File size: 23,997 Bytes
95588df |
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 |
2023-10-25 21:20:06,302 ----------------------------------------------------------------------------------------------------
2023-10-25 21:20:06,303 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 21:20:06,303 ----------------------------------------------------------------------------------------------------
2023-10-25 21:20:06,303 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
2023-10-25 21:20:06,303 ----------------------------------------------------------------------------------------------------
2023-10-25 21:20:06,303 Train: 1085 sentences
2023-10-25 21:20:06,303 (train_with_dev=False, train_with_test=False)
2023-10-25 21:20:06,303 ----------------------------------------------------------------------------------------------------
2023-10-25 21:20:06,303 Training Params:
2023-10-25 21:20:06,303 - learning_rate: "3e-05"
2023-10-25 21:20:06,303 - mini_batch_size: "8"
2023-10-25 21:20:06,304 - max_epochs: "10"
2023-10-25 21:20:06,304 - shuffle: "True"
2023-10-25 21:20:06,304 ----------------------------------------------------------------------------------------------------
2023-10-25 21:20:06,304 Plugins:
2023-10-25 21:20:06,304 - TensorboardLogger
2023-10-25 21:20:06,304 - LinearScheduler | warmup_fraction: '0.1'
2023-10-25 21:20:06,304 ----------------------------------------------------------------------------------------------------
2023-10-25 21:20:06,304 Final evaluation on model from best epoch (best-model.pt)
2023-10-25 21:20:06,304 - metric: "('micro avg', 'f1-score')"
2023-10-25 21:20:06,304 ----------------------------------------------------------------------------------------------------
2023-10-25 21:20:06,304 Computation:
2023-10-25 21:20:06,304 - compute on device: cuda:0
2023-10-25 21:20:06,304 - embedding storage: none
2023-10-25 21:20:06,304 ----------------------------------------------------------------------------------------------------
2023-10-25 21:20:06,304 Model training base path: "hmbench-newseye/sv-dbmdz/bert-base-historic-multilingual-64k-td-cased-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-4"
2023-10-25 21:20:06,304 ----------------------------------------------------------------------------------------------------
2023-10-25 21:20:06,304 ----------------------------------------------------------------------------------------------------
2023-10-25 21:20:06,304 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-25 21:20:07,355 epoch 1 - iter 13/136 - loss 3.19537858 - time (sec): 1.05 - samples/sec: 5006.81 - lr: 0.000003 - momentum: 0.000000
2023-10-25 21:20:08,259 epoch 1 - iter 26/136 - loss 2.76016229 - time (sec): 1.95 - samples/sec: 5426.73 - lr: 0.000006 - momentum: 0.000000
2023-10-25 21:20:09,146 epoch 1 - iter 39/136 - loss 2.31898621 - time (sec): 2.84 - samples/sec: 5352.27 - lr: 0.000008 - momentum: 0.000000
2023-10-25 21:20:10,159 epoch 1 - iter 52/136 - loss 1.87763258 - time (sec): 3.85 - samples/sec: 5221.03 - lr: 0.000011 - momentum: 0.000000
2023-10-25 21:20:11,143 epoch 1 - iter 65/136 - loss 1.62815715 - time (sec): 4.84 - samples/sec: 5007.79 - lr: 0.000014 - momentum: 0.000000
2023-10-25 21:20:12,209 epoch 1 - iter 78/136 - loss 1.39999778 - time (sec): 5.90 - samples/sec: 5074.14 - lr: 0.000017 - momentum: 0.000000
2023-10-25 21:20:13,409 epoch 1 - iter 91/136 - loss 1.24281698 - time (sec): 7.10 - samples/sec: 4954.26 - lr: 0.000020 - momentum: 0.000000
2023-10-25 21:20:14,325 epoch 1 - iter 104/136 - loss 1.13361333 - time (sec): 8.02 - samples/sec: 4962.06 - lr: 0.000023 - momentum: 0.000000
2023-10-25 21:20:15,286 epoch 1 - iter 117/136 - loss 1.04110203 - time (sec): 8.98 - samples/sec: 4950.04 - lr: 0.000026 - momentum: 0.000000
2023-10-25 21:20:16,373 epoch 1 - iter 130/136 - loss 0.96009861 - time (sec): 10.07 - samples/sec: 4950.16 - lr: 0.000028 - momentum: 0.000000
2023-10-25 21:20:16,839 ----------------------------------------------------------------------------------------------------
2023-10-25 21:20:16,839 EPOCH 1 done: loss 0.9272 - lr: 0.000028
2023-10-25 21:20:17,482 DEV : loss 0.16158084571361542 - f1-score (micro avg) 0.6429
2023-10-25 21:20:17,488 saving best model
2023-10-25 21:20:17,990 ----------------------------------------------------------------------------------------------------
2023-10-25 21:20:18,887 epoch 2 - iter 13/136 - loss 0.14478029 - time (sec): 0.90 - samples/sec: 5170.78 - lr: 0.000030 - momentum: 0.000000
2023-10-25 21:20:20,273 epoch 2 - iter 26/136 - loss 0.13903227 - time (sec): 2.28 - samples/sec: 4372.66 - lr: 0.000029 - momentum: 0.000000
2023-10-25 21:20:21,307 epoch 2 - iter 39/136 - loss 0.14383107 - time (sec): 3.32 - samples/sec: 4307.00 - lr: 0.000029 - momentum: 0.000000
2023-10-25 21:20:22,354 epoch 2 - iter 52/136 - loss 0.14741814 - time (sec): 4.36 - samples/sec: 4615.83 - lr: 0.000029 - momentum: 0.000000
2023-10-25 21:20:23,357 epoch 2 - iter 65/136 - loss 0.14244389 - time (sec): 5.37 - samples/sec: 4763.95 - lr: 0.000028 - momentum: 0.000000
2023-10-25 21:20:24,327 epoch 2 - iter 78/136 - loss 0.14786401 - time (sec): 6.34 - samples/sec: 4760.82 - lr: 0.000028 - momentum: 0.000000
2023-10-25 21:20:25,424 epoch 2 - iter 91/136 - loss 0.14575958 - time (sec): 7.43 - samples/sec: 4838.69 - lr: 0.000028 - momentum: 0.000000
2023-10-25 21:20:26,475 epoch 2 - iter 104/136 - loss 0.14340023 - time (sec): 8.48 - samples/sec: 4900.45 - lr: 0.000027 - momentum: 0.000000
2023-10-25 21:20:27,349 epoch 2 - iter 117/136 - loss 0.14202521 - time (sec): 9.36 - samples/sec: 4899.16 - lr: 0.000027 - momentum: 0.000000
2023-10-25 21:20:28,296 epoch 2 - iter 130/136 - loss 0.14295048 - time (sec): 10.31 - samples/sec: 4888.77 - lr: 0.000027 - momentum: 0.000000
2023-10-25 21:20:28,683 ----------------------------------------------------------------------------------------------------
2023-10-25 21:20:28,683 EPOCH 2 done: loss 0.1434 - lr: 0.000027
2023-10-25 21:20:29,924 DEV : loss 0.09990036487579346 - f1-score (micro avg) 0.7468
2023-10-25 21:20:29,930 saving best model
2023-10-25 21:20:30,644 ----------------------------------------------------------------------------------------------------
2023-10-25 21:20:31,648 epoch 3 - iter 13/136 - loss 0.11960243 - time (sec): 1.00 - samples/sec: 4942.09 - lr: 0.000026 - momentum: 0.000000
2023-10-25 21:20:32,586 epoch 3 - iter 26/136 - loss 0.09759495 - time (sec): 1.94 - samples/sec: 5254.25 - lr: 0.000026 - momentum: 0.000000
2023-10-25 21:20:33,650 epoch 3 - iter 39/136 - loss 0.08584252 - time (sec): 3.00 - samples/sec: 5065.02 - lr: 0.000026 - momentum: 0.000000
2023-10-25 21:20:34,640 epoch 3 - iter 52/136 - loss 0.08045591 - time (sec): 3.99 - samples/sec: 5160.60 - lr: 0.000025 - momentum: 0.000000
2023-10-25 21:20:35,645 epoch 3 - iter 65/136 - loss 0.07965884 - time (sec): 5.00 - samples/sec: 5163.72 - lr: 0.000025 - momentum: 0.000000
2023-10-25 21:20:36,647 epoch 3 - iter 78/136 - loss 0.07654363 - time (sec): 6.00 - samples/sec: 5105.24 - lr: 0.000025 - momentum: 0.000000
2023-10-25 21:20:37,586 epoch 3 - iter 91/136 - loss 0.07736097 - time (sec): 6.94 - samples/sec: 5017.45 - lr: 0.000024 - momentum: 0.000000
2023-10-25 21:20:38,660 epoch 3 - iter 104/136 - loss 0.07706710 - time (sec): 8.01 - samples/sec: 4990.12 - lr: 0.000024 - momentum: 0.000000
2023-10-25 21:20:39,700 epoch 3 - iter 117/136 - loss 0.07804906 - time (sec): 9.05 - samples/sec: 4981.63 - lr: 0.000024 - momentum: 0.000000
2023-10-25 21:20:40,611 epoch 3 - iter 130/136 - loss 0.07904440 - time (sec): 9.97 - samples/sec: 4955.02 - lr: 0.000024 - momentum: 0.000000
2023-10-25 21:20:41,066 ----------------------------------------------------------------------------------------------------
2023-10-25 21:20:41,066 EPOCH 3 done: loss 0.0780 - lr: 0.000024
2023-10-25 21:20:42,197 DEV : loss 0.10422874242067337 - f1-score (micro avg) 0.7585
2023-10-25 21:20:42,203 saving best model
2023-10-25 21:20:42,876 ----------------------------------------------------------------------------------------------------
2023-10-25 21:20:44,128 epoch 4 - iter 13/136 - loss 0.04067450 - time (sec): 1.25 - samples/sec: 4058.14 - lr: 0.000023 - momentum: 0.000000
2023-10-25 21:20:45,027 epoch 4 - iter 26/136 - loss 0.05076903 - time (sec): 2.15 - samples/sec: 4540.53 - lr: 0.000023 - momentum: 0.000000
2023-10-25 21:20:46,128 epoch 4 - iter 39/136 - loss 0.04630433 - time (sec): 3.25 - samples/sec: 4867.37 - lr: 0.000022 - momentum: 0.000000
2023-10-25 21:20:47,087 epoch 4 - iter 52/136 - loss 0.04968443 - time (sec): 4.21 - samples/sec: 4902.47 - lr: 0.000022 - momentum: 0.000000
2023-10-25 21:20:48,116 epoch 4 - iter 65/136 - loss 0.04696549 - time (sec): 5.24 - samples/sec: 4932.60 - lr: 0.000022 - momentum: 0.000000
2023-10-25 21:20:49,083 epoch 4 - iter 78/136 - loss 0.04539194 - time (sec): 6.21 - samples/sec: 4925.62 - lr: 0.000021 - momentum: 0.000000
2023-10-25 21:20:50,011 epoch 4 - iter 91/136 - loss 0.04798111 - time (sec): 7.13 - samples/sec: 5005.83 - lr: 0.000021 - momentum: 0.000000
2023-10-25 21:20:50,929 epoch 4 - iter 104/136 - loss 0.04633553 - time (sec): 8.05 - samples/sec: 5037.74 - lr: 0.000021 - momentum: 0.000000
2023-10-25 21:20:51,856 epoch 4 - iter 117/136 - loss 0.04647949 - time (sec): 8.98 - samples/sec: 5046.46 - lr: 0.000021 - momentum: 0.000000
2023-10-25 21:20:52,836 epoch 4 - iter 130/136 - loss 0.04697246 - time (sec): 9.96 - samples/sec: 5034.81 - lr: 0.000020 - momentum: 0.000000
2023-10-25 21:20:53,211 ----------------------------------------------------------------------------------------------------
2023-10-25 21:20:53,211 EPOCH 4 done: loss 0.0462 - lr: 0.000020
2023-10-25 21:20:54,399 DEV : loss 0.10547740757465363 - f1-score (micro avg) 0.7802
2023-10-25 21:20:54,405 saving best model
2023-10-25 21:20:55,135 ----------------------------------------------------------------------------------------------------
2023-10-25 21:20:56,129 epoch 5 - iter 13/136 - loss 0.01663346 - time (sec): 0.99 - samples/sec: 5001.88 - lr: 0.000020 - momentum: 0.000000
2023-10-25 21:20:57,031 epoch 5 - iter 26/136 - loss 0.02629321 - time (sec): 1.89 - samples/sec: 4870.26 - lr: 0.000019 - momentum: 0.000000
2023-10-25 21:20:58,002 epoch 5 - iter 39/136 - loss 0.02794944 - time (sec): 2.86 - samples/sec: 4938.10 - lr: 0.000019 - momentum: 0.000000
2023-10-25 21:20:58,982 epoch 5 - iter 52/136 - loss 0.02661567 - time (sec): 3.84 - samples/sec: 5046.85 - lr: 0.000019 - momentum: 0.000000
2023-10-25 21:20:59,832 epoch 5 - iter 65/136 - loss 0.02489471 - time (sec): 4.69 - samples/sec: 4924.08 - lr: 0.000018 - momentum: 0.000000
2023-10-25 21:21:00,804 epoch 5 - iter 78/136 - loss 0.02726968 - time (sec): 5.66 - samples/sec: 4960.06 - lr: 0.000018 - momentum: 0.000000
2023-10-25 21:21:01,828 epoch 5 - iter 91/136 - loss 0.02863633 - time (sec): 6.69 - samples/sec: 5056.21 - lr: 0.000018 - momentum: 0.000000
2023-10-25 21:21:02,735 epoch 5 - iter 104/136 - loss 0.02720610 - time (sec): 7.60 - samples/sec: 5102.12 - lr: 0.000018 - momentum: 0.000000
2023-10-25 21:21:03,693 epoch 5 - iter 117/136 - loss 0.02758476 - time (sec): 8.55 - samples/sec: 5141.63 - lr: 0.000017 - momentum: 0.000000
2023-10-25 21:21:04,781 epoch 5 - iter 130/136 - loss 0.02952936 - time (sec): 9.64 - samples/sec: 5162.52 - lr: 0.000017 - momentum: 0.000000
2023-10-25 21:21:05,187 ----------------------------------------------------------------------------------------------------
2023-10-25 21:21:05,187 EPOCH 5 done: loss 0.0292 - lr: 0.000017
2023-10-25 21:21:06,376 DEV : loss 0.12244772911071777 - f1-score (micro avg) 0.7927
2023-10-25 21:21:06,382 saving best model
2023-10-25 21:21:07,096 ----------------------------------------------------------------------------------------------------
2023-10-25 21:21:08,438 epoch 6 - iter 13/136 - loss 0.01683556 - time (sec): 1.34 - samples/sec: 3876.93 - lr: 0.000016 - momentum: 0.000000
2023-10-25 21:21:09,366 epoch 6 - iter 26/136 - loss 0.01690694 - time (sec): 2.27 - samples/sec: 4564.47 - lr: 0.000016 - momentum: 0.000000
2023-10-25 21:21:10,317 epoch 6 - iter 39/136 - loss 0.01687007 - time (sec): 3.22 - samples/sec: 4624.06 - lr: 0.000016 - momentum: 0.000000
2023-10-25 21:21:11,401 epoch 6 - iter 52/136 - loss 0.01889602 - time (sec): 4.30 - samples/sec: 4578.36 - lr: 0.000015 - momentum: 0.000000
2023-10-25 21:21:12,369 epoch 6 - iter 65/136 - loss 0.01614339 - time (sec): 5.27 - samples/sec: 4657.62 - lr: 0.000015 - momentum: 0.000000
2023-10-25 21:21:13,359 epoch 6 - iter 78/136 - loss 0.01655885 - time (sec): 6.26 - samples/sec: 4801.57 - lr: 0.000015 - momentum: 0.000000
2023-10-25 21:21:14,288 epoch 6 - iter 91/136 - loss 0.01946751 - time (sec): 7.19 - samples/sec: 4851.68 - lr: 0.000015 - momentum: 0.000000
2023-10-25 21:21:15,295 epoch 6 - iter 104/136 - loss 0.01892046 - time (sec): 8.19 - samples/sec: 4848.33 - lr: 0.000014 - momentum: 0.000000
2023-10-25 21:21:16,230 epoch 6 - iter 117/136 - loss 0.01916421 - time (sec): 9.13 - samples/sec: 4873.62 - lr: 0.000014 - momentum: 0.000000
2023-10-25 21:21:17,157 epoch 6 - iter 130/136 - loss 0.01967563 - time (sec): 10.06 - samples/sec: 4894.41 - lr: 0.000014 - momentum: 0.000000
2023-10-25 21:21:17,580 ----------------------------------------------------------------------------------------------------
2023-10-25 21:21:17,580 EPOCH 6 done: loss 0.0194 - lr: 0.000014
2023-10-25 21:21:18,716 DEV : loss 0.1296090930700302 - f1-score (micro avg) 0.7934
2023-10-25 21:21:18,722 saving best model
2023-10-25 21:21:19,434 ----------------------------------------------------------------------------------------------------
2023-10-25 21:21:20,348 epoch 7 - iter 13/136 - loss 0.01080744 - time (sec): 0.91 - samples/sec: 5668.49 - lr: 0.000013 - momentum: 0.000000
2023-10-25 21:21:21,361 epoch 7 - iter 26/136 - loss 0.01106692 - time (sec): 1.92 - samples/sec: 5326.04 - lr: 0.000013 - momentum: 0.000000
2023-10-25 21:21:22,288 epoch 7 - iter 39/136 - loss 0.01073739 - time (sec): 2.85 - samples/sec: 5389.66 - lr: 0.000012 - momentum: 0.000000
2023-10-25 21:21:23,287 epoch 7 - iter 52/136 - loss 0.01029492 - time (sec): 3.85 - samples/sec: 5354.39 - lr: 0.000012 - momentum: 0.000000
2023-10-25 21:21:24,361 epoch 7 - iter 65/136 - loss 0.01122951 - time (sec): 4.92 - samples/sec: 5259.47 - lr: 0.000012 - momentum: 0.000000
2023-10-25 21:21:25,294 epoch 7 - iter 78/136 - loss 0.01059452 - time (sec): 5.86 - samples/sec: 5231.06 - lr: 0.000012 - momentum: 0.000000
2023-10-25 21:21:26,352 epoch 7 - iter 91/136 - loss 0.01186817 - time (sec): 6.92 - samples/sec: 5155.37 - lr: 0.000011 - momentum: 0.000000
2023-10-25 21:21:27,230 epoch 7 - iter 104/136 - loss 0.01209798 - time (sec): 7.79 - samples/sec: 5208.33 - lr: 0.000011 - momentum: 0.000000
2023-10-25 21:21:28,112 epoch 7 - iter 117/136 - loss 0.01394009 - time (sec): 8.68 - samples/sec: 5185.34 - lr: 0.000011 - momentum: 0.000000
2023-10-25 21:21:29,033 epoch 7 - iter 130/136 - loss 0.01372476 - time (sec): 9.60 - samples/sec: 5200.08 - lr: 0.000010 - momentum: 0.000000
2023-10-25 21:21:29,485 ----------------------------------------------------------------------------------------------------
2023-10-25 21:21:29,486 EPOCH 7 done: loss 0.0136 - lr: 0.000010
2023-10-25 21:21:30,725 DEV : loss 0.1454724222421646 - f1-score (micro avg) 0.792
2023-10-25 21:21:30,732 ----------------------------------------------------------------------------------------------------
2023-10-25 21:21:32,074 epoch 8 - iter 13/136 - loss 0.01218010 - time (sec): 1.34 - samples/sec: 4384.56 - lr: 0.000010 - momentum: 0.000000
2023-10-25 21:21:33,025 epoch 8 - iter 26/136 - loss 0.01693779 - time (sec): 2.29 - samples/sec: 4772.46 - lr: 0.000009 - momentum: 0.000000
2023-10-25 21:21:34,093 epoch 8 - iter 39/136 - loss 0.01703926 - time (sec): 3.36 - samples/sec: 4856.76 - lr: 0.000009 - momentum: 0.000000
2023-10-25 21:21:34,973 epoch 8 - iter 52/136 - loss 0.01542773 - time (sec): 4.24 - samples/sec: 4941.74 - lr: 0.000009 - momentum: 0.000000
2023-10-25 21:21:35,998 epoch 8 - iter 65/136 - loss 0.01460635 - time (sec): 5.26 - samples/sec: 4954.47 - lr: 0.000009 - momentum: 0.000000
2023-10-25 21:21:36,935 epoch 8 - iter 78/136 - loss 0.01371676 - time (sec): 6.20 - samples/sec: 5011.86 - lr: 0.000008 - momentum: 0.000000
2023-10-25 21:21:37,953 epoch 8 - iter 91/136 - loss 0.01319613 - time (sec): 7.22 - samples/sec: 4994.88 - lr: 0.000008 - momentum: 0.000000
2023-10-25 21:21:38,901 epoch 8 - iter 104/136 - loss 0.01203454 - time (sec): 8.17 - samples/sec: 5027.05 - lr: 0.000008 - momentum: 0.000000
2023-10-25 21:21:39,909 epoch 8 - iter 117/136 - loss 0.01120187 - time (sec): 9.18 - samples/sec: 4985.48 - lr: 0.000007 - momentum: 0.000000
2023-10-25 21:21:40,947 epoch 8 - iter 130/136 - loss 0.01051438 - time (sec): 10.21 - samples/sec: 4914.41 - lr: 0.000007 - momentum: 0.000000
2023-10-25 21:21:41,391 ----------------------------------------------------------------------------------------------------
2023-10-25 21:21:41,392 EPOCH 8 done: loss 0.0106 - lr: 0.000007
2023-10-25 21:21:42,626 DEV : loss 0.1543840914964676 - f1-score (micro avg) 0.8059
2023-10-25 21:21:42,632 saving best model
2023-10-25 21:21:43,341 ----------------------------------------------------------------------------------------------------
2023-10-25 21:21:44,407 epoch 9 - iter 13/136 - loss 0.00445394 - time (sec): 1.06 - samples/sec: 4658.17 - lr: 0.000006 - momentum: 0.000000
2023-10-25 21:21:45,462 epoch 9 - iter 26/136 - loss 0.00733912 - time (sec): 2.12 - samples/sec: 5035.08 - lr: 0.000006 - momentum: 0.000000
2023-10-25 21:21:46,388 epoch 9 - iter 39/136 - loss 0.00679164 - time (sec): 3.05 - samples/sec: 5053.37 - lr: 0.000006 - momentum: 0.000000
2023-10-25 21:21:47,268 epoch 9 - iter 52/136 - loss 0.00759414 - time (sec): 3.93 - samples/sec: 4980.00 - lr: 0.000006 - momentum: 0.000000
2023-10-25 21:21:48,269 epoch 9 - iter 65/136 - loss 0.00752308 - time (sec): 4.93 - samples/sec: 5085.53 - lr: 0.000005 - momentum: 0.000000
2023-10-25 21:21:49,176 epoch 9 - iter 78/136 - loss 0.00768783 - time (sec): 5.83 - samples/sec: 5083.57 - lr: 0.000005 - momentum: 0.000000
2023-10-25 21:21:50,246 epoch 9 - iter 91/136 - loss 0.00732468 - time (sec): 6.90 - samples/sec: 5165.41 - lr: 0.000005 - momentum: 0.000000
2023-10-25 21:21:51,141 epoch 9 - iter 104/136 - loss 0.00684016 - time (sec): 7.80 - samples/sec: 5111.40 - lr: 0.000004 - momentum: 0.000000
2023-10-25 21:21:52,117 epoch 9 - iter 117/136 - loss 0.00700631 - time (sec): 8.77 - samples/sec: 5067.64 - lr: 0.000004 - momentum: 0.000000
2023-10-25 21:21:53,131 epoch 9 - iter 130/136 - loss 0.00730047 - time (sec): 9.79 - samples/sec: 5056.15 - lr: 0.000004 - momentum: 0.000000
2023-10-25 21:21:53,632 ----------------------------------------------------------------------------------------------------
2023-10-25 21:21:53,632 EPOCH 9 done: loss 0.0075 - lr: 0.000004
2023-10-25 21:21:54,814 DEV : loss 0.16468970477581024 - f1-score (micro avg) 0.8059
2023-10-25 21:21:54,821 ----------------------------------------------------------------------------------------------------
2023-10-25 21:21:56,267 epoch 10 - iter 13/136 - loss 0.01036517 - time (sec): 1.44 - samples/sec: 3369.07 - lr: 0.000003 - momentum: 0.000000
2023-10-25 21:21:57,242 epoch 10 - iter 26/136 - loss 0.01231050 - time (sec): 2.42 - samples/sec: 4055.70 - lr: 0.000003 - momentum: 0.000000
2023-10-25 21:21:58,278 epoch 10 - iter 39/136 - loss 0.00954086 - time (sec): 3.45 - samples/sec: 4327.08 - lr: 0.000003 - momentum: 0.000000
2023-10-25 21:21:59,305 epoch 10 - iter 52/136 - loss 0.00847034 - time (sec): 4.48 - samples/sec: 4621.28 - lr: 0.000002 - momentum: 0.000000
2023-10-25 21:22:00,240 epoch 10 - iter 65/136 - loss 0.00800394 - time (sec): 5.42 - samples/sec: 4682.11 - lr: 0.000002 - momentum: 0.000000
2023-10-25 21:22:01,177 epoch 10 - iter 78/136 - loss 0.00757495 - time (sec): 6.35 - samples/sec: 4711.89 - lr: 0.000002 - momentum: 0.000000
2023-10-25 21:22:02,192 epoch 10 - iter 91/136 - loss 0.00773310 - time (sec): 7.37 - samples/sec: 4804.46 - lr: 0.000001 - momentum: 0.000000
2023-10-25 21:22:03,206 epoch 10 - iter 104/136 - loss 0.00705573 - time (sec): 8.38 - samples/sec: 4813.82 - lr: 0.000001 - momentum: 0.000000
2023-10-25 21:22:04,042 epoch 10 - iter 117/136 - loss 0.00676122 - time (sec): 9.22 - samples/sec: 4834.45 - lr: 0.000001 - momentum: 0.000000
2023-10-25 21:22:05,026 epoch 10 - iter 130/136 - loss 0.00658402 - time (sec): 10.20 - samples/sec: 4874.95 - lr: 0.000000 - momentum: 0.000000
2023-10-25 21:22:05,439 ----------------------------------------------------------------------------------------------------
2023-10-25 21:22:05,439 EPOCH 10 done: loss 0.0069 - lr: 0.000000
2023-10-25 21:22:06,638 DEV : loss 0.16538743674755096 - f1-score (micro avg) 0.7877
2023-10-25 21:22:07,172 ----------------------------------------------------------------------------------------------------
2023-10-25 21:22:07,174 Loading model from best epoch ...
2023-10-25 21:22:09,130 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
2023-10-25 21:22:11,237
Results:
- F-score (micro) 0.7892
- F-score (macro) 0.7524
- Accuracy 0.664
By class:
precision recall f1-score support
LOC 0.8201 0.8622 0.8406 312
PER 0.6973 0.8750 0.7761 208
ORG 0.5000 0.4545 0.4762 55
HumanProd 0.8462 1.0000 0.9167 22
micro avg 0.7489 0.8342 0.7892 597
macro avg 0.7159 0.7979 0.7524 597
weighted avg 0.7488 0.8342 0.7874 597
2023-10-25 21:22:11,237 ----------------------------------------------------------------------------------------------------
|