File size: 23,886 Bytes
1f0fede |
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 |
2023-10-18 22:11:46,686 ----------------------------------------------------------------------------------------------------
2023-10-18 22:11:46,687 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()
)"
2023-10-18 22:11:46,687 ----------------------------------------------------------------------------------------------------
2023-10-18 22:11:46,687 MultiCorpus: 5777 train + 722 dev + 723 test sentences
- NER_ICDAR_EUROPEANA Corpus: 5777 train + 722 dev + 723 test sentences - /root/.flair/datasets/ner_icdar_europeana/nl
2023-10-18 22:11:46,687 ----------------------------------------------------------------------------------------------------
2023-10-18 22:11:46,687 Train: 5777 sentences
2023-10-18 22:11:46,687 (train_with_dev=False, train_with_test=False)
2023-10-18 22:11:46,687 ----------------------------------------------------------------------------------------------------
2023-10-18 22:11:46,687 Training Params:
2023-10-18 22:11:46,687 - learning_rate: "5e-05"
2023-10-18 22:11:46,687 - mini_batch_size: "8"
2023-10-18 22:11:46,687 - max_epochs: "10"
2023-10-18 22:11:46,687 - shuffle: "True"
2023-10-18 22:11:46,687 ----------------------------------------------------------------------------------------------------
2023-10-18 22:11:46,687 Plugins:
2023-10-18 22:11:46,687 - TensorboardLogger
2023-10-18 22:11:46,687 - LinearScheduler | warmup_fraction: '0.1'
2023-10-18 22:11:46,687 ----------------------------------------------------------------------------------------------------
2023-10-18 22:11:46,687 Final evaluation on model from best epoch (best-model.pt)
2023-10-18 22:11:46,687 - metric: "('micro avg', 'f1-score')"
2023-10-18 22:11:46,687 ----------------------------------------------------------------------------------------------------
2023-10-18 22:11:46,688 Computation:
2023-10-18 22:11:46,688 - compute on device: cuda:0
2023-10-18 22:11:46,688 - embedding storage: none
2023-10-18 22:11:46,688 ----------------------------------------------------------------------------------------------------
2023-10-18 22:11:46,688 Model training base path: "hmbench-icdar/nl-dbmdz/bert-tiny-historic-multilingual-cased-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-1"
2023-10-18 22:11:46,688 ----------------------------------------------------------------------------------------------------
2023-10-18 22:11:46,688 ----------------------------------------------------------------------------------------------------
2023-10-18 22:11:46,688 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-18 22:11:48,521 epoch 1 - iter 72/723 - loss 3.15369193 - time (sec): 1.83 - samples/sec: 9418.43 - lr: 0.000005 - momentum: 0.000000
2023-10-18 22:11:50,353 epoch 1 - iter 144/723 - loss 2.82531096 - time (sec): 3.66 - samples/sec: 9750.79 - lr: 0.000010 - momentum: 0.000000
2023-10-18 22:11:52,174 epoch 1 - iter 216/723 - loss 2.35058830 - time (sec): 5.49 - samples/sec: 9736.98 - lr: 0.000015 - momentum: 0.000000
2023-10-18 22:11:53,972 epoch 1 - iter 288/723 - loss 1.91572462 - time (sec): 7.28 - samples/sec: 9734.87 - lr: 0.000020 - momentum: 0.000000
2023-10-18 22:11:55,779 epoch 1 - iter 360/723 - loss 1.59356684 - time (sec): 9.09 - samples/sec: 9795.92 - lr: 0.000025 - momentum: 0.000000
2023-10-18 22:11:57,603 epoch 1 - iter 432/723 - loss 1.37380455 - time (sec): 10.91 - samples/sec: 9818.38 - lr: 0.000030 - momentum: 0.000000
2023-10-18 22:11:59,430 epoch 1 - iter 504/723 - loss 1.22307055 - time (sec): 12.74 - samples/sec: 9797.58 - lr: 0.000035 - momentum: 0.000000
2023-10-18 22:12:01,273 epoch 1 - iter 576/723 - loss 1.11055697 - time (sec): 14.58 - samples/sec: 9758.58 - lr: 0.000040 - momentum: 0.000000
2023-10-18 22:12:03,063 epoch 1 - iter 648/723 - loss 1.02493184 - time (sec): 16.38 - samples/sec: 9685.83 - lr: 0.000045 - momentum: 0.000000
2023-10-18 22:12:04,793 epoch 1 - iter 720/723 - loss 0.95066192 - time (sec): 18.11 - samples/sec: 9693.53 - lr: 0.000050 - momentum: 0.000000
2023-10-18 22:12:04,877 ----------------------------------------------------------------------------------------------------
2023-10-18 22:12:04,877 EPOCH 1 done: loss 0.9486 - lr: 0.000050
2023-10-18 22:12:06,089 DEV : loss 0.31063002347946167 - f1-score (micro avg) 0.0
2023-10-18 22:12:06,102 ----------------------------------------------------------------------------------------------------
2023-10-18 22:12:07,989 epoch 2 - iter 72/723 - loss 0.23255061 - time (sec): 1.89 - samples/sec: 9861.83 - lr: 0.000049 - momentum: 0.000000
2023-10-18 22:12:09,742 epoch 2 - iter 144/723 - loss 0.23804088 - time (sec): 3.64 - samples/sec: 9834.13 - lr: 0.000049 - momentum: 0.000000
2023-10-18 22:12:11,573 epoch 2 - iter 216/723 - loss 0.24106924 - time (sec): 5.47 - samples/sec: 9842.86 - lr: 0.000048 - momentum: 0.000000
2023-10-18 22:12:13,330 epoch 2 - iter 288/723 - loss 0.23333713 - time (sec): 7.23 - samples/sec: 9913.68 - lr: 0.000048 - momentum: 0.000000
2023-10-18 22:12:15,083 epoch 2 - iter 360/723 - loss 0.22213309 - time (sec): 8.98 - samples/sec: 9890.07 - lr: 0.000047 - momentum: 0.000000
2023-10-18 22:12:16,869 epoch 2 - iter 432/723 - loss 0.21724811 - time (sec): 10.77 - samples/sec: 9985.45 - lr: 0.000047 - momentum: 0.000000
2023-10-18 22:12:18,591 epoch 2 - iter 504/723 - loss 0.21698959 - time (sec): 12.49 - samples/sec: 9901.44 - lr: 0.000046 - momentum: 0.000000
2023-10-18 22:12:20,283 epoch 2 - iter 576/723 - loss 0.21535393 - time (sec): 14.18 - samples/sec: 9886.96 - lr: 0.000046 - momentum: 0.000000
2023-10-18 22:12:22,061 epoch 2 - iter 648/723 - loss 0.21374976 - time (sec): 15.96 - samples/sec: 9885.22 - lr: 0.000045 - momentum: 0.000000
2023-10-18 22:12:23,798 epoch 2 - iter 720/723 - loss 0.20898244 - time (sec): 17.70 - samples/sec: 9933.82 - lr: 0.000044 - momentum: 0.000000
2023-10-18 22:12:23,856 ----------------------------------------------------------------------------------------------------
2023-10-18 22:12:23,856 EPOCH 2 done: loss 0.2092 - lr: 0.000044
2023-10-18 22:12:25,964 DEV : loss 0.22788439691066742 - f1-score (micro avg) 0.3239
2023-10-18 22:12:25,979 saving best model
2023-10-18 22:12:26,010 ----------------------------------------------------------------------------------------------------
2023-10-18 22:12:27,797 epoch 3 - iter 72/723 - loss 0.19078244 - time (sec): 1.79 - samples/sec: 10043.51 - lr: 0.000044 - momentum: 0.000000
2023-10-18 22:12:29,543 epoch 3 - iter 144/723 - loss 0.19123649 - time (sec): 3.53 - samples/sec: 9837.27 - lr: 0.000043 - momentum: 0.000000
2023-10-18 22:12:31,289 epoch 3 - iter 216/723 - loss 0.18622492 - time (sec): 5.28 - samples/sec: 9893.55 - lr: 0.000043 - momentum: 0.000000
2023-10-18 22:12:33,145 epoch 3 - iter 288/723 - loss 0.17538802 - time (sec): 7.13 - samples/sec: 9922.72 - lr: 0.000042 - momentum: 0.000000
2023-10-18 22:12:34,879 epoch 3 - iter 360/723 - loss 0.17611275 - time (sec): 8.87 - samples/sec: 9917.65 - lr: 0.000042 - momentum: 0.000000
2023-10-18 22:12:36,750 epoch 3 - iter 432/723 - loss 0.17600510 - time (sec): 10.74 - samples/sec: 9821.66 - lr: 0.000041 - momentum: 0.000000
2023-10-18 22:12:38,475 epoch 3 - iter 504/723 - loss 0.17653894 - time (sec): 12.46 - samples/sec: 9798.39 - lr: 0.000041 - momentum: 0.000000
2023-10-18 22:12:40,261 epoch 3 - iter 576/723 - loss 0.17866614 - time (sec): 14.25 - samples/sec: 9817.94 - lr: 0.000040 - momentum: 0.000000
2023-10-18 22:12:42,033 epoch 3 - iter 648/723 - loss 0.17625735 - time (sec): 16.02 - samples/sec: 9858.26 - lr: 0.000039 - momentum: 0.000000
2023-10-18 22:12:43,814 epoch 3 - iter 720/723 - loss 0.17698859 - time (sec): 17.80 - samples/sec: 9872.51 - lr: 0.000039 - momentum: 0.000000
2023-10-18 22:12:43,871 ----------------------------------------------------------------------------------------------------
2023-10-18 22:12:43,871 EPOCH 3 done: loss 0.1768 - lr: 0.000039
2023-10-18 22:12:45,631 DEV : loss 0.21366006135940552 - f1-score (micro avg) 0.3814
2023-10-18 22:12:45,646 saving best model
2023-10-18 22:12:45,684 ----------------------------------------------------------------------------------------------------
2023-10-18 22:12:47,421 epoch 4 - iter 72/723 - loss 0.15319449 - time (sec): 1.74 - samples/sec: 10153.35 - lr: 0.000038 - momentum: 0.000000
2023-10-18 22:12:49,178 epoch 4 - iter 144/723 - loss 0.15219794 - time (sec): 3.49 - samples/sec: 9871.73 - lr: 0.000038 - momentum: 0.000000
2023-10-18 22:12:50,904 epoch 4 - iter 216/723 - loss 0.15958723 - time (sec): 5.22 - samples/sec: 10037.56 - lr: 0.000037 - momentum: 0.000000
2023-10-18 22:12:52,729 epoch 4 - iter 288/723 - loss 0.15641675 - time (sec): 7.04 - samples/sec: 10141.55 - lr: 0.000037 - momentum: 0.000000
2023-10-18 22:12:54,430 epoch 4 - iter 360/723 - loss 0.15539065 - time (sec): 8.74 - samples/sec: 10131.07 - lr: 0.000036 - momentum: 0.000000
2023-10-18 22:12:56,173 epoch 4 - iter 432/723 - loss 0.15963557 - time (sec): 10.49 - samples/sec: 10188.50 - lr: 0.000036 - momentum: 0.000000
2023-10-18 22:12:58,321 epoch 4 - iter 504/723 - loss 0.15900354 - time (sec): 12.64 - samples/sec: 9892.63 - lr: 0.000035 - momentum: 0.000000
2023-10-18 22:13:00,068 epoch 4 - iter 576/723 - loss 0.15775232 - time (sec): 14.38 - samples/sec: 9897.82 - lr: 0.000034 - momentum: 0.000000
2023-10-18 22:13:01,795 epoch 4 - iter 648/723 - loss 0.15793963 - time (sec): 16.11 - samples/sec: 9854.22 - lr: 0.000034 - momentum: 0.000000
2023-10-18 22:13:03,655 epoch 4 - iter 720/723 - loss 0.16172244 - time (sec): 17.97 - samples/sec: 9776.09 - lr: 0.000033 - momentum: 0.000000
2023-10-18 22:13:03,711 ----------------------------------------------------------------------------------------------------
2023-10-18 22:13:03,711 EPOCH 4 done: loss 0.1617 - lr: 0.000033
2023-10-18 22:13:05,482 DEV : loss 0.19289655983448029 - f1-score (micro avg) 0.4815
2023-10-18 22:13:05,497 saving best model
2023-10-18 22:13:05,533 ----------------------------------------------------------------------------------------------------
2023-10-18 22:13:07,441 epoch 5 - iter 72/723 - loss 0.16811055 - time (sec): 1.91 - samples/sec: 9536.49 - lr: 0.000033 - momentum: 0.000000
2023-10-18 22:13:09,201 epoch 5 - iter 144/723 - loss 0.15925998 - time (sec): 3.67 - samples/sec: 9799.25 - lr: 0.000032 - momentum: 0.000000
2023-10-18 22:13:10,943 epoch 5 - iter 216/723 - loss 0.15623823 - time (sec): 5.41 - samples/sec: 9557.30 - lr: 0.000032 - momentum: 0.000000
2023-10-18 22:13:12,686 epoch 5 - iter 288/723 - loss 0.15588053 - time (sec): 7.15 - samples/sec: 9547.60 - lr: 0.000031 - momentum: 0.000000
2023-10-18 22:13:14,425 epoch 5 - iter 360/723 - loss 0.15281731 - time (sec): 8.89 - samples/sec: 9559.96 - lr: 0.000031 - momentum: 0.000000
2023-10-18 22:13:16,265 epoch 5 - iter 432/723 - loss 0.15073615 - time (sec): 10.73 - samples/sec: 9668.65 - lr: 0.000030 - momentum: 0.000000
2023-10-18 22:13:17,926 epoch 5 - iter 504/723 - loss 0.14950718 - time (sec): 12.39 - samples/sec: 9809.91 - lr: 0.000029 - momentum: 0.000000
2023-10-18 22:13:19,596 epoch 5 - iter 576/723 - loss 0.14885470 - time (sec): 14.06 - samples/sec: 9898.68 - lr: 0.000029 - momentum: 0.000000
2023-10-18 22:13:21,379 epoch 5 - iter 648/723 - loss 0.15131920 - time (sec): 15.84 - samples/sec: 9902.45 - lr: 0.000028 - momentum: 0.000000
2023-10-18 22:13:23,193 epoch 5 - iter 720/723 - loss 0.15022435 - time (sec): 17.66 - samples/sec: 9943.96 - lr: 0.000028 - momentum: 0.000000
2023-10-18 22:13:23,253 ----------------------------------------------------------------------------------------------------
2023-10-18 22:13:23,253 EPOCH 5 done: loss 0.1504 - lr: 0.000028
2023-10-18 22:13:25,007 DEV : loss 0.1961345225572586 - f1-score (micro avg) 0.4696
2023-10-18 22:13:25,022 ----------------------------------------------------------------------------------------------------
2023-10-18 22:13:26,804 epoch 6 - iter 72/723 - loss 0.13617864 - time (sec): 1.78 - samples/sec: 9559.29 - lr: 0.000027 - momentum: 0.000000
2023-10-18 22:13:28,589 epoch 6 - iter 144/723 - loss 0.14005984 - time (sec): 3.57 - samples/sec: 9628.21 - lr: 0.000027 - momentum: 0.000000
2023-10-18 22:13:30,437 epoch 6 - iter 216/723 - loss 0.15059229 - time (sec): 5.42 - samples/sec: 9626.66 - lr: 0.000026 - momentum: 0.000000
2023-10-18 22:13:32,163 epoch 6 - iter 288/723 - loss 0.15216103 - time (sec): 7.14 - samples/sec: 9579.75 - lr: 0.000026 - momentum: 0.000000
2023-10-18 22:13:34,396 epoch 6 - iter 360/723 - loss 0.14557675 - time (sec): 9.37 - samples/sec: 9231.32 - lr: 0.000025 - momentum: 0.000000
2023-10-18 22:13:36,183 epoch 6 - iter 432/723 - loss 0.14252551 - time (sec): 11.16 - samples/sec: 9279.64 - lr: 0.000024 - momentum: 0.000000
2023-10-18 22:13:37,996 epoch 6 - iter 504/723 - loss 0.14416986 - time (sec): 12.97 - samples/sec: 9456.22 - lr: 0.000024 - momentum: 0.000000
2023-10-18 22:13:39,763 epoch 6 - iter 576/723 - loss 0.14274090 - time (sec): 14.74 - samples/sec: 9548.14 - lr: 0.000023 - momentum: 0.000000
2023-10-18 22:13:41,578 epoch 6 - iter 648/723 - loss 0.14378379 - time (sec): 16.56 - samples/sec: 9609.62 - lr: 0.000023 - momentum: 0.000000
2023-10-18 22:13:43,291 epoch 6 - iter 720/723 - loss 0.14185352 - time (sec): 18.27 - samples/sec: 9606.77 - lr: 0.000022 - momentum: 0.000000
2023-10-18 22:13:43,355 ----------------------------------------------------------------------------------------------------
2023-10-18 22:13:43,356 EPOCH 6 done: loss 0.1414 - lr: 0.000022
2023-10-18 22:13:45,122 DEV : loss 0.19116047024726868 - f1-score (micro avg) 0.4857
2023-10-18 22:13:45,137 saving best model
2023-10-18 22:13:45,174 ----------------------------------------------------------------------------------------------------
2023-10-18 22:13:46,912 epoch 7 - iter 72/723 - loss 0.13529263 - time (sec): 1.74 - samples/sec: 9722.45 - lr: 0.000022 - momentum: 0.000000
2023-10-18 22:13:48,713 epoch 7 - iter 144/723 - loss 0.13613335 - time (sec): 3.54 - samples/sec: 9958.96 - lr: 0.000021 - momentum: 0.000000
2023-10-18 22:13:50,482 epoch 7 - iter 216/723 - loss 0.13441868 - time (sec): 5.31 - samples/sec: 9917.67 - lr: 0.000021 - momentum: 0.000000
2023-10-18 22:13:52,251 epoch 7 - iter 288/723 - loss 0.13760997 - time (sec): 7.08 - samples/sec: 9848.22 - lr: 0.000020 - momentum: 0.000000
2023-10-18 22:13:53,993 epoch 7 - iter 360/723 - loss 0.13573573 - time (sec): 8.82 - samples/sec: 9819.71 - lr: 0.000019 - momentum: 0.000000
2023-10-18 22:13:55,820 epoch 7 - iter 432/723 - loss 0.13560010 - time (sec): 10.65 - samples/sec: 9890.16 - lr: 0.000019 - momentum: 0.000000
2023-10-18 22:13:57,634 epoch 7 - iter 504/723 - loss 0.13364674 - time (sec): 12.46 - samples/sec: 9840.57 - lr: 0.000018 - momentum: 0.000000
2023-10-18 22:13:59,388 epoch 7 - iter 576/723 - loss 0.13450202 - time (sec): 14.21 - samples/sec: 9772.70 - lr: 0.000018 - momentum: 0.000000
2023-10-18 22:14:01,207 epoch 7 - iter 648/723 - loss 0.13529263 - time (sec): 16.03 - samples/sec: 9801.38 - lr: 0.000017 - momentum: 0.000000
2023-10-18 22:14:03,065 epoch 7 - iter 720/723 - loss 0.13363141 - time (sec): 17.89 - samples/sec: 9814.24 - lr: 0.000017 - momentum: 0.000000
2023-10-18 22:14:03,126 ----------------------------------------------------------------------------------------------------
2023-10-18 22:14:03,127 EPOCH 7 done: loss 0.1334 - lr: 0.000017
2023-10-18 22:14:04,891 DEV : loss 0.1891321837902069 - f1-score (micro avg) 0.4815
2023-10-18 22:14:04,906 ----------------------------------------------------------------------------------------------------
2023-10-18 22:14:06,635 epoch 8 - iter 72/723 - loss 0.12504995 - time (sec): 1.73 - samples/sec: 9327.36 - lr: 0.000016 - momentum: 0.000000
2023-10-18 22:14:08,831 epoch 8 - iter 144/723 - loss 0.14497774 - time (sec): 3.92 - samples/sec: 8703.68 - lr: 0.000016 - momentum: 0.000000
2023-10-18 22:14:10,623 epoch 8 - iter 216/723 - loss 0.13421076 - time (sec): 5.72 - samples/sec: 9270.31 - lr: 0.000015 - momentum: 0.000000
2023-10-18 22:14:12,411 epoch 8 - iter 288/723 - loss 0.12980263 - time (sec): 7.50 - samples/sec: 9387.38 - lr: 0.000014 - momentum: 0.000000
2023-10-18 22:14:14,168 epoch 8 - iter 360/723 - loss 0.12913939 - time (sec): 9.26 - samples/sec: 9523.64 - lr: 0.000014 - momentum: 0.000000
2023-10-18 22:14:16,043 epoch 8 - iter 432/723 - loss 0.12544526 - time (sec): 11.14 - samples/sec: 9576.42 - lr: 0.000013 - momentum: 0.000000
2023-10-18 22:14:17,799 epoch 8 - iter 504/723 - loss 0.12520676 - time (sec): 12.89 - samples/sec: 9556.09 - lr: 0.000013 - momentum: 0.000000
2023-10-18 22:14:19,682 epoch 8 - iter 576/723 - loss 0.12476139 - time (sec): 14.78 - samples/sec: 9558.26 - lr: 0.000012 - momentum: 0.000000
2023-10-18 22:14:21,488 epoch 8 - iter 648/723 - loss 0.12586790 - time (sec): 16.58 - samples/sec: 9535.81 - lr: 0.000012 - momentum: 0.000000
2023-10-18 22:14:23,247 epoch 8 - iter 720/723 - loss 0.12811874 - time (sec): 18.34 - samples/sec: 9586.01 - lr: 0.000011 - momentum: 0.000000
2023-10-18 22:14:23,303 ----------------------------------------------------------------------------------------------------
2023-10-18 22:14:23,303 EPOCH 8 done: loss 0.1279 - lr: 0.000011
2023-10-18 22:14:25,072 DEV : loss 0.18201254308223724 - f1-score (micro avg) 0.5072
2023-10-18 22:14:25,087 saving best model
2023-10-18 22:14:25,123 ----------------------------------------------------------------------------------------------------
2023-10-18 22:14:26,935 epoch 9 - iter 72/723 - loss 0.11411652 - time (sec): 1.81 - samples/sec: 10766.66 - lr: 0.000011 - momentum: 0.000000
2023-10-18 22:14:28,686 epoch 9 - iter 144/723 - loss 0.10939075 - time (sec): 3.56 - samples/sec: 10321.51 - lr: 0.000010 - momentum: 0.000000
2023-10-18 22:14:30,443 epoch 9 - iter 216/723 - loss 0.11384251 - time (sec): 5.32 - samples/sec: 10057.20 - lr: 0.000009 - momentum: 0.000000
2023-10-18 22:14:32,214 epoch 9 - iter 288/723 - loss 0.11951860 - time (sec): 7.09 - samples/sec: 9978.23 - lr: 0.000009 - momentum: 0.000000
2023-10-18 22:14:34,027 epoch 9 - iter 360/723 - loss 0.12179869 - time (sec): 8.90 - samples/sec: 9931.67 - lr: 0.000008 - momentum: 0.000000
2023-10-18 22:14:35,758 epoch 9 - iter 432/723 - loss 0.12431945 - time (sec): 10.63 - samples/sec: 9850.04 - lr: 0.000008 - momentum: 0.000000
2023-10-18 22:14:37,499 epoch 9 - iter 504/723 - loss 0.12584957 - time (sec): 12.38 - samples/sec: 9780.76 - lr: 0.000007 - momentum: 0.000000
2023-10-18 22:14:39,356 epoch 9 - iter 576/723 - loss 0.12515093 - time (sec): 14.23 - samples/sec: 9893.29 - lr: 0.000007 - momentum: 0.000000
2023-10-18 22:14:41,122 epoch 9 - iter 648/723 - loss 0.12556555 - time (sec): 16.00 - samples/sec: 9903.79 - lr: 0.000006 - momentum: 0.000000
2023-10-18 22:14:42,893 epoch 9 - iter 720/723 - loss 0.12475240 - time (sec): 17.77 - samples/sec: 9882.46 - lr: 0.000006 - momentum: 0.000000
2023-10-18 22:14:42,962 ----------------------------------------------------------------------------------------------------
2023-10-18 22:14:42,962 EPOCH 9 done: loss 0.1247 - lr: 0.000006
2023-10-18 22:14:45,097 DEV : loss 0.1882598102092743 - f1-score (micro avg) 0.4959
2023-10-18 22:14:45,112 ----------------------------------------------------------------------------------------------------
2023-10-18 22:14:46,869 epoch 10 - iter 72/723 - loss 0.10866994 - time (sec): 1.76 - samples/sec: 9731.03 - lr: 0.000005 - momentum: 0.000000
2023-10-18 22:14:48,671 epoch 10 - iter 144/723 - loss 0.12562583 - time (sec): 3.56 - samples/sec: 9529.64 - lr: 0.000004 - momentum: 0.000000
2023-10-18 22:14:50,455 epoch 10 - iter 216/723 - loss 0.12325727 - time (sec): 5.34 - samples/sec: 9768.34 - lr: 0.000004 - momentum: 0.000000
2023-10-18 22:14:52,279 epoch 10 - iter 288/723 - loss 0.12513734 - time (sec): 7.17 - samples/sec: 9689.81 - lr: 0.000003 - momentum: 0.000000
2023-10-18 22:14:54,208 epoch 10 - iter 360/723 - loss 0.13109443 - time (sec): 9.10 - samples/sec: 9713.73 - lr: 0.000003 - momentum: 0.000000
2023-10-18 22:14:55,958 epoch 10 - iter 432/723 - loss 0.12975067 - time (sec): 10.85 - samples/sec: 9720.21 - lr: 0.000002 - momentum: 0.000000
2023-10-18 22:14:57,713 epoch 10 - iter 504/723 - loss 0.12653515 - time (sec): 12.60 - samples/sec: 9811.00 - lr: 0.000002 - momentum: 0.000000
2023-10-18 22:14:59,552 epoch 10 - iter 576/723 - loss 0.12518131 - time (sec): 14.44 - samples/sec: 9762.73 - lr: 0.000001 - momentum: 0.000000
2023-10-18 22:15:01,283 epoch 10 - iter 648/723 - loss 0.12318594 - time (sec): 16.17 - samples/sec: 9754.98 - lr: 0.000001 - momentum: 0.000000
2023-10-18 22:15:03,005 epoch 10 - iter 720/723 - loss 0.12424305 - time (sec): 17.89 - samples/sec: 9814.44 - lr: 0.000000 - momentum: 0.000000
2023-10-18 22:15:03,068 ----------------------------------------------------------------------------------------------------
2023-10-18 22:15:03,068 EPOCH 10 done: loss 0.1244 - lr: 0.000000
2023-10-18 22:15:04,871 DEV : loss 0.1862431764602661 - f1-score (micro avg) 0.4991
2023-10-18 22:15:04,918 ----------------------------------------------------------------------------------------------------
2023-10-18 22:15:04,918 Loading model from best epoch ...
2023-10-18 22:15:05,001 SequenceTagger predicts: Dictionary with 13 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
2023-10-18 22:15:06,339
Results:
- F-score (micro) 0.5138
- F-score (macro) 0.3616
- Accuracy 0.3566
By class:
precision recall f1-score support
LOC 0.5583 0.6485 0.6000 458
PER 0.4975 0.4212 0.4562 482
ORG 1.0000 0.0145 0.0286 69
micro avg 0.5324 0.4965 0.5138 1009
macro avg 0.6853 0.3614 0.3616 1009
weighted avg 0.5595 0.4965 0.4922 1009
2023-10-18 22:15:06,339 ----------------------------------------------------------------------------------------------------
|