File size: 24,167 Bytes
8734368 |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 |
2023-10-18 14:34:24,829 ----------------------------------------------------------------------------------------------------
2023-10-18 14:34:24,829 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=25, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-18 14:34:24,829 ----------------------------------------------------------------------------------------------------
2023-10-18 14:34:24,830 MultiCorpus: 1100 train + 206 dev + 240 test sentences
- NER_HIPE_2022 Corpus: 1100 train + 206 dev + 240 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/ajmc/de/with_doc_seperator
2023-10-18 14:34:24,830 ----------------------------------------------------------------------------------------------------
2023-10-18 14:34:24,830 Train: 1100 sentences
2023-10-18 14:34:24,830 (train_with_dev=False, train_with_test=False)
2023-10-18 14:34:24,830 ----------------------------------------------------------------------------------------------------
2023-10-18 14:34:24,830 Training Params:
2023-10-18 14:34:24,830 - learning_rate: "5e-05"
2023-10-18 14:34:24,830 - mini_batch_size: "4"
2023-10-18 14:34:24,830 - max_epochs: "10"
2023-10-18 14:34:24,830 - shuffle: "True"
2023-10-18 14:34:24,830 ----------------------------------------------------------------------------------------------------
2023-10-18 14:34:24,830 Plugins:
2023-10-18 14:34:24,830 - TensorboardLogger
2023-10-18 14:34:24,830 - LinearScheduler | warmup_fraction: '0.1'
2023-10-18 14:34:24,830 ----------------------------------------------------------------------------------------------------
2023-10-18 14:34:24,830 Final evaluation on model from best epoch (best-model.pt)
2023-10-18 14:34:24,830 - metric: "('micro avg', 'f1-score')"
2023-10-18 14:34:24,830 ----------------------------------------------------------------------------------------------------
2023-10-18 14:34:24,830 Computation:
2023-10-18 14:34:24,830 - compute on device: cuda:0
2023-10-18 14:34:24,830 - embedding storage: none
2023-10-18 14:34:24,830 ----------------------------------------------------------------------------------------------------
2023-10-18 14:34:24,830 Model training base path: "hmbench-ajmc/de-dbmdz/bert-tiny-historic-multilingual-cased-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-1"
2023-10-18 14:34:24,830 ----------------------------------------------------------------------------------------------------
2023-10-18 14:34:24,830 ----------------------------------------------------------------------------------------------------
2023-10-18 14:34:24,830 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-18 14:34:25,241 epoch 1 - iter 27/275 - loss 3.94457406 - time (sec): 0.41 - samples/sec: 5973.53 - lr: 0.000005 - momentum: 0.000000
2023-10-18 14:34:25,651 epoch 1 - iter 54/275 - loss 3.97957719 - time (sec): 0.82 - samples/sec: 5721.37 - lr: 0.000010 - momentum: 0.000000
2023-10-18 14:34:26,051 epoch 1 - iter 81/275 - loss 3.82844636 - time (sec): 1.22 - samples/sec: 5541.40 - lr: 0.000015 - momentum: 0.000000
2023-10-18 14:34:26,457 epoch 1 - iter 108/275 - loss 3.64553980 - time (sec): 1.63 - samples/sec: 5422.35 - lr: 0.000019 - momentum: 0.000000
2023-10-18 14:34:26,871 epoch 1 - iter 135/275 - loss 3.35546642 - time (sec): 2.04 - samples/sec: 5558.03 - lr: 0.000024 - momentum: 0.000000
2023-10-18 14:34:27,288 epoch 1 - iter 162/275 - loss 3.06649694 - time (sec): 2.46 - samples/sec: 5526.20 - lr: 0.000029 - momentum: 0.000000
2023-10-18 14:34:27,700 epoch 1 - iter 189/275 - loss 2.82331060 - time (sec): 2.87 - samples/sec: 5497.83 - lr: 0.000034 - momentum: 0.000000
2023-10-18 14:34:28,102 epoch 1 - iter 216/275 - loss 2.59254150 - time (sec): 3.27 - samples/sec: 5551.63 - lr: 0.000039 - momentum: 0.000000
2023-10-18 14:34:28,507 epoch 1 - iter 243/275 - loss 2.42230854 - time (sec): 3.68 - samples/sec: 5476.43 - lr: 0.000044 - momentum: 0.000000
2023-10-18 14:34:28,903 epoch 1 - iter 270/275 - loss 2.29568740 - time (sec): 4.07 - samples/sec: 5490.59 - lr: 0.000049 - momentum: 0.000000
2023-10-18 14:34:28,979 ----------------------------------------------------------------------------------------------------
2023-10-18 14:34:28,979 EPOCH 1 done: loss 2.2689 - lr: 0.000049
2023-10-18 14:34:29,218 DEV : loss 0.8229100704193115 - f1-score (micro avg) 0.0
2023-10-18 14:34:29,222 ----------------------------------------------------------------------------------------------------
2023-10-18 14:34:29,620 epoch 2 - iter 27/275 - loss 0.84797441 - time (sec): 0.40 - samples/sec: 6212.02 - lr: 0.000049 - momentum: 0.000000
2023-10-18 14:34:30,019 epoch 2 - iter 54/275 - loss 0.89177898 - time (sec): 0.80 - samples/sec: 5916.38 - lr: 0.000049 - momentum: 0.000000
2023-10-18 14:34:30,418 epoch 2 - iter 81/275 - loss 0.91951425 - time (sec): 1.20 - samples/sec: 5871.17 - lr: 0.000048 - momentum: 0.000000
2023-10-18 14:34:30,826 epoch 2 - iter 108/275 - loss 0.89409210 - time (sec): 1.60 - samples/sec: 5771.55 - lr: 0.000048 - momentum: 0.000000
2023-10-18 14:34:31,201 epoch 2 - iter 135/275 - loss 0.88598127 - time (sec): 1.98 - samples/sec: 5696.78 - lr: 0.000047 - momentum: 0.000000
2023-10-18 14:34:31,610 epoch 2 - iter 162/275 - loss 0.85285616 - time (sec): 2.39 - samples/sec: 5704.02 - lr: 0.000047 - momentum: 0.000000
2023-10-18 14:34:32,021 epoch 2 - iter 189/275 - loss 0.82927359 - time (sec): 2.80 - samples/sec: 5684.11 - lr: 0.000046 - momentum: 0.000000
2023-10-18 14:34:32,418 epoch 2 - iter 216/275 - loss 0.80856883 - time (sec): 3.20 - samples/sec: 5597.78 - lr: 0.000046 - momentum: 0.000000
2023-10-18 14:34:32,831 epoch 2 - iter 243/275 - loss 0.78287459 - time (sec): 3.61 - samples/sec: 5624.79 - lr: 0.000045 - momentum: 0.000000
2023-10-18 14:34:33,243 epoch 2 - iter 270/275 - loss 0.76347923 - time (sec): 4.02 - samples/sec: 5554.65 - lr: 0.000045 - momentum: 0.000000
2023-10-18 14:34:33,315 ----------------------------------------------------------------------------------------------------
2023-10-18 14:34:33,315 EPOCH 2 done: loss 0.7612 - lr: 0.000045
2023-10-18 14:34:33,669 DEV : loss 0.5238796472549438 - f1-score (micro avg) 0.1895
2023-10-18 14:34:33,673 saving best model
2023-10-18 14:34:33,704 ----------------------------------------------------------------------------------------------------
2023-10-18 14:34:34,111 epoch 3 - iter 27/275 - loss 0.63330269 - time (sec): 0.41 - samples/sec: 5574.87 - lr: 0.000044 - momentum: 0.000000
2023-10-18 14:34:34,519 epoch 3 - iter 54/275 - loss 0.57059336 - time (sec): 0.81 - samples/sec: 5551.88 - lr: 0.000043 - momentum: 0.000000
2023-10-18 14:34:34,925 epoch 3 - iter 81/275 - loss 0.59892710 - time (sec): 1.22 - samples/sec: 5646.07 - lr: 0.000043 - momentum: 0.000000
2023-10-18 14:34:35,480 epoch 3 - iter 108/275 - loss 0.57642150 - time (sec): 1.77 - samples/sec: 5218.10 - lr: 0.000042 - momentum: 0.000000
2023-10-18 14:34:35,882 epoch 3 - iter 135/275 - loss 0.57839286 - time (sec): 2.18 - samples/sec: 5263.09 - lr: 0.000042 - momentum: 0.000000
2023-10-18 14:34:36,322 epoch 3 - iter 162/275 - loss 0.56175351 - time (sec): 2.62 - samples/sec: 5233.27 - lr: 0.000041 - momentum: 0.000000
2023-10-18 14:34:36,746 epoch 3 - iter 189/275 - loss 0.55386261 - time (sec): 3.04 - samples/sec: 5233.22 - lr: 0.000041 - momentum: 0.000000
2023-10-18 14:34:37,157 epoch 3 - iter 216/275 - loss 0.55331858 - time (sec): 3.45 - samples/sec: 5239.51 - lr: 0.000040 - momentum: 0.000000
2023-10-18 14:34:37,574 epoch 3 - iter 243/275 - loss 0.55295982 - time (sec): 3.87 - samples/sec: 5236.73 - lr: 0.000040 - momentum: 0.000000
2023-10-18 14:34:37,979 epoch 3 - iter 270/275 - loss 0.55071202 - time (sec): 4.27 - samples/sec: 5252.53 - lr: 0.000039 - momentum: 0.000000
2023-10-18 14:34:38,055 ----------------------------------------------------------------------------------------------------
2023-10-18 14:34:38,055 EPOCH 3 done: loss 0.5485 - lr: 0.000039
2023-10-18 14:34:38,415 DEV : loss 0.382687509059906 - f1-score (micro avg) 0.4746
2023-10-18 14:34:38,419 saving best model
2023-10-18 14:34:38,454 ----------------------------------------------------------------------------------------------------
2023-10-18 14:34:38,866 epoch 4 - iter 27/275 - loss 0.52440291 - time (sec): 0.41 - samples/sec: 5364.19 - lr: 0.000038 - momentum: 0.000000
2023-10-18 14:34:39,254 epoch 4 - iter 54/275 - loss 0.49379413 - time (sec): 0.80 - samples/sec: 5455.30 - lr: 0.000038 - momentum: 0.000000
2023-10-18 14:34:39,668 epoch 4 - iter 81/275 - loss 0.46685441 - time (sec): 1.21 - samples/sec: 5269.79 - lr: 0.000037 - momentum: 0.000000
2023-10-18 14:34:40,086 epoch 4 - iter 108/275 - loss 0.48160009 - time (sec): 1.63 - samples/sec: 5271.61 - lr: 0.000037 - momentum: 0.000000
2023-10-18 14:34:40,522 epoch 4 - iter 135/275 - loss 0.46798370 - time (sec): 2.07 - samples/sec: 5408.94 - lr: 0.000036 - momentum: 0.000000
2023-10-18 14:34:40,947 epoch 4 - iter 162/275 - loss 0.45395448 - time (sec): 2.49 - samples/sec: 5374.43 - lr: 0.000036 - momentum: 0.000000
2023-10-18 14:34:41,367 epoch 4 - iter 189/275 - loss 0.45338045 - time (sec): 2.91 - samples/sec: 5377.17 - lr: 0.000035 - momentum: 0.000000
2023-10-18 14:34:41,776 epoch 4 - iter 216/275 - loss 0.44578276 - time (sec): 3.32 - samples/sec: 5410.32 - lr: 0.000035 - momentum: 0.000000
2023-10-18 14:34:42,192 epoch 4 - iter 243/275 - loss 0.44115371 - time (sec): 3.74 - samples/sec: 5409.32 - lr: 0.000034 - momentum: 0.000000
2023-10-18 14:34:42,606 epoch 4 - iter 270/275 - loss 0.43517083 - time (sec): 4.15 - samples/sec: 5379.63 - lr: 0.000034 - momentum: 0.000000
2023-10-18 14:34:42,692 ----------------------------------------------------------------------------------------------------
2023-10-18 14:34:42,692 EPOCH 4 done: loss 0.4346 - lr: 0.000034
2023-10-18 14:34:43,058 DEV : loss 0.33680784702301025 - f1-score (micro avg) 0.5485
2023-10-18 14:34:43,062 saving best model
2023-10-18 14:34:43,096 ----------------------------------------------------------------------------------------------------
2023-10-18 14:34:43,497 epoch 5 - iter 27/275 - loss 0.34448775 - time (sec): 0.40 - samples/sec: 6541.62 - lr: 0.000033 - momentum: 0.000000
2023-10-18 14:34:43,884 epoch 5 - iter 54/275 - loss 0.35793341 - time (sec): 0.79 - samples/sec: 5918.05 - lr: 0.000032 - momentum: 0.000000
2023-10-18 14:34:44,276 epoch 5 - iter 81/275 - loss 0.37488252 - time (sec): 1.18 - samples/sec: 5855.39 - lr: 0.000032 - momentum: 0.000000
2023-10-18 14:34:44,689 epoch 5 - iter 108/275 - loss 0.38765131 - time (sec): 1.59 - samples/sec: 5872.04 - lr: 0.000031 - momentum: 0.000000
2023-10-18 14:34:45,096 epoch 5 - iter 135/275 - loss 0.38999235 - time (sec): 2.00 - samples/sec: 5703.31 - lr: 0.000031 - momentum: 0.000000
2023-10-18 14:34:45,505 epoch 5 - iter 162/275 - loss 0.38841687 - time (sec): 2.41 - samples/sec: 5644.53 - lr: 0.000030 - momentum: 0.000000
2023-10-18 14:34:45,905 epoch 5 - iter 189/275 - loss 0.38974761 - time (sec): 2.81 - samples/sec: 5610.05 - lr: 0.000030 - momentum: 0.000000
2023-10-18 14:34:46,321 epoch 5 - iter 216/275 - loss 0.39216271 - time (sec): 3.23 - samples/sec: 5552.62 - lr: 0.000029 - momentum: 0.000000
2023-10-18 14:34:46,737 epoch 5 - iter 243/275 - loss 0.38836120 - time (sec): 3.64 - samples/sec: 5498.94 - lr: 0.000029 - momentum: 0.000000
2023-10-18 14:34:47,146 epoch 5 - iter 270/275 - loss 0.39178740 - time (sec): 4.05 - samples/sec: 5524.38 - lr: 0.000028 - momentum: 0.000000
2023-10-18 14:34:47,217 ----------------------------------------------------------------------------------------------------
2023-10-18 14:34:47,217 EPOCH 5 done: loss 0.3892 - lr: 0.000028
2023-10-18 14:34:47,587 DEV : loss 0.286026269197464 - f1-score (micro avg) 0.6083
2023-10-18 14:34:47,591 saving best model
2023-10-18 14:34:47,626 ----------------------------------------------------------------------------------------------------
2023-10-18 14:34:48,018 epoch 6 - iter 27/275 - loss 0.48487282 - time (sec): 0.39 - samples/sec: 5038.91 - lr: 0.000027 - momentum: 0.000000
2023-10-18 14:34:48,409 epoch 6 - iter 54/275 - loss 0.39612221 - time (sec): 0.78 - samples/sec: 5401.34 - lr: 0.000027 - momentum: 0.000000
2023-10-18 14:34:48,806 epoch 6 - iter 81/275 - loss 0.39666419 - time (sec): 1.18 - samples/sec: 5570.88 - lr: 0.000026 - momentum: 0.000000
2023-10-18 14:34:49,200 epoch 6 - iter 108/275 - loss 0.38652012 - time (sec): 1.57 - samples/sec: 5614.65 - lr: 0.000026 - momentum: 0.000000
2023-10-18 14:34:49,610 epoch 6 - iter 135/275 - loss 0.36749916 - time (sec): 1.98 - samples/sec: 5518.23 - lr: 0.000025 - momentum: 0.000000
2023-10-18 14:34:50,028 epoch 6 - iter 162/275 - loss 0.36658781 - time (sec): 2.40 - samples/sec: 5590.13 - lr: 0.000025 - momentum: 0.000000
2023-10-18 14:34:50,430 epoch 6 - iter 189/275 - loss 0.35919349 - time (sec): 2.80 - samples/sec: 5581.01 - lr: 0.000024 - momentum: 0.000000
2023-10-18 14:34:50,827 epoch 6 - iter 216/275 - loss 0.34957161 - time (sec): 3.20 - samples/sec: 5574.48 - lr: 0.000024 - momentum: 0.000000
2023-10-18 14:34:51,248 epoch 6 - iter 243/275 - loss 0.35441459 - time (sec): 3.62 - samples/sec: 5600.18 - lr: 0.000023 - momentum: 0.000000
2023-10-18 14:34:51,648 epoch 6 - iter 270/275 - loss 0.35201354 - time (sec): 4.02 - samples/sec: 5573.54 - lr: 0.000022 - momentum: 0.000000
2023-10-18 14:34:51,725 ----------------------------------------------------------------------------------------------------
2023-10-18 14:34:51,725 EPOCH 6 done: loss 0.3508 - lr: 0.000022
2023-10-18 14:34:52,106 DEV : loss 0.26645413041114807 - f1-score (micro avg) 0.6255
2023-10-18 14:34:52,111 saving best model
2023-10-18 14:34:52,145 ----------------------------------------------------------------------------------------------------
2023-10-18 14:34:52,555 epoch 7 - iter 27/275 - loss 0.37548681 - time (sec): 0.41 - samples/sec: 5082.83 - lr: 0.000022 - momentum: 0.000000
2023-10-18 14:34:52,958 epoch 7 - iter 54/275 - loss 0.34514987 - time (sec): 0.81 - samples/sec: 5253.35 - lr: 0.000021 - momentum: 0.000000
2023-10-18 14:34:53,358 epoch 7 - iter 81/275 - loss 0.35183308 - time (sec): 1.21 - samples/sec: 5228.42 - lr: 0.000021 - momentum: 0.000000
2023-10-18 14:34:53,767 epoch 7 - iter 108/275 - loss 0.34792121 - time (sec): 1.62 - samples/sec: 5148.08 - lr: 0.000020 - momentum: 0.000000
2023-10-18 14:34:54,188 epoch 7 - iter 135/275 - loss 0.34092810 - time (sec): 2.04 - samples/sec: 5228.21 - lr: 0.000020 - momentum: 0.000000
2023-10-18 14:34:54,594 epoch 7 - iter 162/275 - loss 0.33335193 - time (sec): 2.45 - samples/sec: 5352.35 - lr: 0.000019 - momentum: 0.000000
2023-10-18 14:34:55,011 epoch 7 - iter 189/275 - loss 0.33956542 - time (sec): 2.87 - samples/sec: 5392.50 - lr: 0.000019 - momentum: 0.000000
2023-10-18 14:34:55,420 epoch 7 - iter 216/275 - loss 0.33319421 - time (sec): 3.27 - samples/sec: 5427.20 - lr: 0.000018 - momentum: 0.000000
2023-10-18 14:34:55,832 epoch 7 - iter 243/275 - loss 0.32691358 - time (sec): 3.69 - samples/sec: 5474.15 - lr: 0.000017 - momentum: 0.000000
2023-10-18 14:34:56,241 epoch 7 - iter 270/275 - loss 0.32870171 - time (sec): 4.10 - samples/sec: 5479.50 - lr: 0.000017 - momentum: 0.000000
2023-10-18 14:34:56,313 ----------------------------------------------------------------------------------------------------
2023-10-18 14:34:56,313 EPOCH 7 done: loss 0.3272 - lr: 0.000017
2023-10-18 14:34:56,679 DEV : loss 0.2574481666088104 - f1-score (micro avg) 0.6277
2023-10-18 14:34:56,683 saving best model
2023-10-18 14:34:56,718 ----------------------------------------------------------------------------------------------------
2023-10-18 14:34:57,134 epoch 8 - iter 27/275 - loss 0.35232911 - time (sec): 0.42 - samples/sec: 5472.65 - lr: 0.000016 - momentum: 0.000000
2023-10-18 14:34:57,541 epoch 8 - iter 54/275 - loss 0.32320199 - time (sec): 0.82 - samples/sec: 5193.04 - lr: 0.000016 - momentum: 0.000000
2023-10-18 14:34:57,932 epoch 8 - iter 81/275 - loss 0.31388709 - time (sec): 1.21 - samples/sec: 5306.17 - lr: 0.000015 - momentum: 0.000000
2023-10-18 14:34:58,336 epoch 8 - iter 108/275 - loss 0.32588481 - time (sec): 1.62 - samples/sec: 5448.85 - lr: 0.000015 - momentum: 0.000000
2023-10-18 14:34:58,750 epoch 8 - iter 135/275 - loss 0.30666641 - time (sec): 2.03 - samples/sec: 5539.05 - lr: 0.000014 - momentum: 0.000000
2023-10-18 14:34:59,144 epoch 8 - iter 162/275 - loss 0.29869815 - time (sec): 2.43 - samples/sec: 5520.78 - lr: 0.000014 - momentum: 0.000000
2023-10-18 14:34:59,550 epoch 8 - iter 189/275 - loss 0.30212004 - time (sec): 2.83 - samples/sec: 5503.36 - lr: 0.000013 - momentum: 0.000000
2023-10-18 14:34:59,954 epoch 8 - iter 216/275 - loss 0.30866162 - time (sec): 3.24 - samples/sec: 5459.34 - lr: 0.000012 - momentum: 0.000000
2023-10-18 14:35:00,366 epoch 8 - iter 243/275 - loss 0.31213850 - time (sec): 3.65 - samples/sec: 5466.83 - lr: 0.000012 - momentum: 0.000000
2023-10-18 14:35:00,795 epoch 8 - iter 270/275 - loss 0.31265385 - time (sec): 4.08 - samples/sec: 5484.78 - lr: 0.000011 - momentum: 0.000000
2023-10-18 14:35:00,877 ----------------------------------------------------------------------------------------------------
2023-10-18 14:35:00,877 EPOCH 8 done: loss 0.3117 - lr: 0.000011
2023-10-18 14:35:01,258 DEV : loss 0.24906539916992188 - f1-score (micro avg) 0.6434
2023-10-18 14:35:01,263 saving best model
2023-10-18 14:35:01,299 ----------------------------------------------------------------------------------------------------
2023-10-18 14:35:01,725 epoch 9 - iter 27/275 - loss 0.28812589 - time (sec): 0.42 - samples/sec: 5407.92 - lr: 0.000011 - momentum: 0.000000
2023-10-18 14:35:02,140 epoch 9 - iter 54/275 - loss 0.30351124 - time (sec): 0.84 - samples/sec: 5479.49 - lr: 0.000010 - momentum: 0.000000
2023-10-18 14:35:02,566 epoch 9 - iter 81/275 - loss 0.30512938 - time (sec): 1.27 - samples/sec: 5373.65 - lr: 0.000010 - momentum: 0.000000
2023-10-18 14:35:02,965 epoch 9 - iter 108/275 - loss 0.32208951 - time (sec): 1.67 - samples/sec: 5342.06 - lr: 0.000009 - momentum: 0.000000
2023-10-18 14:35:03,367 epoch 9 - iter 135/275 - loss 0.32851618 - time (sec): 2.07 - samples/sec: 5341.07 - lr: 0.000009 - momentum: 0.000000
2023-10-18 14:35:03,774 epoch 9 - iter 162/275 - loss 0.32796046 - time (sec): 2.47 - samples/sec: 5362.39 - lr: 0.000008 - momentum: 0.000000
2023-10-18 14:35:04,187 epoch 9 - iter 189/275 - loss 0.31928635 - time (sec): 2.89 - samples/sec: 5363.87 - lr: 0.000007 - momentum: 0.000000
2023-10-18 14:35:04,598 epoch 9 - iter 216/275 - loss 0.31189310 - time (sec): 3.30 - samples/sec: 5409.81 - lr: 0.000007 - momentum: 0.000000
2023-10-18 14:35:05,011 epoch 9 - iter 243/275 - loss 0.31212107 - time (sec): 3.71 - samples/sec: 5502.41 - lr: 0.000006 - momentum: 0.000000
2023-10-18 14:35:05,409 epoch 9 - iter 270/275 - loss 0.30980213 - time (sec): 4.11 - samples/sec: 5439.67 - lr: 0.000006 - momentum: 0.000000
2023-10-18 14:35:05,483 ----------------------------------------------------------------------------------------------------
2023-10-18 14:35:05,483 EPOCH 9 done: loss 0.3114 - lr: 0.000006
2023-10-18 14:35:05,873 DEV : loss 0.24717433750629425 - f1-score (micro avg) 0.6471
2023-10-18 14:35:05,878 saving best model
2023-10-18 14:35:05,918 ----------------------------------------------------------------------------------------------------
2023-10-18 14:35:06,353 epoch 10 - iter 27/275 - loss 0.25692152 - time (sec): 0.43 - samples/sec: 5151.95 - lr: 0.000005 - momentum: 0.000000
2023-10-18 14:35:06,767 epoch 10 - iter 54/275 - loss 0.28814194 - time (sec): 0.85 - samples/sec: 5358.13 - lr: 0.000005 - momentum: 0.000000
2023-10-18 14:35:07,169 epoch 10 - iter 81/275 - loss 0.29696314 - time (sec): 1.25 - samples/sec: 5240.54 - lr: 0.000004 - momentum: 0.000000
2023-10-18 14:35:07,563 epoch 10 - iter 108/275 - loss 0.29047393 - time (sec): 1.64 - samples/sec: 5305.80 - lr: 0.000004 - momentum: 0.000000
2023-10-18 14:35:07,974 epoch 10 - iter 135/275 - loss 0.30111439 - time (sec): 2.06 - samples/sec: 5365.85 - lr: 0.000003 - momentum: 0.000000
2023-10-18 14:35:08,388 epoch 10 - iter 162/275 - loss 0.30748718 - time (sec): 2.47 - samples/sec: 5416.22 - lr: 0.000002 - momentum: 0.000000
2023-10-18 14:35:08,811 epoch 10 - iter 189/275 - loss 0.31512691 - time (sec): 2.89 - samples/sec: 5459.24 - lr: 0.000002 - momentum: 0.000000
2023-10-18 14:35:09,217 epoch 10 - iter 216/275 - loss 0.30589871 - time (sec): 3.30 - samples/sec: 5491.21 - lr: 0.000001 - momentum: 0.000000
2023-10-18 14:35:09,630 epoch 10 - iter 243/275 - loss 0.30215041 - time (sec): 3.71 - samples/sec: 5434.46 - lr: 0.000001 - momentum: 0.000000
2023-10-18 14:35:10,047 epoch 10 - iter 270/275 - loss 0.29793489 - time (sec): 4.13 - samples/sec: 5420.59 - lr: 0.000000 - momentum: 0.000000
2023-10-18 14:35:10,122 ----------------------------------------------------------------------------------------------------
2023-10-18 14:35:10,122 EPOCH 10 done: loss 0.2969 - lr: 0.000000
2023-10-18 14:35:10,495 DEV : loss 0.2449171245098114 - f1-score (micro avg) 0.644
2023-10-18 14:35:10,529 ----------------------------------------------------------------------------------------------------
2023-10-18 14:35:10,529 Loading model from best epoch ...
2023-10-18 14:35:10,603 SequenceTagger predicts: Dictionary with 25 tags: O, S-scope, B-scope, E-scope, I-scope, S-pers, B-pers, E-pers, I-pers, S-work, B-work, E-work, I-work, S-loc, B-loc, E-loc, I-loc, S-object, B-object, E-object, I-object, S-date, B-date, E-date, I-date
2023-10-18 14:35:10,884
Results:
- F-score (micro) 0.6606
- F-score (macro) 0.3933
- Accuracy 0.501
By class:
precision recall f1-score support
scope 0.6216 0.6534 0.6371 176
pers 0.8962 0.7422 0.8120 128
work 0.4583 0.5946 0.5176 74
object 0.0000 0.0000 0.0000 2
loc 0.0000 0.0000 0.0000 2
micro avg 0.6563 0.6649 0.6606 382
macro avg 0.3952 0.3980 0.3933 382
weighted avg 0.6755 0.6649 0.6659 382
2023-10-18 14:35:10,884 ----------------------------------------------------------------------------------------------------
|