File size: 23,867 Bytes
71ebc80 |
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 21:25:37,600 ----------------------------------------------------------------------------------------------------
2023-10-25 21:25:37,601 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 21:25:37,601 ----------------------------------------------------------------------------------------------------
2023-10-25 21:25:37,601 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 21:25:37,601 ----------------------------------------------------------------------------------------------------
2023-10-25 21:25:37,601 Train: 1166 sentences
2023-10-25 21:25:37,601 (train_with_dev=False, train_with_test=False)
2023-10-25 21:25:37,601 ----------------------------------------------------------------------------------------------------
2023-10-25 21:25:37,601 Training Params:
2023-10-25 21:25:37,601 - learning_rate: "3e-05"
2023-10-25 21:25:37,601 - mini_batch_size: "8"
2023-10-25 21:25:37,601 - max_epochs: "10"
2023-10-25 21:25:37,601 - shuffle: "True"
2023-10-25 21:25:37,601 ----------------------------------------------------------------------------------------------------
2023-10-25 21:25:37,601 Plugins:
2023-10-25 21:25:37,601 - TensorboardLogger
2023-10-25 21:25:37,601 - LinearScheduler | warmup_fraction: '0.1'
2023-10-25 21:25:37,601 ----------------------------------------------------------------------------------------------------
2023-10-25 21:25:37,602 Final evaluation on model from best epoch (best-model.pt)
2023-10-25 21:25:37,602 - metric: "('micro avg', 'f1-score')"
2023-10-25 21:25:37,602 ----------------------------------------------------------------------------------------------------
2023-10-25 21:25:37,602 Computation:
2023-10-25 21:25:37,602 - compute on device: cuda:0
2023-10-25 21:25:37,602 - embedding storage: none
2023-10-25 21:25:37,602 ----------------------------------------------------------------------------------------------------
2023-10-25 21:25:37,602 Model training base path: "hmbench-newseye/fi-dbmdz/bert-base-historic-multilingual-64k-td-cased-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-5"
2023-10-25 21:25:37,602 ----------------------------------------------------------------------------------------------------
2023-10-25 21:25:37,602 ----------------------------------------------------------------------------------------------------
2023-10-25 21:25:37,602 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-25 21:25:38,456 epoch 1 - iter 14/146 - loss 2.60294278 - time (sec): 0.85 - samples/sec: 4554.56 - lr: 0.000003 - momentum: 0.000000
2023-10-25 21:25:39,251 epoch 1 - iter 28/146 - loss 2.23371535 - time (sec): 1.65 - samples/sec: 4624.06 - lr: 0.000006 - momentum: 0.000000
2023-10-25 21:25:40,234 epoch 1 - iter 42/146 - loss 1.80138209 - time (sec): 2.63 - samples/sec: 4632.61 - lr: 0.000008 - momentum: 0.000000
2023-10-25 21:25:41,163 epoch 1 - iter 56/146 - loss 1.53922194 - time (sec): 3.56 - samples/sec: 4611.81 - lr: 0.000011 - momentum: 0.000000
2023-10-25 21:25:42,048 epoch 1 - iter 70/146 - loss 1.31289657 - time (sec): 4.44 - samples/sec: 4710.80 - lr: 0.000014 - momentum: 0.000000
2023-10-25 21:25:42,819 epoch 1 - iter 84/146 - loss 1.19311406 - time (sec): 5.22 - samples/sec: 4682.69 - lr: 0.000017 - momentum: 0.000000
2023-10-25 21:25:43,898 epoch 1 - iter 98/146 - loss 1.06894666 - time (sec): 6.30 - samples/sec: 4636.30 - lr: 0.000020 - momentum: 0.000000
2023-10-25 21:25:44,851 epoch 1 - iter 112/146 - loss 0.96000516 - time (sec): 7.25 - samples/sec: 4688.87 - lr: 0.000023 - momentum: 0.000000
2023-10-25 21:25:45,925 epoch 1 - iter 126/146 - loss 0.87262223 - time (sec): 8.32 - samples/sec: 4714.46 - lr: 0.000026 - momentum: 0.000000
2023-10-25 21:25:46,732 epoch 1 - iter 140/146 - loss 0.82531461 - time (sec): 9.13 - samples/sec: 4677.43 - lr: 0.000029 - momentum: 0.000000
2023-10-25 21:25:47,124 ----------------------------------------------------------------------------------------------------
2023-10-25 21:25:47,125 EPOCH 1 done: loss 0.8031 - lr: 0.000029
2023-10-25 21:25:47,636 DEV : loss 0.17106929421424866 - f1-score (micro avg) 0.5556
2023-10-25 21:25:47,641 saving best model
2023-10-25 21:25:48,125 ----------------------------------------------------------------------------------------------------
2023-10-25 21:25:49,058 epoch 2 - iter 14/146 - loss 0.21725174 - time (sec): 0.93 - samples/sec: 4670.16 - lr: 0.000030 - momentum: 0.000000
2023-10-25 21:25:49,982 epoch 2 - iter 28/146 - loss 0.26028088 - time (sec): 1.86 - samples/sec: 4639.64 - lr: 0.000029 - momentum: 0.000000
2023-10-25 21:25:50,829 epoch 2 - iter 42/146 - loss 0.23972876 - time (sec): 2.70 - samples/sec: 4534.87 - lr: 0.000029 - momentum: 0.000000
2023-10-25 21:25:51,852 epoch 2 - iter 56/146 - loss 0.21456830 - time (sec): 3.73 - samples/sec: 4529.17 - lr: 0.000029 - momentum: 0.000000
2023-10-25 21:25:52,760 epoch 2 - iter 70/146 - loss 0.20980275 - time (sec): 4.63 - samples/sec: 4586.61 - lr: 0.000028 - momentum: 0.000000
2023-10-25 21:25:53,548 epoch 2 - iter 84/146 - loss 0.20500047 - time (sec): 5.42 - samples/sec: 4663.26 - lr: 0.000028 - momentum: 0.000000
2023-10-25 21:25:54,385 epoch 2 - iter 98/146 - loss 0.20211037 - time (sec): 6.26 - samples/sec: 4703.40 - lr: 0.000028 - momentum: 0.000000
2023-10-25 21:25:55,213 epoch 2 - iter 112/146 - loss 0.19844997 - time (sec): 7.09 - samples/sec: 4734.97 - lr: 0.000027 - momentum: 0.000000
2023-10-25 21:25:56,153 epoch 2 - iter 126/146 - loss 0.19974107 - time (sec): 8.03 - samples/sec: 4721.44 - lr: 0.000027 - momentum: 0.000000
2023-10-25 21:25:57,151 epoch 2 - iter 140/146 - loss 0.19108156 - time (sec): 9.03 - samples/sec: 4716.34 - lr: 0.000027 - momentum: 0.000000
2023-10-25 21:25:57,479 ----------------------------------------------------------------------------------------------------
2023-10-25 21:25:57,479 EPOCH 2 done: loss 0.1879 - lr: 0.000027
2023-10-25 21:25:58,559 DEV : loss 0.11886167526245117 - f1-score (micro avg) 0.628
2023-10-25 21:25:58,564 saving best model
2023-10-25 21:25:59,181 ----------------------------------------------------------------------------------------------------
2023-10-25 21:26:00,384 epoch 3 - iter 14/146 - loss 0.09759941 - time (sec): 1.20 - samples/sec: 4773.89 - lr: 0.000026 - momentum: 0.000000
2023-10-25 21:26:01,355 epoch 3 - iter 28/146 - loss 0.10716313 - time (sec): 2.17 - samples/sec: 4677.63 - lr: 0.000026 - momentum: 0.000000
2023-10-25 21:26:02,184 epoch 3 - iter 42/146 - loss 0.10722639 - time (sec): 3.00 - samples/sec: 4725.88 - lr: 0.000026 - momentum: 0.000000
2023-10-25 21:26:03,077 epoch 3 - iter 56/146 - loss 0.10523426 - time (sec): 3.89 - samples/sec: 4598.51 - lr: 0.000025 - momentum: 0.000000
2023-10-25 21:26:04,013 epoch 3 - iter 70/146 - loss 0.10392697 - time (sec): 4.83 - samples/sec: 4631.03 - lr: 0.000025 - momentum: 0.000000
2023-10-25 21:26:04,838 epoch 3 - iter 84/146 - loss 0.10454309 - time (sec): 5.65 - samples/sec: 4614.89 - lr: 0.000025 - momentum: 0.000000
2023-10-25 21:26:05,722 epoch 3 - iter 98/146 - loss 0.09893681 - time (sec): 6.54 - samples/sec: 4658.43 - lr: 0.000024 - momentum: 0.000000
2023-10-25 21:26:06,558 epoch 3 - iter 112/146 - loss 0.09875386 - time (sec): 7.38 - samples/sec: 4621.33 - lr: 0.000024 - momentum: 0.000000
2023-10-25 21:26:07,479 epoch 3 - iter 126/146 - loss 0.09913673 - time (sec): 8.30 - samples/sec: 4648.08 - lr: 0.000024 - momentum: 0.000000
2023-10-25 21:26:08,263 epoch 3 - iter 140/146 - loss 0.09719812 - time (sec): 9.08 - samples/sec: 4722.37 - lr: 0.000024 - momentum: 0.000000
2023-10-25 21:26:08,631 ----------------------------------------------------------------------------------------------------
2023-10-25 21:26:08,631 EPOCH 3 done: loss 0.0987 - lr: 0.000024
2023-10-25 21:26:09,551 DEV : loss 0.10990928113460541 - f1-score (micro avg) 0.7281
2023-10-25 21:26:09,556 saving best model
2023-10-25 21:26:10,167 ----------------------------------------------------------------------------------------------------
2023-10-25 21:26:10,976 epoch 4 - iter 14/146 - loss 0.08318771 - time (sec): 0.81 - samples/sec: 4774.15 - lr: 0.000023 - momentum: 0.000000
2023-10-25 21:26:11,965 epoch 4 - iter 28/146 - loss 0.06355907 - time (sec): 1.80 - samples/sec: 4570.83 - lr: 0.000023 - momentum: 0.000000
2023-10-25 21:26:12,846 epoch 4 - iter 42/146 - loss 0.05940453 - time (sec): 2.68 - samples/sec: 4712.30 - lr: 0.000022 - momentum: 0.000000
2023-10-25 21:26:13,683 epoch 4 - iter 56/146 - loss 0.05624618 - time (sec): 3.51 - samples/sec: 4578.48 - lr: 0.000022 - momentum: 0.000000
2023-10-25 21:26:14,614 epoch 4 - iter 70/146 - loss 0.05687332 - time (sec): 4.44 - samples/sec: 4793.07 - lr: 0.000022 - momentum: 0.000000
2023-10-25 21:26:15,618 epoch 4 - iter 84/146 - loss 0.05866038 - time (sec): 5.45 - samples/sec: 4832.66 - lr: 0.000021 - momentum: 0.000000
2023-10-25 21:26:16,470 epoch 4 - iter 98/146 - loss 0.05691631 - time (sec): 6.30 - samples/sec: 4897.68 - lr: 0.000021 - momentum: 0.000000
2023-10-25 21:26:17,311 epoch 4 - iter 112/146 - loss 0.06125083 - time (sec): 7.14 - samples/sec: 4852.90 - lr: 0.000021 - momentum: 0.000000
2023-10-25 21:26:18,415 epoch 4 - iter 126/146 - loss 0.06362815 - time (sec): 8.25 - samples/sec: 4759.95 - lr: 0.000021 - momentum: 0.000000
2023-10-25 21:26:19,217 epoch 4 - iter 140/146 - loss 0.06182318 - time (sec): 9.05 - samples/sec: 4740.79 - lr: 0.000020 - momentum: 0.000000
2023-10-25 21:26:19,541 ----------------------------------------------------------------------------------------------------
2023-10-25 21:26:19,541 EPOCH 4 done: loss 0.0618 - lr: 0.000020
2023-10-25 21:26:20,462 DEV : loss 0.0920729711651802 - f1-score (micro avg) 0.7702
2023-10-25 21:26:20,467 saving best model
2023-10-25 21:26:20,948 ----------------------------------------------------------------------------------------------------
2023-10-25 21:26:21,871 epoch 5 - iter 14/146 - loss 0.03669754 - time (sec): 0.92 - samples/sec: 4823.31 - lr: 0.000020 - momentum: 0.000000
2023-10-25 21:26:22,778 epoch 5 - iter 28/146 - loss 0.03920318 - time (sec): 1.83 - samples/sec: 4663.07 - lr: 0.000019 - momentum: 0.000000
2023-10-25 21:26:23,664 epoch 5 - iter 42/146 - loss 0.03549894 - time (sec): 2.71 - samples/sec: 4675.88 - lr: 0.000019 - momentum: 0.000000
2023-10-25 21:26:24,500 epoch 5 - iter 56/146 - loss 0.03524770 - time (sec): 3.55 - samples/sec: 4765.02 - lr: 0.000019 - momentum: 0.000000
2023-10-25 21:26:25,348 epoch 5 - iter 70/146 - loss 0.03474532 - time (sec): 4.40 - samples/sec: 4816.39 - lr: 0.000018 - momentum: 0.000000
2023-10-25 21:26:26,121 epoch 5 - iter 84/146 - loss 0.03887679 - time (sec): 5.17 - samples/sec: 4820.26 - lr: 0.000018 - momentum: 0.000000
2023-10-25 21:26:27,253 epoch 5 - iter 98/146 - loss 0.04147485 - time (sec): 6.30 - samples/sec: 4715.08 - lr: 0.000018 - momentum: 0.000000
2023-10-25 21:26:28,113 epoch 5 - iter 112/146 - loss 0.04145546 - time (sec): 7.16 - samples/sec: 4758.84 - lr: 0.000018 - momentum: 0.000000
2023-10-25 21:26:29,050 epoch 5 - iter 126/146 - loss 0.04076736 - time (sec): 8.10 - samples/sec: 4785.00 - lr: 0.000017 - momentum: 0.000000
2023-10-25 21:26:29,887 epoch 5 - iter 140/146 - loss 0.03908635 - time (sec): 8.94 - samples/sec: 4809.16 - lr: 0.000017 - momentum: 0.000000
2023-10-25 21:26:30,239 ----------------------------------------------------------------------------------------------------
2023-10-25 21:26:30,239 EPOCH 5 done: loss 0.0394 - lr: 0.000017
2023-10-25 21:26:31,319 DEV : loss 0.11569201946258545 - f1-score (micro avg) 0.7261
2023-10-25 21:26:31,324 ----------------------------------------------------------------------------------------------------
2023-10-25 21:26:32,244 epoch 6 - iter 14/146 - loss 0.03038153 - time (sec): 0.92 - samples/sec: 5188.57 - lr: 0.000016 - momentum: 0.000000
2023-10-25 21:26:33,136 epoch 6 - iter 28/146 - loss 0.03130799 - time (sec): 1.81 - samples/sec: 4906.54 - lr: 0.000016 - momentum: 0.000000
2023-10-25 21:26:33,995 epoch 6 - iter 42/146 - loss 0.02641538 - time (sec): 2.67 - samples/sec: 4942.07 - lr: 0.000016 - momentum: 0.000000
2023-10-25 21:26:34,916 epoch 6 - iter 56/146 - loss 0.03146130 - time (sec): 3.59 - samples/sec: 4891.39 - lr: 0.000015 - momentum: 0.000000
2023-10-25 21:26:35,779 epoch 6 - iter 70/146 - loss 0.03030920 - time (sec): 4.45 - samples/sec: 4938.51 - lr: 0.000015 - momentum: 0.000000
2023-10-25 21:26:36,627 epoch 6 - iter 84/146 - loss 0.02957139 - time (sec): 5.30 - samples/sec: 4865.32 - lr: 0.000015 - momentum: 0.000000
2023-10-25 21:26:37,593 epoch 6 - iter 98/146 - loss 0.02770553 - time (sec): 6.27 - samples/sec: 4793.74 - lr: 0.000015 - momentum: 0.000000
2023-10-25 21:26:38,516 epoch 6 - iter 112/146 - loss 0.02699072 - time (sec): 7.19 - samples/sec: 4747.91 - lr: 0.000014 - momentum: 0.000000
2023-10-25 21:26:39,578 epoch 6 - iter 126/146 - loss 0.02595757 - time (sec): 8.25 - samples/sec: 4701.74 - lr: 0.000014 - momentum: 0.000000
2023-10-25 21:26:40,381 epoch 6 - iter 140/146 - loss 0.02607979 - time (sec): 9.06 - samples/sec: 4699.00 - lr: 0.000014 - momentum: 0.000000
2023-10-25 21:26:40,763 ----------------------------------------------------------------------------------------------------
2023-10-25 21:26:40,764 EPOCH 6 done: loss 0.0259 - lr: 0.000014
2023-10-25 21:26:41,690 DEV : loss 0.12295451760292053 - f1-score (micro avg) 0.7401
2023-10-25 21:26:41,696 ----------------------------------------------------------------------------------------------------
2023-10-25 21:26:42,577 epoch 7 - iter 14/146 - loss 0.02415144 - time (sec): 0.88 - samples/sec: 5234.42 - lr: 0.000013 - momentum: 0.000000
2023-10-25 21:26:43,586 epoch 7 - iter 28/146 - loss 0.03236498 - time (sec): 1.89 - samples/sec: 4997.27 - lr: 0.000013 - momentum: 0.000000
2023-10-25 21:26:44,412 epoch 7 - iter 42/146 - loss 0.03067054 - time (sec): 2.71 - samples/sec: 4934.20 - lr: 0.000012 - momentum: 0.000000
2023-10-25 21:26:45,263 epoch 7 - iter 56/146 - loss 0.02611424 - time (sec): 3.57 - samples/sec: 4843.92 - lr: 0.000012 - momentum: 0.000000
2023-10-25 21:26:46,053 epoch 7 - iter 70/146 - loss 0.02398715 - time (sec): 4.36 - samples/sec: 4818.12 - lr: 0.000012 - momentum: 0.000000
2023-10-25 21:26:47,091 epoch 7 - iter 84/146 - loss 0.02253424 - time (sec): 5.39 - samples/sec: 4774.42 - lr: 0.000012 - momentum: 0.000000
2023-10-25 21:26:47,983 epoch 7 - iter 98/146 - loss 0.02258653 - time (sec): 6.29 - samples/sec: 4831.43 - lr: 0.000011 - momentum: 0.000000
2023-10-25 21:26:48,818 epoch 7 - iter 112/146 - loss 0.02125092 - time (sec): 7.12 - samples/sec: 4805.93 - lr: 0.000011 - momentum: 0.000000
2023-10-25 21:26:49,735 epoch 7 - iter 126/146 - loss 0.02038749 - time (sec): 8.04 - samples/sec: 4787.52 - lr: 0.000011 - momentum: 0.000000
2023-10-25 21:26:50,593 epoch 7 - iter 140/146 - loss 0.02016293 - time (sec): 8.90 - samples/sec: 4762.46 - lr: 0.000010 - momentum: 0.000000
2023-10-25 21:26:51,047 ----------------------------------------------------------------------------------------------------
2023-10-25 21:26:51,048 EPOCH 7 done: loss 0.0195 - lr: 0.000010
2023-10-25 21:26:51,973 DEV : loss 0.1449124813079834 - f1-score (micro avg) 0.7158
2023-10-25 21:26:51,978 ----------------------------------------------------------------------------------------------------
2023-10-25 21:26:52,924 epoch 8 - iter 14/146 - loss 0.01568499 - time (sec): 0.95 - samples/sec: 4492.68 - lr: 0.000010 - momentum: 0.000000
2023-10-25 21:26:53,864 epoch 8 - iter 28/146 - loss 0.01851063 - time (sec): 1.88 - samples/sec: 4485.90 - lr: 0.000009 - momentum: 0.000000
2023-10-25 21:26:54,698 epoch 8 - iter 42/146 - loss 0.01478075 - time (sec): 2.72 - samples/sec: 4596.52 - lr: 0.000009 - momentum: 0.000000
2023-10-25 21:26:55,645 epoch 8 - iter 56/146 - loss 0.01476943 - time (sec): 3.67 - samples/sec: 4680.34 - lr: 0.000009 - momentum: 0.000000
2023-10-25 21:26:56,459 epoch 8 - iter 70/146 - loss 0.01459552 - time (sec): 4.48 - samples/sec: 4686.56 - lr: 0.000009 - momentum: 0.000000
2023-10-25 21:26:57,274 epoch 8 - iter 84/146 - loss 0.01588161 - time (sec): 5.30 - samples/sec: 4758.94 - lr: 0.000008 - momentum: 0.000000
2023-10-25 21:26:58,098 epoch 8 - iter 98/146 - loss 0.01515460 - time (sec): 6.12 - samples/sec: 4735.78 - lr: 0.000008 - momentum: 0.000000
2023-10-25 21:26:59,012 epoch 8 - iter 112/146 - loss 0.01564759 - time (sec): 7.03 - samples/sec: 4711.07 - lr: 0.000008 - momentum: 0.000000
2023-10-25 21:26:59,942 epoch 8 - iter 126/146 - loss 0.01592133 - time (sec): 7.96 - samples/sec: 4714.23 - lr: 0.000007 - momentum: 0.000000
2023-10-25 21:27:00,945 epoch 8 - iter 140/146 - loss 0.01519304 - time (sec): 8.97 - samples/sec: 4745.85 - lr: 0.000007 - momentum: 0.000000
2023-10-25 21:27:01,340 ----------------------------------------------------------------------------------------------------
2023-10-25 21:27:01,340 EPOCH 8 done: loss 0.0159 - lr: 0.000007
2023-10-25 21:27:02,424 DEV : loss 0.14217789471149445 - f1-score (micro avg) 0.755
2023-10-25 21:27:02,430 ----------------------------------------------------------------------------------------------------
2023-10-25 21:27:03,382 epoch 9 - iter 14/146 - loss 0.00940644 - time (sec): 0.95 - samples/sec: 5197.42 - lr: 0.000006 - momentum: 0.000000
2023-10-25 21:27:04,214 epoch 9 - iter 28/146 - loss 0.01224290 - time (sec): 1.78 - samples/sec: 5124.09 - lr: 0.000006 - momentum: 0.000000
2023-10-25 21:27:05,021 epoch 9 - iter 42/146 - loss 0.01134241 - time (sec): 2.59 - samples/sec: 5030.86 - lr: 0.000006 - momentum: 0.000000
2023-10-25 21:27:05,958 epoch 9 - iter 56/146 - loss 0.01127674 - time (sec): 3.53 - samples/sec: 5002.42 - lr: 0.000006 - momentum: 0.000000
2023-10-25 21:27:06,926 epoch 9 - iter 70/146 - loss 0.01508017 - time (sec): 4.50 - samples/sec: 4872.49 - lr: 0.000005 - momentum: 0.000000
2023-10-25 21:27:07,920 epoch 9 - iter 84/146 - loss 0.01540634 - time (sec): 5.49 - samples/sec: 4773.73 - lr: 0.000005 - momentum: 0.000000
2023-10-25 21:27:08,821 epoch 9 - iter 98/146 - loss 0.01387577 - time (sec): 6.39 - samples/sec: 4775.83 - lr: 0.000005 - momentum: 0.000000
2023-10-25 21:27:09,723 epoch 9 - iter 112/146 - loss 0.01346312 - time (sec): 7.29 - samples/sec: 4762.16 - lr: 0.000004 - momentum: 0.000000
2023-10-25 21:27:10,619 epoch 9 - iter 126/146 - loss 0.01415713 - time (sec): 8.19 - samples/sec: 4712.60 - lr: 0.000004 - momentum: 0.000000
2023-10-25 21:27:11,529 epoch 9 - iter 140/146 - loss 0.01466539 - time (sec): 9.10 - samples/sec: 4690.74 - lr: 0.000004 - momentum: 0.000000
2023-10-25 21:27:11,876 ----------------------------------------------------------------------------------------------------
2023-10-25 21:27:11,877 EPOCH 9 done: loss 0.0142 - lr: 0.000004
2023-10-25 21:27:12,796 DEV : loss 0.14243099093437195 - f1-score (micro avg) 0.7387
2023-10-25 21:27:12,801 ----------------------------------------------------------------------------------------------------
2023-10-25 21:27:13,674 epoch 10 - iter 14/146 - loss 0.00899830 - time (sec): 0.87 - samples/sec: 4927.08 - lr: 0.000003 - momentum: 0.000000
2023-10-25 21:27:14,530 epoch 10 - iter 28/146 - loss 0.00557953 - time (sec): 1.73 - samples/sec: 4608.78 - lr: 0.000003 - momentum: 0.000000
2023-10-25 21:27:15,417 epoch 10 - iter 42/146 - loss 0.01050729 - time (sec): 2.61 - samples/sec: 4620.84 - lr: 0.000003 - momentum: 0.000000
2023-10-25 21:27:16,202 epoch 10 - iter 56/146 - loss 0.01109768 - time (sec): 3.40 - samples/sec: 4647.40 - lr: 0.000002 - momentum: 0.000000
2023-10-25 21:27:17,172 epoch 10 - iter 70/146 - loss 0.01198941 - time (sec): 4.37 - samples/sec: 4708.51 - lr: 0.000002 - momentum: 0.000000
2023-10-25 21:27:18,082 epoch 10 - iter 84/146 - loss 0.01146295 - time (sec): 5.28 - samples/sec: 4686.69 - lr: 0.000002 - momentum: 0.000000
2023-10-25 21:27:18,904 epoch 10 - iter 98/146 - loss 0.01083949 - time (sec): 6.10 - samples/sec: 4773.97 - lr: 0.000001 - momentum: 0.000000
2023-10-25 21:27:19,896 epoch 10 - iter 112/146 - loss 0.01100900 - time (sec): 7.09 - samples/sec: 4798.30 - lr: 0.000001 - momentum: 0.000000
2023-10-25 21:27:20,832 epoch 10 - iter 126/146 - loss 0.01016506 - time (sec): 8.03 - samples/sec: 4761.08 - lr: 0.000001 - momentum: 0.000000
2023-10-25 21:27:21,781 epoch 10 - iter 140/146 - loss 0.01013197 - time (sec): 8.98 - samples/sec: 4791.19 - lr: 0.000000 - momentum: 0.000000
2023-10-25 21:27:22,091 ----------------------------------------------------------------------------------------------------
2023-10-25 21:27:22,091 EPOCH 10 done: loss 0.0099 - lr: 0.000000
2023-10-25 21:27:23,012 DEV : loss 0.14977367222309113 - f1-score (micro avg) 0.7419
2023-10-25 21:27:23,485 ----------------------------------------------------------------------------------------------------
2023-10-25 21:27:23,486 Loading model from best epoch ...
2023-10-25 21:27:25,065 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 21:27:26,606
Results:
- F-score (micro) 0.7631
- F-score (macro) 0.6653
- Accuracy 0.6408
By class:
precision recall f1-score support
PER 0.7919 0.8420 0.8162 348
LOC 0.7026 0.8238 0.7584 261
ORG 0.5111 0.4423 0.4742 52
HumanProd 0.5556 0.6818 0.6122 22
micro avg 0.7299 0.7994 0.7631 683
macro avg 0.6403 0.6975 0.6653 683
weighted avg 0.7288 0.7994 0.7615 683
2023-10-25 21:27:26,606 ----------------------------------------------------------------------------------------------------
|