File size: 24,035 Bytes
9e33766 |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 |
2023-10-25 21:31:24,935 ----------------------------------------------------------------------------------------------------
2023-10-25 21:31:24,936 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 21:31:24,936 ----------------------------------------------------------------------------------------------------
2023-10-25 21:31:24,936 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 21:31:24,936 ----------------------------------------------------------------------------------------------------
2023-10-25 21:31:24,937 Train: 1085 sentences
2023-10-25 21:31:24,937 (train_with_dev=False, train_with_test=False)
2023-10-25 21:31:24,937 ----------------------------------------------------------------------------------------------------
2023-10-25 21:31:24,937 Training Params:
2023-10-25 21:31:24,937 - learning_rate: "3e-05"
2023-10-25 21:31:24,937 - mini_batch_size: "8"
2023-10-25 21:31:24,937 - max_epochs: "10"
2023-10-25 21:31:24,937 - shuffle: "True"
2023-10-25 21:31:24,937 ----------------------------------------------------------------------------------------------------
2023-10-25 21:31:24,937 Plugins:
2023-10-25 21:31:24,937 - TensorboardLogger
2023-10-25 21:31:24,937 - LinearScheduler | warmup_fraction: '0.1'
2023-10-25 21:31:24,937 ----------------------------------------------------------------------------------------------------
2023-10-25 21:31:24,937 Final evaluation on model from best epoch (best-model.pt)
2023-10-25 21:31:24,937 - metric: "('micro avg', 'f1-score')"
2023-10-25 21:31:24,937 ----------------------------------------------------------------------------------------------------
2023-10-25 21:31:24,937 Computation:
2023-10-25 21:31:24,937 - compute on device: cuda:0
2023-10-25 21:31:24,937 - embedding storage: none
2023-10-25 21:31:24,937 ----------------------------------------------------------------------------------------------------
2023-10-25 21:31:24,937 Model training base path: "hmbench-newseye/sv-dbmdz/bert-base-historic-multilingual-64k-td-cased-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-5"
2023-10-25 21:31:24,937 ----------------------------------------------------------------------------------------------------
2023-10-25 21:31:24,937 ----------------------------------------------------------------------------------------------------
2023-10-25 21:31:24,937 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-25 21:31:25,855 epoch 1 - iter 13/136 - loss 2.67862976 - time (sec): 0.92 - samples/sec: 5455.64 - lr: 0.000003 - momentum: 0.000000
2023-10-25 21:31:26,870 epoch 1 - iter 26/136 - loss 2.31862643 - time (sec): 1.93 - samples/sec: 5241.39 - lr: 0.000006 - momentum: 0.000000
2023-10-25 21:31:27,857 epoch 1 - iter 39/136 - loss 1.79659135 - time (sec): 2.92 - samples/sec: 5245.24 - lr: 0.000008 - momentum: 0.000000
2023-10-25 21:31:28,918 epoch 1 - iter 52/136 - loss 1.46309717 - time (sec): 3.98 - samples/sec: 5273.26 - lr: 0.000011 - momentum: 0.000000
2023-10-25 21:31:29,964 epoch 1 - iter 65/136 - loss 1.28129944 - time (sec): 5.03 - samples/sec: 5140.43 - lr: 0.000014 - momentum: 0.000000
2023-10-25 21:31:31,013 epoch 1 - iter 78/136 - loss 1.14224358 - time (sec): 6.07 - samples/sec: 5077.33 - lr: 0.000017 - momentum: 0.000000
2023-10-25 21:31:32,039 epoch 1 - iter 91/136 - loss 1.02123046 - time (sec): 7.10 - samples/sec: 5116.90 - lr: 0.000020 - momentum: 0.000000
2023-10-25 21:31:33,016 epoch 1 - iter 104/136 - loss 0.93704629 - time (sec): 8.08 - samples/sec: 5082.36 - lr: 0.000023 - momentum: 0.000000
2023-10-25 21:31:34,048 epoch 1 - iter 117/136 - loss 0.86134920 - time (sec): 9.11 - samples/sec: 5049.79 - lr: 0.000026 - momentum: 0.000000
2023-10-25 21:31:34,972 epoch 1 - iter 130/136 - loss 0.81315635 - time (sec): 10.03 - samples/sec: 4978.28 - lr: 0.000028 - momentum: 0.000000
2023-10-25 21:31:35,395 ----------------------------------------------------------------------------------------------------
2023-10-25 21:31:35,395 EPOCH 1 done: loss 0.7886 - lr: 0.000028
2023-10-25 21:31:36,496 DEV : loss 0.15381869673728943 - f1-score (micro avg) 0.6475
2023-10-25 21:31:36,503 saving best model
2023-10-25 21:31:37,020 ----------------------------------------------------------------------------------------------------
2023-10-25 21:31:38,009 epoch 2 - iter 13/136 - loss 0.14729279 - time (sec): 0.99 - samples/sec: 5174.73 - lr: 0.000030 - momentum: 0.000000
2023-10-25 21:31:38,993 epoch 2 - iter 26/136 - loss 0.17121608 - time (sec): 1.97 - samples/sec: 5363.99 - lr: 0.000029 - momentum: 0.000000
2023-10-25 21:31:40,025 epoch 2 - iter 39/136 - loss 0.16370443 - time (sec): 3.00 - samples/sec: 4965.75 - lr: 0.000029 - momentum: 0.000000
2023-10-25 21:31:41,014 epoch 2 - iter 52/136 - loss 0.15904671 - time (sec): 3.99 - samples/sec: 4935.00 - lr: 0.000029 - momentum: 0.000000
2023-10-25 21:31:41,969 epoch 2 - iter 65/136 - loss 0.15169814 - time (sec): 4.95 - samples/sec: 4959.65 - lr: 0.000028 - momentum: 0.000000
2023-10-25 21:31:42,925 epoch 2 - iter 78/136 - loss 0.15515706 - time (sec): 5.90 - samples/sec: 5047.51 - lr: 0.000028 - momentum: 0.000000
2023-10-25 21:31:43,922 epoch 2 - iter 91/136 - loss 0.15141408 - time (sec): 6.90 - samples/sec: 5069.46 - lr: 0.000028 - momentum: 0.000000
2023-10-25 21:31:44,959 epoch 2 - iter 104/136 - loss 0.15001351 - time (sec): 7.94 - samples/sec: 4974.03 - lr: 0.000027 - momentum: 0.000000
2023-10-25 21:31:45,937 epoch 2 - iter 117/136 - loss 0.14937968 - time (sec): 8.92 - samples/sec: 5053.65 - lr: 0.000027 - momentum: 0.000000
2023-10-25 21:31:46,900 epoch 2 - iter 130/136 - loss 0.14671523 - time (sec): 9.88 - samples/sec: 5019.57 - lr: 0.000027 - momentum: 0.000000
2023-10-25 21:31:47,358 ----------------------------------------------------------------------------------------------------
2023-10-25 21:31:47,359 EPOCH 2 done: loss 0.1457 - lr: 0.000027
2023-10-25 21:31:48,659 DEV : loss 0.10822859406471252 - f1-score (micro avg) 0.7601
2023-10-25 21:31:48,666 saving best model
2023-10-25 21:31:49,411 ----------------------------------------------------------------------------------------------------
2023-10-25 21:31:50,343 epoch 3 - iter 13/136 - loss 0.09159492 - time (sec): 0.93 - samples/sec: 4581.24 - lr: 0.000026 - momentum: 0.000000
2023-10-25 21:31:51,270 epoch 3 - iter 26/136 - loss 0.09188390 - time (sec): 1.86 - samples/sec: 4829.89 - lr: 0.000026 - momentum: 0.000000
2023-10-25 21:31:52,317 epoch 3 - iter 39/136 - loss 0.07817653 - time (sec): 2.90 - samples/sec: 4913.41 - lr: 0.000026 - momentum: 0.000000
2023-10-25 21:31:53,237 epoch 3 - iter 52/136 - loss 0.07930223 - time (sec): 3.82 - samples/sec: 4985.05 - lr: 0.000025 - momentum: 0.000000
2023-10-25 21:31:54,301 epoch 3 - iter 65/136 - loss 0.07711853 - time (sec): 4.89 - samples/sec: 4930.78 - lr: 0.000025 - momentum: 0.000000
2023-10-25 21:31:55,353 epoch 3 - iter 78/136 - loss 0.07404006 - time (sec): 5.94 - samples/sec: 5104.97 - lr: 0.000025 - momentum: 0.000000
2023-10-25 21:31:56,460 epoch 3 - iter 91/136 - loss 0.07528584 - time (sec): 7.05 - samples/sec: 5062.43 - lr: 0.000024 - momentum: 0.000000
2023-10-25 21:31:57,382 epoch 3 - iter 104/136 - loss 0.07545838 - time (sec): 7.97 - samples/sec: 5024.83 - lr: 0.000024 - momentum: 0.000000
2023-10-25 21:31:58,355 epoch 3 - iter 117/136 - loss 0.07590779 - time (sec): 8.94 - samples/sec: 4957.33 - lr: 0.000024 - momentum: 0.000000
2023-10-25 21:31:59,312 epoch 3 - iter 130/136 - loss 0.07606308 - time (sec): 9.90 - samples/sec: 4978.39 - lr: 0.000024 - momentum: 0.000000
2023-10-25 21:31:59,829 ----------------------------------------------------------------------------------------------------
2023-10-25 21:31:59,830 EPOCH 3 done: loss 0.0758 - lr: 0.000024
2023-10-25 21:32:01,578 DEV : loss 0.10140043497085571 - f1-score (micro avg) 0.7633
2023-10-25 21:32:01,585 saving best model
2023-10-25 21:32:02,308 ----------------------------------------------------------------------------------------------------
2023-10-25 21:32:03,306 epoch 4 - iter 13/136 - loss 0.04720147 - time (sec): 1.00 - samples/sec: 5398.05 - lr: 0.000023 - momentum: 0.000000
2023-10-25 21:32:04,364 epoch 4 - iter 26/136 - loss 0.04723812 - time (sec): 2.05 - samples/sec: 5475.58 - lr: 0.000023 - momentum: 0.000000
2023-10-25 21:32:05,451 epoch 4 - iter 39/136 - loss 0.04375497 - time (sec): 3.14 - samples/sec: 5265.04 - lr: 0.000022 - momentum: 0.000000
2023-10-25 21:32:06,345 epoch 4 - iter 52/136 - loss 0.04366552 - time (sec): 4.04 - samples/sec: 5215.14 - lr: 0.000022 - momentum: 0.000000
2023-10-25 21:32:07,242 epoch 4 - iter 65/136 - loss 0.04276438 - time (sec): 4.93 - samples/sec: 5140.59 - lr: 0.000022 - momentum: 0.000000
2023-10-25 21:32:08,243 epoch 4 - iter 78/136 - loss 0.04392016 - time (sec): 5.93 - samples/sec: 5061.38 - lr: 0.000021 - momentum: 0.000000
2023-10-25 21:32:09,338 epoch 4 - iter 91/136 - loss 0.04355417 - time (sec): 7.03 - samples/sec: 5010.34 - lr: 0.000021 - momentum: 0.000000
2023-10-25 21:32:10,401 epoch 4 - iter 104/136 - loss 0.04553449 - time (sec): 8.09 - samples/sec: 5030.91 - lr: 0.000021 - momentum: 0.000000
2023-10-25 21:32:11,297 epoch 4 - iter 117/136 - loss 0.04613281 - time (sec): 8.99 - samples/sec: 5012.27 - lr: 0.000021 - momentum: 0.000000
2023-10-25 21:32:12,347 epoch 4 - iter 130/136 - loss 0.04531020 - time (sec): 10.04 - samples/sec: 4970.93 - lr: 0.000020 - momentum: 0.000000
2023-10-25 21:32:12,767 ----------------------------------------------------------------------------------------------------
2023-10-25 21:32:12,767 EPOCH 4 done: loss 0.0456 - lr: 0.000020
2023-10-25 21:32:14,146 DEV : loss 0.10330618172883987 - f1-score (micro avg) 0.8
2023-10-25 21:32:14,152 saving best model
2023-10-25 21:32:14,883 ----------------------------------------------------------------------------------------------------
2023-10-25 21:32:15,913 epoch 5 - iter 13/136 - loss 0.02499789 - time (sec): 1.03 - samples/sec: 4912.36 - lr: 0.000020 - momentum: 0.000000
2023-10-25 21:32:16,874 epoch 5 - iter 26/136 - loss 0.01992048 - time (sec): 1.99 - samples/sec: 4736.67 - lr: 0.000019 - momentum: 0.000000
2023-10-25 21:32:17,823 epoch 5 - iter 39/136 - loss 0.02764478 - time (sec): 2.94 - samples/sec: 4782.33 - lr: 0.000019 - momentum: 0.000000
2023-10-25 21:32:18,798 epoch 5 - iter 52/136 - loss 0.02728486 - time (sec): 3.91 - samples/sec: 4825.41 - lr: 0.000019 - momentum: 0.000000
2023-10-25 21:32:19,695 epoch 5 - iter 65/136 - loss 0.03034797 - time (sec): 4.81 - samples/sec: 4828.95 - lr: 0.000018 - momentum: 0.000000
2023-10-25 21:32:20,826 epoch 5 - iter 78/136 - loss 0.03063707 - time (sec): 5.94 - samples/sec: 4901.53 - lr: 0.000018 - momentum: 0.000000
2023-10-25 21:32:22,099 epoch 5 - iter 91/136 - loss 0.03029661 - time (sec): 7.21 - samples/sec: 4867.24 - lr: 0.000018 - momentum: 0.000000
2023-10-25 21:32:23,087 epoch 5 - iter 104/136 - loss 0.03057002 - time (sec): 8.20 - samples/sec: 4899.67 - lr: 0.000018 - momentum: 0.000000
2023-10-25 21:32:23,977 epoch 5 - iter 117/136 - loss 0.03238794 - time (sec): 9.09 - samples/sec: 4905.19 - lr: 0.000017 - momentum: 0.000000
2023-10-25 21:32:24,939 epoch 5 - iter 130/136 - loss 0.03113197 - time (sec): 10.05 - samples/sec: 4943.76 - lr: 0.000017 - momentum: 0.000000
2023-10-25 21:32:25,365 ----------------------------------------------------------------------------------------------------
2023-10-25 21:32:25,366 EPOCH 5 done: loss 0.0304 - lr: 0.000017
2023-10-25 21:32:27,025 DEV : loss 0.1195770800113678 - f1-score (micro avg) 0.8037
2023-10-25 21:32:27,032 saving best model
2023-10-25 21:32:27,749 ----------------------------------------------------------------------------------------------------
2023-10-25 21:32:28,803 epoch 6 - iter 13/136 - loss 0.01756530 - time (sec): 1.05 - samples/sec: 5366.79 - lr: 0.000016 - momentum: 0.000000
2023-10-25 21:32:29,850 epoch 6 - iter 26/136 - loss 0.02348632 - time (sec): 2.10 - samples/sec: 5058.44 - lr: 0.000016 - momentum: 0.000000
2023-10-25 21:32:30,796 epoch 6 - iter 39/136 - loss 0.02052357 - time (sec): 3.05 - samples/sec: 5110.69 - lr: 0.000016 - momentum: 0.000000
2023-10-25 21:32:31,849 epoch 6 - iter 52/136 - loss 0.02138897 - time (sec): 4.10 - samples/sec: 4940.68 - lr: 0.000015 - momentum: 0.000000
2023-10-25 21:32:32,811 epoch 6 - iter 65/136 - loss 0.02488717 - time (sec): 5.06 - samples/sec: 4853.29 - lr: 0.000015 - momentum: 0.000000
2023-10-25 21:32:33,815 epoch 6 - iter 78/136 - loss 0.02260196 - time (sec): 6.06 - samples/sec: 4945.45 - lr: 0.000015 - momentum: 0.000000
2023-10-25 21:32:34,799 epoch 6 - iter 91/136 - loss 0.02378811 - time (sec): 7.05 - samples/sec: 4971.81 - lr: 0.000015 - momentum: 0.000000
2023-10-25 21:32:35,887 epoch 6 - iter 104/136 - loss 0.02407465 - time (sec): 8.14 - samples/sec: 5025.23 - lr: 0.000014 - momentum: 0.000000
2023-10-25 21:32:36,916 epoch 6 - iter 117/136 - loss 0.02377918 - time (sec): 9.17 - samples/sec: 4953.62 - lr: 0.000014 - momentum: 0.000000
2023-10-25 21:32:37,826 epoch 6 - iter 130/136 - loss 0.02268322 - time (sec): 10.08 - samples/sec: 5001.50 - lr: 0.000014 - momentum: 0.000000
2023-10-25 21:32:38,197 ----------------------------------------------------------------------------------------------------
2023-10-25 21:32:38,197 EPOCH 6 done: loss 0.0222 - lr: 0.000014
2023-10-25 21:32:39,383 DEV : loss 0.1351010650396347 - f1-score (micro avg) 0.7927
2023-10-25 21:32:39,390 ----------------------------------------------------------------------------------------------------
2023-10-25 21:32:40,439 epoch 7 - iter 13/136 - loss 0.01667780 - time (sec): 1.05 - samples/sec: 4199.34 - lr: 0.000013 - momentum: 0.000000
2023-10-25 21:32:41,346 epoch 7 - iter 26/136 - loss 0.01388388 - time (sec): 1.95 - samples/sec: 4535.34 - lr: 0.000013 - momentum: 0.000000
2023-10-25 21:32:42,346 epoch 7 - iter 39/136 - loss 0.01450612 - time (sec): 2.95 - samples/sec: 4460.74 - lr: 0.000012 - momentum: 0.000000
2023-10-25 21:32:43,416 epoch 7 - iter 52/136 - loss 0.01658526 - time (sec): 4.03 - samples/sec: 4680.42 - lr: 0.000012 - momentum: 0.000000
2023-10-25 21:32:44,330 epoch 7 - iter 65/136 - loss 0.01539127 - time (sec): 4.94 - samples/sec: 4724.57 - lr: 0.000012 - momentum: 0.000000
2023-10-25 21:32:45,264 epoch 7 - iter 78/136 - loss 0.01885348 - time (sec): 5.87 - samples/sec: 4879.61 - lr: 0.000012 - momentum: 0.000000
2023-10-25 21:32:46,287 epoch 7 - iter 91/136 - loss 0.01876729 - time (sec): 6.90 - samples/sec: 4938.69 - lr: 0.000011 - momentum: 0.000000
2023-10-25 21:32:47,232 epoch 7 - iter 104/136 - loss 0.01819473 - time (sec): 7.84 - samples/sec: 4961.39 - lr: 0.000011 - momentum: 0.000000
2023-10-25 21:32:48,233 epoch 7 - iter 117/136 - loss 0.01693316 - time (sec): 8.84 - samples/sec: 5000.54 - lr: 0.000011 - momentum: 0.000000
2023-10-25 21:32:49,163 epoch 7 - iter 130/136 - loss 0.01647128 - time (sec): 9.77 - samples/sec: 5032.42 - lr: 0.000010 - momentum: 0.000000
2023-10-25 21:32:49,680 ----------------------------------------------------------------------------------------------------
2023-10-25 21:32:49,681 EPOCH 7 done: loss 0.0166 - lr: 0.000010
2023-10-25 21:32:50,858 DEV : loss 0.1340150088071823 - f1-score (micro avg) 0.8
2023-10-25 21:32:50,865 ----------------------------------------------------------------------------------------------------
2023-10-25 21:32:52,199 epoch 8 - iter 13/136 - loss 0.00560534 - time (sec): 1.33 - samples/sec: 3356.32 - lr: 0.000010 - momentum: 0.000000
2023-10-25 21:32:53,265 epoch 8 - iter 26/136 - loss 0.00976307 - time (sec): 2.40 - samples/sec: 4207.42 - lr: 0.000009 - momentum: 0.000000
2023-10-25 21:32:54,304 epoch 8 - iter 39/136 - loss 0.01070634 - time (sec): 3.44 - samples/sec: 4583.64 - lr: 0.000009 - momentum: 0.000000
2023-10-25 21:32:55,296 epoch 8 - iter 52/136 - loss 0.01208915 - time (sec): 4.43 - samples/sec: 4643.84 - lr: 0.000009 - momentum: 0.000000
2023-10-25 21:32:56,323 epoch 8 - iter 65/136 - loss 0.01109595 - time (sec): 5.46 - samples/sec: 4568.73 - lr: 0.000009 - momentum: 0.000000
2023-10-25 21:32:57,330 epoch 8 - iter 78/136 - loss 0.01173631 - time (sec): 6.46 - samples/sec: 4733.52 - lr: 0.000008 - momentum: 0.000000
2023-10-25 21:32:58,296 epoch 8 - iter 91/136 - loss 0.01105965 - time (sec): 7.43 - samples/sec: 4775.36 - lr: 0.000008 - momentum: 0.000000
2023-10-25 21:32:59,316 epoch 8 - iter 104/136 - loss 0.01148285 - time (sec): 8.45 - samples/sec: 4777.31 - lr: 0.000008 - momentum: 0.000000
2023-10-25 21:33:00,234 epoch 8 - iter 117/136 - loss 0.01174659 - time (sec): 9.37 - samples/sec: 4800.88 - lr: 0.000007 - momentum: 0.000000
2023-10-25 21:33:01,265 epoch 8 - iter 130/136 - loss 0.01071709 - time (sec): 10.40 - samples/sec: 4801.33 - lr: 0.000007 - momentum: 0.000000
2023-10-25 21:33:01,692 ----------------------------------------------------------------------------------------------------
2023-10-25 21:33:01,693 EPOCH 8 done: loss 0.0121 - lr: 0.000007
2023-10-25 21:33:02,845 DEV : loss 0.16605901718139648 - f1-score (micro avg) 0.8051
2023-10-25 21:33:02,852 saving best model
2023-10-25 21:33:03,579 ----------------------------------------------------------------------------------------------------
2023-10-25 21:33:04,564 epoch 9 - iter 13/136 - loss 0.00519371 - time (sec): 0.98 - samples/sec: 4935.82 - lr: 0.000006 - momentum: 0.000000
2023-10-25 21:33:05,403 epoch 9 - iter 26/136 - loss 0.00741030 - time (sec): 1.81 - samples/sec: 4742.54 - lr: 0.000006 - momentum: 0.000000
2023-10-25 21:33:06,394 epoch 9 - iter 39/136 - loss 0.00860202 - time (sec): 2.81 - samples/sec: 4879.48 - lr: 0.000006 - momentum: 0.000000
2023-10-25 21:33:07,354 epoch 9 - iter 52/136 - loss 0.00994720 - time (sec): 3.76 - samples/sec: 4840.00 - lr: 0.000006 - momentum: 0.000000
2023-10-25 21:33:08,435 epoch 9 - iter 65/136 - loss 0.00914803 - time (sec): 4.85 - samples/sec: 4892.40 - lr: 0.000005 - momentum: 0.000000
2023-10-25 21:33:09,545 epoch 9 - iter 78/136 - loss 0.00869169 - time (sec): 5.96 - samples/sec: 4949.71 - lr: 0.000005 - momentum: 0.000000
2023-10-25 21:33:10,573 epoch 9 - iter 91/136 - loss 0.00816170 - time (sec): 6.98 - samples/sec: 5004.09 - lr: 0.000005 - momentum: 0.000000
2023-10-25 21:33:11,706 epoch 9 - iter 104/136 - loss 0.00799395 - time (sec): 8.12 - samples/sec: 5036.45 - lr: 0.000004 - momentum: 0.000000
2023-10-25 21:33:12,646 epoch 9 - iter 117/136 - loss 0.00901235 - time (sec): 9.06 - samples/sec: 5069.49 - lr: 0.000004 - momentum: 0.000000
2023-10-25 21:33:13,534 epoch 9 - iter 130/136 - loss 0.00901908 - time (sec): 9.95 - samples/sec: 5056.08 - lr: 0.000004 - momentum: 0.000000
2023-10-25 21:33:13,901 ----------------------------------------------------------------------------------------------------
2023-10-25 21:33:13,901 EPOCH 9 done: loss 0.0090 - lr: 0.000004
2023-10-25 21:33:15,092 DEV : loss 0.17188507318496704 - f1-score (micro avg) 0.8124
2023-10-25 21:33:15,098 saving best model
2023-10-25 21:33:15,802 ----------------------------------------------------------------------------------------------------
2023-10-25 21:33:16,783 epoch 10 - iter 13/136 - loss 0.00872606 - time (sec): 0.97 - samples/sec: 4629.98 - lr: 0.000003 - momentum: 0.000000
2023-10-25 21:33:17,714 epoch 10 - iter 26/136 - loss 0.01279988 - time (sec): 1.91 - samples/sec: 4849.49 - lr: 0.000003 - momentum: 0.000000
2023-10-25 21:33:19,030 epoch 10 - iter 39/136 - loss 0.00977020 - time (sec): 3.22 - samples/sec: 4685.56 - lr: 0.000003 - momentum: 0.000000
2023-10-25 21:33:19,887 epoch 10 - iter 52/136 - loss 0.00929420 - time (sec): 4.08 - samples/sec: 4725.22 - lr: 0.000002 - momentum: 0.000000
2023-10-25 21:33:20,819 epoch 10 - iter 65/136 - loss 0.00906255 - time (sec): 5.01 - samples/sec: 4785.44 - lr: 0.000002 - momentum: 0.000000
2023-10-25 21:33:21,900 epoch 10 - iter 78/136 - loss 0.00778176 - time (sec): 6.09 - samples/sec: 4778.21 - lr: 0.000002 - momentum: 0.000000
2023-10-25 21:33:22,977 epoch 10 - iter 91/136 - loss 0.00714978 - time (sec): 7.17 - samples/sec: 4762.80 - lr: 0.000001 - momentum: 0.000000
2023-10-25 21:33:23,940 epoch 10 - iter 104/136 - loss 0.00657170 - time (sec): 8.13 - samples/sec: 4838.79 - lr: 0.000001 - momentum: 0.000000
2023-10-25 21:33:24,878 epoch 10 - iter 117/136 - loss 0.00701785 - time (sec): 9.07 - samples/sec: 4910.58 - lr: 0.000001 - momentum: 0.000000
2023-10-25 21:33:25,928 epoch 10 - iter 130/136 - loss 0.00745080 - time (sec): 10.12 - samples/sec: 4919.23 - lr: 0.000000 - momentum: 0.000000
2023-10-25 21:33:26,456 ----------------------------------------------------------------------------------------------------
2023-10-25 21:33:26,457 EPOCH 10 done: loss 0.0077 - lr: 0.000000
2023-10-25 21:33:27,624 DEV : loss 0.16700904071331024 - f1-score (micro avg) 0.8145
2023-10-25 21:33:27,630 saving best model
2023-10-25 21:33:28,840 ----------------------------------------------------------------------------------------------------
2023-10-25 21:33:28,841 Loading model from best epoch ...
2023-10-25 21:33:30,739 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 21:33:32,729
Results:
- F-score (micro) 0.7818
- F-score (macro) 0.7411
- Accuracy 0.657
By class:
precision recall f1-score support
LOC 0.7959 0.8622 0.8277 312
PER 0.6894 0.8750 0.7712 208
ORG 0.5400 0.4909 0.5143 55
HumanProd 0.8000 0.9091 0.8511 22
micro avg 0.7356 0.8342 0.7818 597
macro avg 0.7063 0.7843 0.7411 597
weighted avg 0.7353 0.8342 0.7800 597
2023-10-25 21:33:32,729 ----------------------------------------------------------------------------------------------------
|