:18:22,610 Plugins:
2023-10-16 20:18:22,610 - LinearScheduler | warmup_fraction: '0.1'
2023-10-16 20:18:22,610 ----------------------------------------------------------------------------------------------------
2023-10-16 20:18:22,610 Final evaluation on model from best epoch (best-model.pt)
2023-10-16 20:18:22,610 - metric: "('micro avg', 'f1-score')"
2023-10-16 20:18:22,610 ----------------------------------------------------------------------------------------------------
2023-10-16 20:18:22,610 Computation:
2023-10-16 20:18:22,610 - compute on device: cuda:0
2023-10-16 20:18:22,610 - embedding storage: none
2023-10-16 20:18:22,610 ----------------------------------------------------------------------------------------------------
2023-10-16 20:18:22,610 Model training base path: "hmbench-newseye/sv-dbmdz/bert-base-historic-multilingual-cased-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-4"
2023-10-16 20:18:22,610 ----------------------------------------------------------------------------------------------------
2023-10-16 20:18:22,610 ----------------------------------------------------------------------------------------------------
2023-10-16 20:18:23,900 epoch 1 - iter 13/136 - loss 2.93136957 - time (sec): 1.29 - samples/sec: 3687.97 - lr: 0.000004 - momentum: 0.000000
2023-10-16 20:18:25,583 epoch 1 - iter 26/136 - loss 2.61761304 - time (sec): 2.97 - samples/sec: 3560.79 - lr: 0.000009 - momentum: 0.000000
2023-10-16 20:18:27,189 epoch 1 - iter 39/136 - loss 2.10292372 - time (sec): 4.58 - samples/sec: 3480.42 - lr: 0.000014 - momentum: 0.000000
2023-10-16 20:18:28,654 epoch 1 - iter 52/136 - loss 1.69932419 - time (sec): 6.04 - samples/sec: 3541.03 - lr: 0.000019 - momentum: 0.000000
2023-10-16 20:18:30,218 epoch 1 - iter 65/136 - loss 1.46083257 - time (sec): 7.61 - samples/sec: 3509.89 - lr: 0.000024 - momentum: 0.000000
2023-10-16 20:18:31,384 epoch 1 - iter 78/136 - loss 1.32435059 - time (sec): 8.77 - samples/sec: 3550.71 - lr: 0.000028 - momentum: 0.000000
2023-10-16 20:18:32,960 epoch 1 - iter 91/136 - loss 1.19352850 - time (sec): 10.35 - samples/sec: 3485.30 - lr: 0.000033 - momentum: 0.000000
2023-10-16 20:18:34,200 epoch 1 - iter 104/136 - loss 1.10278115 - time (sec): 11.59 - samples/sec: 3501.34 - lr: 0.000038 - momentum: 0.000000
2023-10-16 20:18:35,599 epoch 1 - iter 117/136 - loss 1.01082586 - time (sec): 12.99 - samples/sec: 3499.00 - lr: 0.000043 - momentum: 0.000000
2023-10-16 20:18:36,934 epoch 1 - iter 130/136 - loss 0.93871330 - time (sec): 14.32 - samples/sec: 3480.74 - lr: 0.000047 - momentum: 0.000000
2023-10-16 20:18:37,489 ----------------------------------------------------------------------------------------------------
2023-10-16 20:18:37,489 EPOCH 1 done: loss 0.9142 - lr: 0.000047
2023-10-16 20:18:38,542 DEV : loss 0.17539678514003754 - f1-score (micro avg) 0.6643
2023-10-16 20:18:38,546 saving best model
2023-10-16 20:18:38,883 ----------------------------------------------------------------------------------------------------
2023-10-16 20:18:40,211 epoch 2 - iter 13/136 - loss 0.22301634 - time (sec): 1.33 - samples/sec: 3488.45 - lr: 0.000050 - momentum: 0.000000
2023-10-16 20:18:41,522 epoch 2 - iter 26/136 - loss 0.18431598 - time (sec): 2.64 - samples/sec: 3631.46 - lr: 0.000049 - momentum: 0.000000
2023-10-16 20:18:42,978 epoch 2 - iter 39/136 - loss 0.16140995 - time (sec): 4.09 - samples/sec: 3598.49 - lr: 0.000048 - momentum: 0.000000
2023-10-16 20:18:44,206 epoch 2 - iter 52/136 - loss 0.17264568 - time (sec): 5.32 - samples/sec: 3718.67 - lr: 0.000048 - momentum: 0.000000
2023-10-16 20:18:45,431 epoch 2 - iter 65/136 - loss 0.18654532 - time (sec): 6.55 - samples/sec: 3641.40 - lr: 0.000047 - momentum: 0.000000
2023-10-16 20:18:46,877 epoch 2 - iter 78/136 - loss 0.18635291 - time (sec): 7.99 - samples/sec: 3574.94 - lr: 0.000047 - momentum: 0.000000
2023-10-16 20:18:48,524 epoch 2 - iter 91/136 - loss 0.17676252 - time (sec): 9.64 - samples/sec: 3538.90 - lr: 0.000046 - momentum: 0.000000
2023-10-16 20:18:49,939 epoch 2 - iter 104/136 - loss 0.16965661 - time (sec): 11.05 - samples/sec: 3571.79 - lr: 0.000046 - momentum: 0.000000
2023-10-16 20:18:51,488 epoch 2 - iter 117/136 - loss 0.16802195 - time (sec): 12.60 - samples/sec: 3573.08 - lr: 0.000045 - momentum: 0.000000
2023-10-16 20:18:52,961 epoch 2 - iter 130/136 - loss 0.16391174 - time (sec): 14.08 - samples/sec: 3553.33 - lr: 0.000045 - momentum: 0.000000
2023-10-16 20:18:53,453 ----------------------------------------------------------------------------------------------------
2023-10-16 20:18:53,453 EPOCH 2 done: loss 0.1653 - lr: 0.000045
2023-10-16 20:18:54,968 DEV : loss 0.14248891174793243 - f1-score (micro avg) 0.691
2023-10-16 20:18:54,972 saving best model
2023-10-16 20:18:55,435 ----------------------------------------------------------------------------------------------------
2023-10-16 20:18:56,982 epoch 3 - iter 13/136 - loss 0.10095872 - time (sec): 1.55 - samples/sec: 3543.51 - lr: 0.000044 - momentum: 0.000000
2023-10-16 20:18:58,419 epoch 3 - iter 26/136 - loss 0.10478173 - time (sec): 2.98 - samples/sec: 3660.03 - lr: 0.000043 - momentum: 0.000000
2023-10-16 20:18:59,853 epoch 3 - iter 39/136 - loss 0.09423993 - time (sec): 4.42 - samples/sec: 3552.60 - lr: 0.000043 - momentum: 0.000000
2023-10-16 20:19:01,429 epoch 3 - iter 52/136 - loss 0.09587897 - time (sec): 5.99 - samples/sec: 3509.06 - lr: 0.000042 - momentum: 0.000000
2023-10-16 20:19:02,600 epoch 3 - iter 65/136 - loss 0.09997083 - time (sec): 7.16 - samples/sec: 3492.72 - lr: 0.000042 - momentum: 0.000000
2023-10-16 20:19:03,950 epoch 3 - iter 78/136 - loss 0.09751295 - time (sec): 8.51 - samples/sec: 3489.66 - lr: 0.000041 - momentum: 0.000000
2023-10-16 20:19:05,435 epoch 3 - iter 91/136 - loss 0.09486572 - time (sec): 10.00 - samples/sec: 3532.44 - lr: 0.000041 - momentum: 0.000000
2023-10-16 20:19:06,726 epoch 3 - iter 104/136 - loss 0.09199902 - time (sec): 11.29 - samples/sec: 3559.17 - lr: 0.000040 - momentum: 0.000000
2023-10-16 20:19:08,133 epoch 3 - iter 117/136 - loss 0.08768794 - time (sec): 12.70 - samples/sec: 3551.74 - lr: 0.000040 - momentum: 0.000000
2023-10-16 20:19:09,588 epoch 3 - iter 130/136 - loss 0.08763564 - time (sec): 14.15 - samples/sec: 3526.52 - lr: 0.000039 - momentum: 0.000000
2023-10-16 20:19:10,164 ----------------------------------------------------------------------------------------------------
2023-10-16 20:19:10,165 EPOCH 3 done: loss 0.0866 - lr: 0.000039
2023-10-16 20:19:11,832 DEV : loss 0.10237669199705124 - f1-score (micro avg) 0.8268
2023-10-16 20:19:11,836 saving best model
2023-10-16 20:19:12,290 ----------------------------------------------------------------------------------------------------
2023-10-16 20:19:13,716 epoch 4 - iter 13/136 - loss 0.08431575 - time (sec): 1.42 - samples/sec: 3427.07 - lr: 0.000038 - momentum: 0.000000
2023-10-16 20:19:15,238 epoch 4 - iter 26/136 - loss 0.06713900 - time (sec): 2.95 - samples/sec: 3482.32 - lr: 0.000038 - momentum: 0.000000
2023-10-16 20:19:16,482 epoch 4 - iter 39/136 - loss 0.05765410 - time (sec): 4.19 - samples/sec: 3551.75 - lr: 0.000037 - momentum: 0.000000
2023-10-16 20:19:17,877 epoch 4 - iter 52/136 - loss 0.05665901 - time (sec): 5.58 - samples/sec: 3682.64 - lr: 0.000037 - momentum: 0.000000
2023-10-16 20:19:19,212 epoch 4 - iter 65/136 - loss 0.05979582 - time (sec): 6.92 - samples/sec: 3637.82 - lr: 0.000036 - momentum: 0.000000
2023-10-16 20:19:20,790 epoch 4 - iter 78/136 - loss 0.05790229 - time (sec): 8.50 - samples/sec: 3623.54 - lr: 0.000036 - momentum: 0.000000
2023-10-16 20:19:22,120 epoch 4 - iter 91/136 - loss 0.05693213 - time (sec): 9.83 - samples/sec: 3606.14 - lr: 0.000035 - momentum: 0.000000
2023-10-16 20:19:23,630 epoch 4 - iter 104/136 - loss 0.05628982 - time (sec): 11.34 - samples/sec: 3579.50 - lr: 0.000035 - momentum: 0.000000
2023-10-16 20:19:25,016 epoch 4 - iter 117/136 - loss 0.05352644 - time (sec): 12.72 - samples/sec: 3590.41 - lr: 0.000034 - momentum: 0.000000
2023-10-16 20:19:26,241 epoch 4 - iter 130/136 - loss 0.05384621 - time (sec): 13.95 - samples/sec: 3567.78 - lr: 0.000034 - momentum: 0.000000
2023-10-16 20:19:26,852 ----------------------------------------------------------------------------------------------------
2023-10-16 20:19:26,852 EPOCH 4 done: loss 0.0531 - lr: 0.000034
2023-10-16 20:19:28,342 DEV : loss 0.1108362078666687 - f1-score (micro avg) 0.792
2023-10-16 20:19:28,347 ----------------------------------------------------------------------------------------------------
2023-10-16 20:19:29,788 epoch 5 - iter 13/136 - loss 0.04268788 - time (sec): 1.44 - samples/sec: 3396.75 - lr: 0.000033 - momentum: 0.000000
2023-10-16 20:19:31,041 epoch 5 - iter 26/136 - loss 0.03670220 - time (sec): 2.69 - samples/sec: 3419.76 - lr: 0.000032 - momentum: 0.000000
2023-10-16 20:19:32,272 epoch 5 - iter 39/136 - loss 0.04010108 - time (sec): 3.92 - samples/sec: 3646.72 - lr: 0.000032 - momentum: 0.000000
2023-10-16 20:19:33,713 epoch 5 - iter 52/136 - loss 0.04362300 - time (sec): 5.37 - samples/sec: 3515.55 - lr: 0.000031 - momentum: 0.000000
2023-10-16 20:19:35,275 epoch 5 - iter 65/136 - loss 0.03800478 - time (sec): 6.93 - samples/sec: 3481.28 - lr: 0.000031 - momentum: 0.000000
2023-10-16 20:19:36,599 epoch 5 - iter 78/136 - loss 0.03549079 - time (sec): 8.25 - samples/sec: 3546.96 - lr: 0.000030 - momentum: 0.000000
2023-10-16 20:19:37,986 epoch 5 - iter 91/136 - loss 0.03458614 - time (sec): 9.64 - samples/sec: 3543.08 - lr: 0.000030 - momentum: 0.000000
2023-10-16 20:19:39,580 epoch 5 - iter 104/136 - loss 0.03257155 - time (sec): 11.23 - samples/sec: 3549.79 - lr: 0.000029 - momentum: 0.000000
2023-10-16 20:19:40,937 epoch 5 - iter 117/136 - loss 0.03505263 - time (sec): 12.59 - samples/sec: 3533.61 - lr: 0.000029 - momentum: 0.000000
2023-10-16 20:19:42,362 epoch 5 - iter 130/136 - loss 0.03522794 - time (sec): 14.01 - samples/sec: 3540.70 - lr: 0.000028 - momentum: 0.000000
2023-10-16 20:19:43,035 ----------------------------------------------------------------------------------------------------
2023-10-16 20:19:43,036 EPOCH 5 done: loss 0.0344 - lr: 0.000028
2023-10-16 20:19:44,757 DEV : loss 0.12420879304409027 - f1-score (micro avg) 0.8015
2023-10-16 20:19:44,762 ----------------------------------------------------------------------------------------------------
2023-10-16 20:19:46,094 epoch 6 - iter 13/136 - loss 0.01916100 - time (sec): 1.33 - samples/sec: 3326.19 - lr: 0.000027 - momentum: 0.000000
2023-10-16 20:19:47,511 epoch 6 - iter 26/136 - loss 0.02420091 - time (sec): 2.75 - samples/sec: 3358.03 - lr: 0.000027 - momentum: 0.000000
2023-10-16 20:19:49,067 epoch 6 - iter 39/136 - loss 0.02128442 - time (sec): 4.30 - samples/sec: 3389.40 - lr: 0.000026 - momentum: 0.000000
2023-10-16 20:19:50,714 epoch 6 - iter 52/136 - loss 0.02244990 - time (sec): 5.95 - samples/sec: 3469.10 - lr: 0.000026 - momentum: 0.000000
2023-10-16 20:19:52,211 epoch 6 - iter 65/136 - loss 0.02199431 - time (sec): 7.45 - samples/sec: 3475.38 - lr: 0.000025 - momentum: 0.000000
2023-10-16 20:19:53,518 epoch 6 - iter 78/136 - loss 0.02434853 - time (sec): 8.76 - samples/sec: 3477.60 - lr: 0.000025 - momentum: 0.000000
2023-10-16 20:19:54,879 epoch 6 - iter 91/136 - loss 0.02513952 - time (sec): 10.12 - samples/sec: 3445.86 - lr: 0.000024 - momentum: 0.000000
2023-10-16 20:19:56,241 epoch 6 - iter 104/136 - loss 0.02469020 - time (sec): 11.48 - samples/sec: 3465.60 - lr: 0.000024 - momentum: 0.000000
2023-10-16 20:19:57,761 epoch 6 - iter 117/136 - loss 0.02336871 - time (sec): 13.00 - samples/sec: 3473.17 - lr: 0.000023 - momentum: 0.000000
2023-10-16 20:19:59,055 epoch 6 - iter 130/136 - loss 0.02303322 - time (sec): 14.29 - samples/sec: 3518.84 - lr: 0.000023 - momentum: 0.000000
2023-10-16 20:19:59,504 ----------------------------------------------------------------------------------------------------
2023-10-16 20:19:59,505 EPOCH 6 done: loss 0.0235 - lr: 0.000023
2023-10-16 20:20:01,009 DEV : loss 0.1299697607755661 - f1-score (micro avg) 0.82
2023-10-16 20:20:01,015 ----------------------------------------------------------------------------------------------------
2023-10-16 20:20:02,378 epoch 7 - iter 13/136 - loss 0.02201266 - time (sec): 1.36 - samples/sec: 3644.27 - lr: 0.000022 - momentum: 0.000000
2023-10-16 20:20:04,027 epoch 7 - iter 26/136 - loss 0.01585502 - time (sec): 3.01 - samples/sec: 3607.62 - lr: 0.000021 - momentum: 0.000000
2023-10-16 20:20:05,485 epoch 7 - iter 39/136 - loss 0.01813674 - time (sec): 4.47 - samples/sec: 3493.68 - lr: 0.000021 - momentum: 0.000000
2023-10-16 20:20:07,060 epoch 7 - iter 52/136 - loss 0.01977797 - time (sec): 6.04 - samples/sec: 3551.44 - lr: 0.000020 - momentum: 0.000000
2023-10-16 20:20:08,369 epoch 7 - iter 65/136 - loss 0.01878455 - time (sec): 7.35 - samples/sec: 3549.15 - lr: 0.000020 - momentum: 0.000000
2023-10-16 20:20:09,893 epoch 7 - iter 78/136 - loss 0.01847453 - time (sec): 8.88 - samples/sec: 3555.26 - lr: 0.000019 - momentum: 0.000000
2023-10-16 20:20:11,172 epoch 7 - iter 91/136 - loss 0.01767757 - time (sec): 10.16 - samples/sec: 3562.64 - lr: 0.000019 - momentum: 0.000000
2023-10-16 20:20:12,435 epoch 7 - iter 104/136 - loss 0.01671361 - time (sec): 11.42 - samples/sec: 3581.10 - lr: 0.000018 - momentum: 0.000000
2023-10-16 20:20:13,763 epoch 7 - iter 117/136 - loss 0.01655957 - time (sec): 12.75 - samples/sec: 3559.24 - lr: 0.000018 - momentum: 0.000000
2023-10-16 20:20:15,116 epoch 7 - iter 130/136 - loss 0.01641038 - time (sec): 14.10 - samples/sec: 3556.12 - lr: 0.000017 - momentum: 0.000000
2023-10-16 20:20:15,632 ----------------------------------------------------------------------------------------------------
2023-10-16 20:20:15,632 EPOCH 7 done: loss 0.0160 - lr: 0.000017
2023-10-16 20:20:17,148 DEV : loss 0.13152331113815308 - f1-score (micro avg) 0.8315
2023-10-16 20:20:17,154 saving best model
2023-10-16 20:20:17,607 ----------------------------------------------------------------------------------------------------
2023-10-16 20:20:19,128 epoch 8 - iter 13/136 - loss 0.01402109 - time (sec): 1.52 - samples/sec: 2768.32 - lr: 0.000016 - momentum: 0.000000
2023-10-16 20:20:20,447 epoch 8 - iter 26/136 - loss 0.01902129 - time (sec): 2.84 - samples/sec: 3178.52 - lr: 0.000016 - momentum: 0.000000
2023-10-16 20:20:21,607 epoch 8 - iter 39/136 - loss 0.01608610 - time (sec): 4.00 - samples/sec: 3256.67 - lr: 0.000015 - momentum: 0.000000
2023-10-16 20:20:23,057 epoch 8 - iter 52/136 - loss 0.01326992 - time (sec): 5.45 - samples/sec: 3408.57 - lr: 0.000015 - momentum: 0.000000
2023-10-16 20:20:24,346 epoch 8 - iter 65/136 - loss 0.01198353 - time (sec): 6.74 - samples/sec: 3474.97 - lr: 0.000014 - momentum: 0.000000
2023-10-16 20:20:25,975 epoch 8 - iter 78/136 - loss 0.01138949 - time (sec): 8.37 - samples/sec: 3455.20 - lr: 0.000014 - momentum: 0.000000
2023-10-16 20:20:27,579 epoch 8 - iter 91/136 - loss 0.01277453 - time (sec): 9.97 - samples/sec: 3455.56 - lr: 0.000013 - momentum: 0.000000
2023-10-16 20:20:28,873 epoch 8 - iter 104/136 - loss 0.01291802 - time (sec): 11.26 - samples/sec: 3477.15 - lr: 0.000013 - momentum: 0.000000
2023-10-16 20:20:30,352 epoch 8 - iter 117/136 - loss 0.01246003 - time (sec): 12.74 - samples/sec: 3471.33 - lr: 0.000012 - momentum: 0.000000
2023-10-16 20:20:31,831 epoch 8 - iter 130/136 - loss 0.01146785 - time (sec): 14.22 - samples/sec: 3491.93 - lr: 0.000012 - momentum: 0.000000
2023-10-16 20:20:32,488 ----------------------------------------------------------------------------------------------------
2023-10-16 20:20:32,488 EPOCH 8 done: loss 0.0123 - lr: 0.000012
2023-10-16 20:20:33,976 DEV : loss 0.15120868384838104 - f1-score (micro avg) 0.8192
2023-10-16 20:20:33,981 ----------------------------------------------------------------------------------------------------
2023-10-16 20:20:35,606 epoch 9 - iter 13/136 - loss 0.00940542 - time (sec): 1.62 - samples/sec: 3592.17 - lr: 0.000011 - momentum: 0.000000
2023-10-16 20:20:36,941 epoch 9 - iter 26/136 - loss 0.00670356 - time (sec): 2.96 - samples/sec: 3633.67 - lr: 0.000010 - momentum: 0.000000
2023-10-16 20:20:38,165 epoch 9 - iter 39/136 - loss 0.00947777 - time (sec): 4.18 - samples/sec: 3542.21 - lr: 0.000010 - momentum: 0.000000
2023-10-16 20:20:39,685 epoch 9 - iter 52/136 - loss 0.00891002 - time (sec): 5.70 - samples/sec: 3538.63 - lr: 0.000009 - momentum: 0.000000
2023-10-16 20:20:41,079 epoch 9 - iter 65/136 - loss 0.00918508 - time (sec): 7.10 - samples/sec: 3529.78 - lr: 0.000009 - momentum: 0.000000
2023-10-16 20:20:42,396 epoch 9 - iter 78/136 - loss 0.00934079 - time (sec): 8.41 - samples/sec: 3592.02 - lr: 0.000008 - momentum: 0.000000
2023-10-16 20:20:43,945 epoch 9 - iter 91/136 - loss 0.00873731 - time (sec): 9.96 - samples/sec: 3549.63 - lr: 0.000008 - momentum: 0.000000
2023-10-16 20:20:45,515 epoch 9 - iter 104/136 - loss 0.00839287 - time (sec): 11.53 - samples/sec: 3560.14 - lr: 0.000007 - momentum: 0.000000
2023-10-16 20:20:46,878 epoch 9 - iter 117/136 - loss 0.00857232 - time (sec): 12.90 - samples/sec: 3579.75 - lr: 0.000007 - momentum: 0.000000
2023-10-16 20:20:48,163 epoch 9 - iter 130/136 - loss 0.01058054 - time (sec): 14.18 - samples/sec: 3549.23 - lr: 0.000006 - momentum: 0.000000
2023-10-16 20:20:48,734 ----------------------------------------------------------------------------------------------------
2023-10-16 20:20:48,735 EPOCH 9 done: loss 0.0106 - lr: 0.000006
2023-10-16 20:20:50,191 DEV : loss 0.14968341588974 - f1-score (micro avg) 0.8175
2023-10-16 20:20:50,195 ----------------------------------------------------------------------------------------------------
2023-10-16 20:20:51,766 epoch 10 - iter 13/136 - loss 0.00201920 - time (sec): 1.57 - samples/sec: 3363.38 - lr: 0.000005 - momentum: 0.000000
2023-10-16 20:20:52,880 epoch 10 - iter 26/136 - loss 0.00257635 - time (sec): 2.68 - samples/sec: 3418.54 - lr: 0.000005 - momentum: 0.000000
2023-10-16 20:20:54,377 epoch 10 - iter 39/136 - loss 0.00284480 - time (sec): 4.18 - samples/sec: 3228.03 - lr: 0.000004 - momentum: 0.000000
2023-10-16 20:20:55,815 epoch 10 - iter 52/136 - loss 0.00540897 - time (sec): 5.62 - samples/sec: 3338.60 - lr: 0.000004 - momentum: 0.000000
2023-10-16 20:20:57,292 epoch 10 - iter 65/136 - loss 0.00496823 - time (sec): 7.10 - samples/sec: 3352.13 - lr: 0.000003 - momentum: 0.000000
2023-10-16 20:20:58,618 epoch 10 - iter 78/136 - loss 0.00534676 - time (sec): 8.42 - samples/sec: 3351.23 - lr: 0.000003 - momentum: 0.000000
2023-10-16 20:21:00,051 epoch 10 - iter 91/136 - loss 0.00574925 - time (sec): 9.85 - samples/sec: 3392.70 - lr: 0.000002 - momentum: 0.000000
2023-10-16 20:21:01,333 epoch 10 - iter 104/136 - loss 0.00664022 - time (sec): 11.14 - samples/sec: 3429.83 - lr: 0.000002 - momentum: 0.000000
2023-10-16 20:21:02,990 epoch 10 - iter 117/136 - loss 0.00799959 - time (sec): 12.79 - samples/sec: 3422.62 - lr: 0.000001 - momentum: 0.000000
2023-10-16 20:21:04,396 epoch 10 - iter 130/136 - loss 0.00826072 - time (sec): 14.20 - samples/sec: 3483.83 - lr: 0.000000 - momentum: 0.000000
2023-10-16 20:21:05,155 ----------------------------------------------------------------------------------------------------
2023-10-16 20:21:05,155 EPOCH 10 done: loss 0.0080 - lr: 0.000000
2023-10-16 20:21:06,676 DEV : loss 0.15046021342277527 - f1-score (micro avg) 0.8324
2023-10-16 20:21:06,681 saving best model
2023-10-16 20:21:07,513 ----------------------------------------------------------------------------------------------------
2023-10-16 20:21:07,514 Loading model from best epoch ...
2023-10-16 20:21:09,210 SequenceTagger predicts: Dictionary with 17 tags: O, S-LOC, B-LOC, E-LOC, I-LOC, S-PER, B-PER, E-PER, I-PER, S-HumanProd, B-HumanProd, E-HumanProd, I-HumanProd, S-ORG, B-ORG, E-ORG, I-ORG
2023-10-16 20:21:11,380
Results:
- F-score (micro) 0.7767
- F-score (macro) 0.7354
- Accuracy 0.6543
By class:
precision recall f1-score support
LOC 0.8112 0.8814 0.8449 312
PER 0.6628 0.8221 0.7339 208
ORG 0.5306 0.4727 0.5000 55
HumanProd 0.7586 1.0000 0.8627 22
micro avg 0.7319 0.8275 0.7767 597
macro avg 0.6908 0.7941 0.7354 597
weighted avg 0.7317 0.8275 0.7751 597
2023-10-16 20:21:11,380 ----------------------------------------------------------------------------------------------------
|