File size: 24,188 Bytes
39a1928 |
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 |
2023-10-20 09:56:17,330 ----------------------------------------------------------------------------------------------------
2023-10-20 09:56:17,330 Model: "SequenceTagger(
(embeddings): TransformerWordEmbeddings(
(model): BertModel(
(embeddings): BertEmbeddings(
(word_embeddings): Embedding(32001, 128)
(position_embeddings): Embedding(512, 128)
(token_type_embeddings): Embedding(2, 128)
(LayerNorm): LayerNorm((128,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(encoder): BertEncoder(
(layer): ModuleList(
(0-1): 2 x BertLayer(
(attention): BertAttention(
(self): BertSelfAttention(
(query): Linear(in_features=128, out_features=128, bias=True)
(key): Linear(in_features=128, out_features=128, bias=True)
(value): Linear(in_features=128, out_features=128, bias=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(output): BertSelfOutput(
(dense): Linear(in_features=128, out_features=128, bias=True)
(LayerNorm): LayerNorm((128,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
(intermediate): BertIntermediate(
(dense): Linear(in_features=128, out_features=512, bias=True)
(intermediate_act_fn): GELUActivation()
)
(output): BertOutput(
(dense): Linear(in_features=512, out_features=128, bias=True)
(LayerNorm): LayerNorm((128,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
)
(pooler): BertPooler(
(dense): Linear(in_features=128, out_features=128, bias=True)
(activation): Tanh()
)
)
)
(locked_dropout): LockedDropout(p=0.5)
(linear): Linear(in_features=128, out_features=13, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-20 09:56:17,330 ----------------------------------------------------------------------------------------------------
2023-10-20 09:56:17,330 MultiCorpus: 6183 train + 680 dev + 2113 test sentences
- NER_HIPE_2022 Corpus: 6183 train + 680 dev + 2113 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/topres19th/en/with_doc_seperator
2023-10-20 09:56:17,330 ----------------------------------------------------------------------------------------------------
2023-10-20 09:56:17,330 Train: 6183 sentences
2023-10-20 09:56:17,330 (train_with_dev=False, train_with_test=False)
2023-10-20 09:56:17,330 ----------------------------------------------------------------------------------------------------
2023-10-20 09:56:17,330 Training Params:
2023-10-20 09:56:17,330 - learning_rate: "5e-05"
2023-10-20 09:56:17,330 - mini_batch_size: "4"
2023-10-20 09:56:17,330 - max_epochs: "10"
2023-10-20 09:56:17,330 - shuffle: "True"
2023-10-20 09:56:17,330 ----------------------------------------------------------------------------------------------------
2023-10-20 09:56:17,330 Plugins:
2023-10-20 09:56:17,331 - TensorboardLogger
2023-10-20 09:56:17,331 - LinearScheduler | warmup_fraction: '0.1'
2023-10-20 09:56:17,331 ----------------------------------------------------------------------------------------------------
2023-10-20 09:56:17,331 Final evaluation on model from best epoch (best-model.pt)
2023-10-20 09:56:17,331 - metric: "('micro avg', 'f1-score')"
2023-10-20 09:56:17,331 ----------------------------------------------------------------------------------------------------
2023-10-20 09:56:17,331 Computation:
2023-10-20 09:56:17,331 - compute on device: cuda:0
2023-10-20 09:56:17,331 - embedding storage: none
2023-10-20 09:56:17,331 ----------------------------------------------------------------------------------------------------
2023-10-20 09:56:17,331 Model training base path: "hmbench-topres19th/en-dbmdz/bert-tiny-historic-multilingual-cased-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-4"
2023-10-20 09:56:17,331 ----------------------------------------------------------------------------------------------------
2023-10-20 09:56:17,331 ----------------------------------------------------------------------------------------------------
2023-10-20 09:56:17,331 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-20 09:56:19,693 epoch 1 - iter 154/1546 - loss 3.62102707 - time (sec): 2.36 - samples/sec: 5236.15 - lr: 0.000005 - momentum: 0.000000
2023-10-20 09:56:22,114 epoch 1 - iter 308/1546 - loss 3.13477517 - time (sec): 4.78 - samples/sec: 5182.14 - lr: 0.000010 - momentum: 0.000000
2023-10-20 09:56:24,433 epoch 1 - iter 462/1546 - loss 2.46338633 - time (sec): 7.10 - samples/sec: 5146.78 - lr: 0.000015 - momentum: 0.000000
2023-10-20 09:56:26,539 epoch 1 - iter 616/1546 - loss 1.93374896 - time (sec): 9.21 - samples/sec: 5337.75 - lr: 0.000020 - momentum: 0.000000
2023-10-20 09:56:28,673 epoch 1 - iter 770/1546 - loss 1.61995330 - time (sec): 11.34 - samples/sec: 5337.89 - lr: 0.000025 - momentum: 0.000000
2023-10-20 09:56:30,909 epoch 1 - iter 924/1546 - loss 1.39359053 - time (sec): 13.58 - samples/sec: 5362.23 - lr: 0.000030 - momentum: 0.000000
2023-10-20 09:56:33,199 epoch 1 - iter 1078/1546 - loss 1.23035252 - time (sec): 15.87 - samples/sec: 5350.81 - lr: 0.000035 - momentum: 0.000000
2023-10-20 09:56:35,625 epoch 1 - iter 1232/1546 - loss 1.09712317 - time (sec): 18.29 - samples/sec: 5368.14 - lr: 0.000040 - momentum: 0.000000
2023-10-20 09:56:37,924 epoch 1 - iter 1386/1546 - loss 0.99421232 - time (sec): 20.59 - samples/sec: 5403.80 - lr: 0.000045 - momentum: 0.000000
2023-10-20 09:56:40,172 epoch 1 - iter 1540/1546 - loss 0.91571471 - time (sec): 22.84 - samples/sec: 5422.50 - lr: 0.000050 - momentum: 0.000000
2023-10-20 09:56:40,282 ----------------------------------------------------------------------------------------------------
2023-10-20 09:56:40,282 EPOCH 1 done: loss 0.9134 - lr: 0.000050
2023-10-20 09:56:41,270 DEV : loss 0.11716283857822418 - f1-score (micro avg) 0.0247
2023-10-20 09:56:41,282 saving best model
2023-10-20 09:56:41,311 ----------------------------------------------------------------------------------------------------
2023-10-20 09:56:43,719 epoch 2 - iter 154/1546 - loss 0.19306759 - time (sec): 2.41 - samples/sec: 5487.96 - lr: 0.000049 - momentum: 0.000000
2023-10-20 09:56:46,138 epoch 2 - iter 308/1546 - loss 0.19527201 - time (sec): 4.83 - samples/sec: 5486.99 - lr: 0.000049 - momentum: 0.000000
2023-10-20 09:56:48,446 epoch 2 - iter 462/1546 - loss 0.19787880 - time (sec): 7.13 - samples/sec: 5214.32 - lr: 0.000048 - momentum: 0.000000
2023-10-20 09:56:50,833 epoch 2 - iter 616/1546 - loss 0.19678261 - time (sec): 9.52 - samples/sec: 5154.17 - lr: 0.000048 - momentum: 0.000000
2023-10-20 09:56:53,210 epoch 2 - iter 770/1546 - loss 0.19605129 - time (sec): 11.90 - samples/sec: 5121.25 - lr: 0.000047 - momentum: 0.000000
2023-10-20 09:56:55,564 epoch 2 - iter 924/1546 - loss 0.19154674 - time (sec): 14.25 - samples/sec: 5156.29 - lr: 0.000047 - momentum: 0.000000
2023-10-20 09:56:57,968 epoch 2 - iter 1078/1546 - loss 0.18535945 - time (sec): 16.66 - samples/sec: 5202.28 - lr: 0.000046 - momentum: 0.000000
2023-10-20 09:57:00,295 epoch 2 - iter 1232/1546 - loss 0.18750537 - time (sec): 18.98 - samples/sec: 5175.91 - lr: 0.000046 - momentum: 0.000000
2023-10-20 09:57:02,650 epoch 2 - iter 1386/1546 - loss 0.18188318 - time (sec): 21.34 - samples/sec: 5162.78 - lr: 0.000045 - momentum: 0.000000
2023-10-20 09:57:05,066 epoch 2 - iter 1540/1546 - loss 0.18010251 - time (sec): 23.75 - samples/sec: 5209.21 - lr: 0.000044 - momentum: 0.000000
2023-10-20 09:57:05,155 ----------------------------------------------------------------------------------------------------
2023-10-20 09:57:05,155 EPOCH 2 done: loss 0.1797 - lr: 0.000044
2023-10-20 09:57:06,238 DEV : loss 0.09568006545305252 - f1-score (micro avg) 0.4505
2023-10-20 09:57:06,250 saving best model
2023-10-20 09:57:06,289 ----------------------------------------------------------------------------------------------------
2023-10-20 09:57:08,762 epoch 3 - iter 154/1546 - loss 0.15246759 - time (sec): 2.47 - samples/sec: 5045.79 - lr: 0.000044 - momentum: 0.000000
2023-10-20 09:57:11,131 epoch 3 - iter 308/1546 - loss 0.15171707 - time (sec): 4.84 - samples/sec: 4969.29 - lr: 0.000043 - momentum: 0.000000
2023-10-20 09:57:13,473 epoch 3 - iter 462/1546 - loss 0.14819470 - time (sec): 7.18 - samples/sec: 5146.44 - lr: 0.000043 - momentum: 0.000000
2023-10-20 09:57:15,811 epoch 3 - iter 616/1546 - loss 0.14757401 - time (sec): 9.52 - samples/sec: 5193.61 - lr: 0.000042 - momentum: 0.000000
2023-10-20 09:57:18,159 epoch 3 - iter 770/1546 - loss 0.14865086 - time (sec): 11.87 - samples/sec: 5251.53 - lr: 0.000042 - momentum: 0.000000
2023-10-20 09:57:20,503 epoch 3 - iter 924/1546 - loss 0.15129778 - time (sec): 14.21 - samples/sec: 5149.30 - lr: 0.000041 - momentum: 0.000000
2023-10-20 09:57:22,888 epoch 3 - iter 1078/1546 - loss 0.14983046 - time (sec): 16.60 - samples/sec: 5184.46 - lr: 0.000041 - momentum: 0.000000
2023-10-20 09:57:25,248 epoch 3 - iter 1232/1546 - loss 0.14979943 - time (sec): 18.96 - samples/sec: 5220.74 - lr: 0.000040 - momentum: 0.000000
2023-10-20 09:57:27,623 epoch 3 - iter 1386/1546 - loss 0.15151253 - time (sec): 21.33 - samples/sec: 5212.66 - lr: 0.000039 - momentum: 0.000000
2023-10-20 09:57:30,024 epoch 3 - iter 1540/1546 - loss 0.14973869 - time (sec): 23.73 - samples/sec: 5220.95 - lr: 0.000039 - momentum: 0.000000
2023-10-20 09:57:30,119 ----------------------------------------------------------------------------------------------------
2023-10-20 09:57:30,119 EPOCH 3 done: loss 0.1500 - lr: 0.000039
2023-10-20 09:57:31,211 DEV : loss 0.09354293346405029 - f1-score (micro avg) 0.5438
2023-10-20 09:57:31,223 saving best model
2023-10-20 09:57:31,257 ----------------------------------------------------------------------------------------------------
2023-10-20 09:57:33,586 epoch 4 - iter 154/1546 - loss 0.15714911 - time (sec): 2.33 - samples/sec: 5661.53 - lr: 0.000038 - momentum: 0.000000
2023-10-20 09:57:35,970 epoch 4 - iter 308/1546 - loss 0.14974834 - time (sec): 4.71 - samples/sec: 5257.99 - lr: 0.000038 - momentum: 0.000000
2023-10-20 09:57:38,295 epoch 4 - iter 462/1546 - loss 0.14721188 - time (sec): 7.04 - samples/sec: 5145.14 - lr: 0.000037 - momentum: 0.000000
2023-10-20 09:57:40,675 epoch 4 - iter 616/1546 - loss 0.14478087 - time (sec): 9.42 - samples/sec: 5248.76 - lr: 0.000037 - momentum: 0.000000
2023-10-20 09:57:43,001 epoch 4 - iter 770/1546 - loss 0.14146413 - time (sec): 11.74 - samples/sec: 5205.67 - lr: 0.000036 - momentum: 0.000000
2023-10-20 09:57:45,386 epoch 4 - iter 924/1546 - loss 0.13681608 - time (sec): 14.13 - samples/sec: 5194.75 - lr: 0.000036 - momentum: 0.000000
2023-10-20 09:57:47,777 epoch 4 - iter 1078/1546 - loss 0.13568701 - time (sec): 16.52 - samples/sec: 5199.00 - lr: 0.000035 - momentum: 0.000000
2023-10-20 09:57:50,225 epoch 4 - iter 1232/1546 - loss 0.13598660 - time (sec): 18.97 - samples/sec: 5200.73 - lr: 0.000034 - momentum: 0.000000
2023-10-20 09:57:52,616 epoch 4 - iter 1386/1546 - loss 0.13476872 - time (sec): 21.36 - samples/sec: 5213.71 - lr: 0.000034 - momentum: 0.000000
2023-10-20 09:57:54,877 epoch 4 - iter 1540/1546 - loss 0.13332730 - time (sec): 23.62 - samples/sec: 5236.93 - lr: 0.000033 - momentum: 0.000000
2023-10-20 09:57:54,964 ----------------------------------------------------------------------------------------------------
2023-10-20 09:57:54,965 EPOCH 4 done: loss 0.1335 - lr: 0.000033
2023-10-20 09:57:56,336 DEV : loss 0.08910585939884186 - f1-score (micro avg) 0.5781
2023-10-20 09:57:56,348 saving best model
2023-10-20 09:57:56,389 ----------------------------------------------------------------------------------------------------
2023-10-20 09:57:58,684 epoch 5 - iter 154/1546 - loss 0.11074956 - time (sec): 2.29 - samples/sec: 5207.64 - lr: 0.000033 - momentum: 0.000000
2023-10-20 09:58:01,079 epoch 5 - iter 308/1546 - loss 0.12438368 - time (sec): 4.69 - samples/sec: 5411.83 - lr: 0.000032 - momentum: 0.000000
2023-10-20 09:58:03,466 epoch 5 - iter 462/1546 - loss 0.13150006 - time (sec): 7.08 - samples/sec: 5360.25 - lr: 0.000032 - momentum: 0.000000
2023-10-20 09:58:05,767 epoch 5 - iter 616/1546 - loss 0.12690522 - time (sec): 9.38 - samples/sec: 5263.95 - lr: 0.000031 - momentum: 0.000000
2023-10-20 09:58:08,159 epoch 5 - iter 770/1546 - loss 0.12550895 - time (sec): 11.77 - samples/sec: 5297.28 - lr: 0.000031 - momentum: 0.000000
2023-10-20 09:58:10,480 epoch 5 - iter 924/1546 - loss 0.12198991 - time (sec): 14.09 - samples/sec: 5273.19 - lr: 0.000030 - momentum: 0.000000
2023-10-20 09:58:12,851 epoch 5 - iter 1078/1546 - loss 0.12094948 - time (sec): 16.46 - samples/sec: 5285.15 - lr: 0.000029 - momentum: 0.000000
2023-10-20 09:58:15,325 epoch 5 - iter 1232/1546 - loss 0.12368525 - time (sec): 18.94 - samples/sec: 5235.62 - lr: 0.000029 - momentum: 0.000000
2023-10-20 09:58:17,724 epoch 5 - iter 1386/1546 - loss 0.12543156 - time (sec): 21.33 - samples/sec: 5235.20 - lr: 0.000028 - momentum: 0.000000
2023-10-20 09:58:20,123 epoch 5 - iter 1540/1546 - loss 0.12463983 - time (sec): 23.73 - samples/sec: 5213.99 - lr: 0.000028 - momentum: 0.000000
2023-10-20 09:58:20,217 ----------------------------------------------------------------------------------------------------
2023-10-20 09:58:20,217 EPOCH 5 done: loss 0.1245 - lr: 0.000028
2023-10-20 09:58:21,313 DEV : loss 0.09435312449932098 - f1-score (micro avg) 0.5978
2023-10-20 09:58:21,326 saving best model
2023-10-20 09:58:21,360 ----------------------------------------------------------------------------------------------------
2023-10-20 09:58:23,747 epoch 6 - iter 154/1546 - loss 0.11053617 - time (sec): 2.39 - samples/sec: 4737.34 - lr: 0.000027 - momentum: 0.000000
2023-10-20 09:58:26,136 epoch 6 - iter 308/1546 - loss 0.11721437 - time (sec): 4.78 - samples/sec: 4972.77 - lr: 0.000027 - momentum: 0.000000
2023-10-20 09:58:28,544 epoch 6 - iter 462/1546 - loss 0.11731031 - time (sec): 7.18 - samples/sec: 5050.51 - lr: 0.000026 - momentum: 0.000000
2023-10-20 09:58:30,911 epoch 6 - iter 616/1546 - loss 0.11093231 - time (sec): 9.55 - samples/sec: 5101.05 - lr: 0.000026 - momentum: 0.000000
2023-10-20 09:58:33,258 epoch 6 - iter 770/1546 - loss 0.11841360 - time (sec): 11.90 - samples/sec: 5111.39 - lr: 0.000025 - momentum: 0.000000
2023-10-20 09:58:35,581 epoch 6 - iter 924/1546 - loss 0.11998999 - time (sec): 14.22 - samples/sec: 5129.85 - lr: 0.000024 - momentum: 0.000000
2023-10-20 09:58:37,939 epoch 6 - iter 1078/1546 - loss 0.11704457 - time (sec): 16.58 - samples/sec: 5155.77 - lr: 0.000024 - momentum: 0.000000
2023-10-20 09:58:40,326 epoch 6 - iter 1232/1546 - loss 0.11636736 - time (sec): 18.97 - samples/sec: 5193.59 - lr: 0.000023 - momentum: 0.000000
2023-10-20 09:58:42,726 epoch 6 - iter 1386/1546 - loss 0.11459251 - time (sec): 21.37 - samples/sec: 5200.43 - lr: 0.000023 - momentum: 0.000000
2023-10-20 09:58:45,072 epoch 6 - iter 1540/1546 - loss 0.11638062 - time (sec): 23.71 - samples/sec: 5211.77 - lr: 0.000022 - momentum: 0.000000
2023-10-20 09:58:45,178 ----------------------------------------------------------------------------------------------------
2023-10-20 09:58:45,179 EPOCH 6 done: loss 0.1158 - lr: 0.000022
2023-10-20 09:58:46,284 DEV : loss 0.09260376542806625 - f1-score (micro avg) 0.6344
2023-10-20 09:58:46,296 saving best model
2023-10-20 09:58:46,333 ----------------------------------------------------------------------------------------------------
2023-10-20 09:58:48,697 epoch 7 - iter 154/1546 - loss 0.10982928 - time (sec): 2.36 - samples/sec: 5341.51 - lr: 0.000022 - momentum: 0.000000
2023-10-20 09:58:51,116 epoch 7 - iter 308/1546 - loss 0.10926222 - time (sec): 4.78 - samples/sec: 5195.27 - lr: 0.000021 - momentum: 0.000000
2023-10-20 09:58:53,492 epoch 7 - iter 462/1546 - loss 0.10878928 - time (sec): 7.16 - samples/sec: 5171.38 - lr: 0.000021 - momentum: 0.000000
2023-10-20 09:58:55,841 epoch 7 - iter 616/1546 - loss 0.11396563 - time (sec): 9.51 - samples/sec: 5221.24 - lr: 0.000020 - momentum: 0.000000
2023-10-20 09:58:58,270 epoch 7 - iter 770/1546 - loss 0.11096524 - time (sec): 11.94 - samples/sec: 5316.49 - lr: 0.000019 - momentum: 0.000000
2023-10-20 09:59:00,633 epoch 7 - iter 924/1546 - loss 0.10959747 - time (sec): 14.30 - samples/sec: 5299.06 - lr: 0.000019 - momentum: 0.000000
2023-10-20 09:59:03,007 epoch 7 - iter 1078/1546 - loss 0.11069285 - time (sec): 16.67 - samples/sec: 5260.74 - lr: 0.000018 - momentum: 0.000000
2023-10-20 09:59:05,348 epoch 7 - iter 1232/1546 - loss 0.11156839 - time (sec): 19.01 - samples/sec: 5246.88 - lr: 0.000018 - momentum: 0.000000
2023-10-20 09:59:07,702 epoch 7 - iter 1386/1546 - loss 0.11104971 - time (sec): 21.37 - samples/sec: 5237.77 - lr: 0.000017 - momentum: 0.000000
2023-10-20 09:59:10,067 epoch 7 - iter 1540/1546 - loss 0.10993191 - time (sec): 23.73 - samples/sec: 5217.63 - lr: 0.000017 - momentum: 0.000000
2023-10-20 09:59:10,167 ----------------------------------------------------------------------------------------------------
2023-10-20 09:59:10,167 EPOCH 7 done: loss 0.1098 - lr: 0.000017
2023-10-20 09:59:11,270 DEV : loss 0.09885262697935104 - f1-score (micro avg) 0.6147
2023-10-20 09:59:11,281 ----------------------------------------------------------------------------------------------------
2023-10-20 09:59:13,620 epoch 8 - iter 154/1546 - loss 0.10457091 - time (sec): 2.34 - samples/sec: 5371.89 - lr: 0.000016 - momentum: 0.000000
2023-10-20 09:59:16,035 epoch 8 - iter 308/1546 - loss 0.09573930 - time (sec): 4.75 - samples/sec: 5244.20 - lr: 0.000016 - momentum: 0.000000
2023-10-20 09:59:18,416 epoch 8 - iter 462/1546 - loss 0.09757855 - time (sec): 7.13 - samples/sec: 5222.20 - lr: 0.000015 - momentum: 0.000000
2023-10-20 09:59:20,846 epoch 8 - iter 616/1546 - loss 0.10232804 - time (sec): 9.56 - samples/sec: 5196.35 - lr: 0.000014 - momentum: 0.000000
2023-10-20 09:59:23,246 epoch 8 - iter 770/1546 - loss 0.10450911 - time (sec): 11.96 - samples/sec: 5213.17 - lr: 0.000014 - momentum: 0.000000
2023-10-20 09:59:25,609 epoch 8 - iter 924/1546 - loss 0.10466820 - time (sec): 14.33 - samples/sec: 5201.41 - lr: 0.000013 - momentum: 0.000000
2023-10-20 09:59:27,944 epoch 8 - iter 1078/1546 - loss 0.10690035 - time (sec): 16.66 - samples/sec: 5204.45 - lr: 0.000013 - momentum: 0.000000
2023-10-20 09:59:30,324 epoch 8 - iter 1232/1546 - loss 0.10601181 - time (sec): 19.04 - samples/sec: 5206.44 - lr: 0.000012 - momentum: 0.000000
2023-10-20 09:59:32,724 epoch 8 - iter 1386/1546 - loss 0.10561856 - time (sec): 21.44 - samples/sec: 5197.08 - lr: 0.000012 - momentum: 0.000000
2023-10-20 09:59:35,091 epoch 8 - iter 1540/1546 - loss 0.10482255 - time (sec): 23.81 - samples/sec: 5198.72 - lr: 0.000011 - momentum: 0.000000
2023-10-20 09:59:35,188 ----------------------------------------------------------------------------------------------------
2023-10-20 09:59:35,188 EPOCH 8 done: loss 0.1048 - lr: 0.000011
2023-10-20 09:59:36,269 DEV : loss 0.1057317703962326 - f1-score (micro avg) 0.6261
2023-10-20 09:59:36,281 ----------------------------------------------------------------------------------------------------
2023-10-20 09:59:38,590 epoch 9 - iter 154/1546 - loss 0.11356698 - time (sec): 2.31 - samples/sec: 5151.09 - lr: 0.000011 - momentum: 0.000000
2023-10-20 09:59:40,915 epoch 9 - iter 308/1546 - loss 0.10406692 - time (sec): 4.63 - samples/sec: 5176.55 - lr: 0.000010 - momentum: 0.000000
2023-10-20 09:59:43,279 epoch 9 - iter 462/1546 - loss 0.10401387 - time (sec): 7.00 - samples/sec: 5196.44 - lr: 0.000009 - momentum: 0.000000
2023-10-20 09:59:45,663 epoch 9 - iter 616/1546 - loss 0.09614905 - time (sec): 9.38 - samples/sec: 5264.89 - lr: 0.000009 - momentum: 0.000000
2023-10-20 09:59:48,049 epoch 9 - iter 770/1546 - loss 0.09323588 - time (sec): 11.77 - samples/sec: 5219.67 - lr: 0.000008 - momentum: 0.000000
2023-10-20 09:59:50,422 epoch 9 - iter 924/1546 - loss 0.09618676 - time (sec): 14.14 - samples/sec: 5189.75 - lr: 0.000008 - momentum: 0.000000
2023-10-20 09:59:52,822 epoch 9 - iter 1078/1546 - loss 0.09686063 - time (sec): 16.54 - samples/sec: 5188.16 - lr: 0.000007 - momentum: 0.000000
2023-10-20 09:59:55,194 epoch 9 - iter 1232/1546 - loss 0.09966681 - time (sec): 18.91 - samples/sec: 5213.59 - lr: 0.000007 - momentum: 0.000000
2023-10-20 09:59:57,578 epoch 9 - iter 1386/1546 - loss 0.09979067 - time (sec): 21.30 - samples/sec: 5253.50 - lr: 0.000006 - momentum: 0.000000
2023-10-20 09:59:59,886 epoch 9 - iter 1540/1546 - loss 0.10020466 - time (sec): 23.60 - samples/sec: 5245.93 - lr: 0.000006 - momentum: 0.000000
2023-10-20 09:59:59,978 ----------------------------------------------------------------------------------------------------
2023-10-20 09:59:59,978 EPOCH 9 done: loss 0.1004 - lr: 0.000006
2023-10-20 10:00:01,073 DEV : loss 0.10489093512296677 - f1-score (micro avg) 0.6565
2023-10-20 10:00:01,085 saving best model
2023-10-20 10:00:01,123 ----------------------------------------------------------------------------------------------------
2023-10-20 10:00:03,228 epoch 10 - iter 154/1546 - loss 0.11356945 - time (sec): 2.10 - samples/sec: 5568.14 - lr: 0.000005 - momentum: 0.000000
2023-10-20 10:00:05,574 epoch 10 - iter 308/1546 - loss 0.10220205 - time (sec): 4.45 - samples/sec: 5477.44 - lr: 0.000004 - momentum: 0.000000
2023-10-20 10:00:07,947 epoch 10 - iter 462/1546 - loss 0.09862619 - time (sec): 6.82 - samples/sec: 5456.90 - lr: 0.000004 - momentum: 0.000000
2023-10-20 10:00:10,344 epoch 10 - iter 616/1546 - loss 0.09815601 - time (sec): 9.22 - samples/sec: 5398.30 - lr: 0.000003 - momentum: 0.000000
2023-10-20 10:00:12,711 epoch 10 - iter 770/1546 - loss 0.09919748 - time (sec): 11.59 - samples/sec: 5350.37 - lr: 0.000003 - momentum: 0.000000
2023-10-20 10:00:15,063 epoch 10 - iter 924/1546 - loss 0.10078099 - time (sec): 13.94 - samples/sec: 5288.38 - lr: 0.000002 - momentum: 0.000000
2023-10-20 10:00:17,441 epoch 10 - iter 1078/1546 - loss 0.10013561 - time (sec): 16.32 - samples/sec: 5287.68 - lr: 0.000002 - momentum: 0.000000
2023-10-20 10:00:19,854 epoch 10 - iter 1232/1546 - loss 0.09648546 - time (sec): 18.73 - samples/sec: 5296.75 - lr: 0.000001 - momentum: 0.000000
2023-10-20 10:00:22,219 epoch 10 - iter 1386/1546 - loss 0.09661131 - time (sec): 21.10 - samples/sec: 5270.05 - lr: 0.000001 - momentum: 0.000000
2023-10-20 10:00:24,599 epoch 10 - iter 1540/1546 - loss 0.09632277 - time (sec): 23.48 - samples/sec: 5267.14 - lr: 0.000000 - momentum: 0.000000
2023-10-20 10:00:24,695 ----------------------------------------------------------------------------------------------------
2023-10-20 10:00:24,696 EPOCH 10 done: loss 0.0962 - lr: 0.000000
2023-10-20 10:00:25,790 DEV : loss 0.10527437180280685 - f1-score (micro avg) 0.6468
2023-10-20 10:00:25,836 ----------------------------------------------------------------------------------------------------
2023-10-20 10:00:25,836 Loading model from best epoch ...
2023-10-20 10:00:25,912 SequenceTagger predicts: Dictionary with 13 tags: O, S-LOC, B-LOC, E-LOC, I-LOC, S-BUILDING, B-BUILDING, E-BUILDING, I-BUILDING, S-STREET, B-STREET, E-STREET, I-STREET
2023-10-20 10:00:28,836
Results:
- F-score (micro) 0.5986
- F-score (macro) 0.342
- Accuracy 0.4367
By class:
precision recall f1-score support
LOC 0.6831 0.6723 0.6777 946
BUILDING 0.2317 0.1027 0.1423 185
STREET 0.5833 0.1250 0.2059 56
micro avg 0.6459 0.5577 0.5986 1187
macro avg 0.4994 0.3000 0.3420 1187
weighted avg 0.6081 0.5577 0.5720 1187
2023-10-20 10:00:28,836 ----------------------------------------------------------------------------------------------------
|