File size: 24,146 Bytes
6deadec |
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 18:09:46,064 ----------------------------------------------------------------------------------------------------
2023-10-18 18:09:46,064 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=21, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-18 18:09:46,064 ----------------------------------------------------------------------------------------------------
2023-10-18 18:09:46,064 MultiCorpus: 3575 train + 1235 dev + 1266 test sentences
- NER_HIPE_2022 Corpus: 3575 train + 1235 dev + 1266 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/hipe2020/de/with_doc_seperator
2023-10-18 18:09:46,064 ----------------------------------------------------------------------------------------------------
2023-10-18 18:09:46,065 Train: 3575 sentences
2023-10-18 18:09:46,065 (train_with_dev=False, train_with_test=False)
2023-10-18 18:09:46,065 ----------------------------------------------------------------------------------------------------
2023-10-18 18:09:46,065 Training Params:
2023-10-18 18:09:46,065 - learning_rate: "5e-05"
2023-10-18 18:09:46,065 - mini_batch_size: "8"
2023-10-18 18:09:46,065 - max_epochs: "10"
2023-10-18 18:09:46,065 - shuffle: "True"
2023-10-18 18:09:46,065 ----------------------------------------------------------------------------------------------------
2023-10-18 18:09:46,065 Plugins:
2023-10-18 18:09:46,065 - TensorboardLogger
2023-10-18 18:09:46,065 - LinearScheduler | warmup_fraction: '0.1'
2023-10-18 18:09:46,065 ----------------------------------------------------------------------------------------------------
2023-10-18 18:09:46,065 Final evaluation on model from best epoch (best-model.pt)
2023-10-18 18:09:46,065 - metric: "('micro avg', 'f1-score')"
2023-10-18 18:09:46,065 ----------------------------------------------------------------------------------------------------
2023-10-18 18:09:46,065 Computation:
2023-10-18 18:09:46,065 - compute on device: cuda:0
2023-10-18 18:09:46,065 - embedding storage: none
2023-10-18 18:09:46,065 ----------------------------------------------------------------------------------------------------
2023-10-18 18:09:46,065 Model training base path: "hmbench-hipe2020/de-dbmdz/bert-tiny-historic-multilingual-cased-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-3"
2023-10-18 18:09:46,065 ----------------------------------------------------------------------------------------------------
2023-10-18 18:09:46,065 ----------------------------------------------------------------------------------------------------
2023-10-18 18:09:46,065 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-18 18:09:47,053 epoch 1 - iter 44/447 - loss 3.48107062 - time (sec): 0.99 - samples/sec: 8149.30 - lr: 0.000005 - momentum: 0.000000
2023-10-18 18:09:48,036 epoch 1 - iter 88/447 - loss 3.27194561 - time (sec): 1.97 - samples/sec: 8188.94 - lr: 0.000010 - momentum: 0.000000
2023-10-18 18:09:49,024 epoch 1 - iter 132/447 - loss 2.95857671 - time (sec): 2.96 - samples/sec: 8487.10 - lr: 0.000015 - momentum: 0.000000
2023-10-18 18:09:50,056 epoch 1 - iter 176/447 - loss 2.57672294 - time (sec): 3.99 - samples/sec: 8501.78 - lr: 0.000020 - momentum: 0.000000
2023-10-18 18:09:51,047 epoch 1 - iter 220/447 - loss 2.24069636 - time (sec): 4.98 - samples/sec: 8528.40 - lr: 0.000024 - momentum: 0.000000
2023-10-18 18:09:52,027 epoch 1 - iter 264/447 - loss 1.99699733 - time (sec): 5.96 - samples/sec: 8479.72 - lr: 0.000029 - momentum: 0.000000
2023-10-18 18:09:52,989 epoch 1 - iter 308/447 - loss 1.80609338 - time (sec): 6.92 - samples/sec: 8448.04 - lr: 0.000034 - momentum: 0.000000
2023-10-18 18:09:53,982 epoch 1 - iter 352/447 - loss 1.65672493 - time (sec): 7.92 - samples/sec: 8441.08 - lr: 0.000039 - momentum: 0.000000
2023-10-18 18:09:55,020 epoch 1 - iter 396/447 - loss 1.51394066 - time (sec): 8.95 - samples/sec: 8605.93 - lr: 0.000044 - momentum: 0.000000
2023-10-18 18:09:56,001 epoch 1 - iter 440/447 - loss 1.42253798 - time (sec): 9.94 - samples/sec: 8579.71 - lr: 0.000049 - momentum: 0.000000
2023-10-18 18:09:56,152 ----------------------------------------------------------------------------------------------------
2023-10-18 18:09:56,152 EPOCH 1 done: loss 1.4116 - lr: 0.000049
2023-10-18 18:09:58,311 DEV : loss 0.440503865480423 - f1-score (micro avg) 0.0
2023-10-18 18:09:58,335 ----------------------------------------------------------------------------------------------------
2023-10-18 18:09:59,384 epoch 2 - iter 44/447 - loss 0.55508516 - time (sec): 1.05 - samples/sec: 8395.32 - lr: 0.000049 - momentum: 0.000000
2023-10-18 18:10:00,396 epoch 2 - iter 88/447 - loss 0.52727644 - time (sec): 2.06 - samples/sec: 8360.90 - lr: 0.000049 - momentum: 0.000000
2023-10-18 18:10:01,408 epoch 2 - iter 132/447 - loss 0.52183207 - time (sec): 3.07 - samples/sec: 8233.32 - lr: 0.000048 - momentum: 0.000000
2023-10-18 18:10:02,425 epoch 2 - iter 176/447 - loss 0.52157574 - time (sec): 4.09 - samples/sec: 8126.32 - lr: 0.000048 - momentum: 0.000000
2023-10-18 18:10:03,453 epoch 2 - iter 220/447 - loss 0.51770155 - time (sec): 5.12 - samples/sec: 8137.91 - lr: 0.000047 - momentum: 0.000000
2023-10-18 18:10:04,457 epoch 2 - iter 264/447 - loss 0.50585354 - time (sec): 6.12 - samples/sec: 8201.03 - lr: 0.000047 - momentum: 0.000000
2023-10-18 18:10:05,438 epoch 2 - iter 308/447 - loss 0.50399864 - time (sec): 7.10 - samples/sec: 8201.77 - lr: 0.000046 - momentum: 0.000000
2023-10-18 18:10:06,437 epoch 2 - iter 352/447 - loss 0.49361795 - time (sec): 8.10 - samples/sec: 8250.70 - lr: 0.000046 - momentum: 0.000000
2023-10-18 18:10:07,518 epoch 2 - iter 396/447 - loss 0.48893978 - time (sec): 9.18 - samples/sec: 8357.10 - lr: 0.000045 - momentum: 0.000000
2023-10-18 18:10:08,526 epoch 2 - iter 440/447 - loss 0.48392438 - time (sec): 10.19 - samples/sec: 8352.26 - lr: 0.000045 - momentum: 0.000000
2023-10-18 18:10:08,682 ----------------------------------------------------------------------------------------------------
2023-10-18 18:10:08,683 EPOCH 2 done: loss 0.4823 - lr: 0.000045
2023-10-18 18:10:13,841 DEV : loss 0.34371984004974365 - f1-score (micro avg) 0.113
2023-10-18 18:10:13,866 saving best model
2023-10-18 18:10:13,903 ----------------------------------------------------------------------------------------------------
2023-10-18 18:10:15,010 epoch 3 - iter 44/447 - loss 0.43301365 - time (sec): 1.11 - samples/sec: 7369.43 - lr: 0.000044 - momentum: 0.000000
2023-10-18 18:10:16,142 epoch 3 - iter 88/447 - loss 0.44758670 - time (sec): 2.24 - samples/sec: 7560.07 - lr: 0.000043 - momentum: 0.000000
2023-10-18 18:10:17,243 epoch 3 - iter 132/447 - loss 0.42991796 - time (sec): 3.34 - samples/sec: 7834.85 - lr: 0.000043 - momentum: 0.000000
2023-10-18 18:10:18,256 epoch 3 - iter 176/447 - loss 0.43302866 - time (sec): 4.35 - samples/sec: 8004.92 - lr: 0.000042 - momentum: 0.000000
2023-10-18 18:10:19,252 epoch 3 - iter 220/447 - loss 0.42209371 - time (sec): 5.35 - samples/sec: 8108.14 - lr: 0.000042 - momentum: 0.000000
2023-10-18 18:10:20,220 epoch 3 - iter 264/447 - loss 0.40578864 - time (sec): 6.32 - samples/sec: 8228.82 - lr: 0.000041 - momentum: 0.000000
2023-10-18 18:10:21,223 epoch 3 - iter 308/447 - loss 0.40322907 - time (sec): 7.32 - samples/sec: 8256.58 - lr: 0.000041 - momentum: 0.000000
2023-10-18 18:10:22,195 epoch 3 - iter 352/447 - loss 0.39866877 - time (sec): 8.29 - samples/sec: 8245.34 - lr: 0.000040 - momentum: 0.000000
2023-10-18 18:10:23,204 epoch 3 - iter 396/447 - loss 0.39598782 - time (sec): 9.30 - samples/sec: 8291.45 - lr: 0.000040 - momentum: 0.000000
2023-10-18 18:10:24,193 epoch 3 - iter 440/447 - loss 0.39667531 - time (sec): 10.29 - samples/sec: 8277.73 - lr: 0.000039 - momentum: 0.000000
2023-10-18 18:10:24,350 ----------------------------------------------------------------------------------------------------
2023-10-18 18:10:24,350 EPOCH 3 done: loss 0.3956 - lr: 0.000039
2023-10-18 18:10:29,527 DEV : loss 0.32860174775123596 - f1-score (micro avg) 0.285
2023-10-18 18:10:29,551 saving best model
2023-10-18 18:10:29,586 ----------------------------------------------------------------------------------------------------
2023-10-18 18:10:30,520 epoch 4 - iter 44/447 - loss 0.36943001 - time (sec): 0.93 - samples/sec: 9106.31 - lr: 0.000038 - momentum: 0.000000
2023-10-18 18:10:31,531 epoch 4 - iter 88/447 - loss 0.38745357 - time (sec): 1.94 - samples/sec: 8838.65 - lr: 0.000038 - momentum: 0.000000
2023-10-18 18:10:32,506 epoch 4 - iter 132/447 - loss 0.38115552 - time (sec): 2.92 - samples/sec: 8839.02 - lr: 0.000037 - momentum: 0.000000
2023-10-18 18:10:33,512 epoch 4 - iter 176/447 - loss 0.37121101 - time (sec): 3.93 - samples/sec: 8895.14 - lr: 0.000037 - momentum: 0.000000
2023-10-18 18:10:34,550 epoch 4 - iter 220/447 - loss 0.36960639 - time (sec): 4.96 - samples/sec: 8984.03 - lr: 0.000036 - momentum: 0.000000
2023-10-18 18:10:35,551 epoch 4 - iter 264/447 - loss 0.36995927 - time (sec): 5.96 - samples/sec: 8809.47 - lr: 0.000036 - momentum: 0.000000
2023-10-18 18:10:36,550 epoch 4 - iter 308/447 - loss 0.36353926 - time (sec): 6.96 - samples/sec: 8733.96 - lr: 0.000035 - momentum: 0.000000
2023-10-18 18:10:37,569 epoch 4 - iter 352/447 - loss 0.36657074 - time (sec): 7.98 - samples/sec: 8625.94 - lr: 0.000035 - momentum: 0.000000
2023-10-18 18:10:38,546 epoch 4 - iter 396/447 - loss 0.36628777 - time (sec): 8.96 - samples/sec: 8561.72 - lr: 0.000034 - momentum: 0.000000
2023-10-18 18:10:39,540 epoch 4 - iter 440/447 - loss 0.36059543 - time (sec): 9.95 - samples/sec: 8580.60 - lr: 0.000033 - momentum: 0.000000
2023-10-18 18:10:39,696 ----------------------------------------------------------------------------------------------------
2023-10-18 18:10:39,696 EPOCH 4 done: loss 0.3608 - lr: 0.000033
2023-10-18 18:10:44,932 DEV : loss 0.320486456155777 - f1-score (micro avg) 0.3043
2023-10-18 18:10:44,956 saving best model
2023-10-18 18:10:44,995 ----------------------------------------------------------------------------------------------------
2023-10-18 18:10:46,000 epoch 5 - iter 44/447 - loss 0.32352808 - time (sec): 1.00 - samples/sec: 8351.93 - lr: 0.000033 - momentum: 0.000000
2023-10-18 18:10:47,066 epoch 5 - iter 88/447 - loss 0.32347254 - time (sec): 2.07 - samples/sec: 8722.05 - lr: 0.000032 - momentum: 0.000000
2023-10-18 18:10:48,107 epoch 5 - iter 132/447 - loss 0.33387921 - time (sec): 3.11 - samples/sec: 8659.51 - lr: 0.000032 - momentum: 0.000000
2023-10-18 18:10:49,084 epoch 5 - iter 176/447 - loss 0.32884812 - time (sec): 4.09 - samples/sec: 8638.09 - lr: 0.000031 - momentum: 0.000000
2023-10-18 18:10:50,071 epoch 5 - iter 220/447 - loss 0.33619392 - time (sec): 5.07 - samples/sec: 8511.54 - lr: 0.000031 - momentum: 0.000000
2023-10-18 18:10:51,057 epoch 5 - iter 264/447 - loss 0.33428705 - time (sec): 6.06 - samples/sec: 8467.93 - lr: 0.000030 - momentum: 0.000000
2023-10-18 18:10:52,045 epoch 5 - iter 308/447 - loss 0.33761251 - time (sec): 7.05 - samples/sec: 8500.30 - lr: 0.000030 - momentum: 0.000000
2023-10-18 18:10:53,101 epoch 5 - iter 352/447 - loss 0.33546904 - time (sec): 8.10 - samples/sec: 8513.28 - lr: 0.000029 - momentum: 0.000000
2023-10-18 18:10:54,129 epoch 5 - iter 396/447 - loss 0.33425057 - time (sec): 9.13 - samples/sec: 8480.39 - lr: 0.000028 - momentum: 0.000000
2023-10-18 18:10:55,110 epoch 5 - iter 440/447 - loss 0.33363531 - time (sec): 10.11 - samples/sec: 8441.10 - lr: 0.000028 - momentum: 0.000000
2023-10-18 18:10:55,257 ----------------------------------------------------------------------------------------------------
2023-10-18 18:10:55,257 EPOCH 5 done: loss 0.3316 - lr: 0.000028
2023-10-18 18:11:00,464 DEV : loss 0.3082502484321594 - f1-score (micro avg) 0.3129
2023-10-18 18:11:00,488 saving best model
2023-10-18 18:11:00,526 ----------------------------------------------------------------------------------------------------
2023-10-18 18:11:01,586 epoch 6 - iter 44/447 - loss 0.27288611 - time (sec): 1.06 - samples/sec: 8701.04 - lr: 0.000027 - momentum: 0.000000
2023-10-18 18:11:02,564 epoch 6 - iter 88/447 - loss 0.30392757 - time (sec): 2.04 - samples/sec: 8602.53 - lr: 0.000027 - momentum: 0.000000
2023-10-18 18:11:03,489 epoch 6 - iter 132/447 - loss 0.32248259 - time (sec): 2.96 - samples/sec: 8719.10 - lr: 0.000026 - momentum: 0.000000
2023-10-18 18:11:04,445 epoch 6 - iter 176/447 - loss 0.32428223 - time (sec): 3.92 - samples/sec: 8749.25 - lr: 0.000026 - momentum: 0.000000
2023-10-18 18:11:05,456 epoch 6 - iter 220/447 - loss 0.32839831 - time (sec): 4.93 - samples/sec: 8750.03 - lr: 0.000025 - momentum: 0.000000
2023-10-18 18:11:06,438 epoch 6 - iter 264/447 - loss 0.32721286 - time (sec): 5.91 - samples/sec: 8725.73 - lr: 0.000025 - momentum: 0.000000
2023-10-18 18:11:07,434 epoch 6 - iter 308/447 - loss 0.31843729 - time (sec): 6.91 - samples/sec: 8635.69 - lr: 0.000024 - momentum: 0.000000
2023-10-18 18:11:08,439 epoch 6 - iter 352/447 - loss 0.31455378 - time (sec): 7.91 - samples/sec: 8592.91 - lr: 0.000023 - momentum: 0.000000
2023-10-18 18:11:09,472 epoch 6 - iter 396/447 - loss 0.30630728 - time (sec): 8.95 - samples/sec: 8597.27 - lr: 0.000023 - momentum: 0.000000
2023-10-18 18:11:10,441 epoch 6 - iter 440/447 - loss 0.30982808 - time (sec): 9.91 - samples/sec: 8575.59 - lr: 0.000022 - momentum: 0.000000
2023-10-18 18:11:10,601 ----------------------------------------------------------------------------------------------------
2023-10-18 18:11:10,601 EPOCH 6 done: loss 0.3103 - lr: 0.000022
2023-10-18 18:11:15,515 DEV : loss 0.30133387446403503 - f1-score (micro avg) 0.3376
2023-10-18 18:11:15,539 saving best model
2023-10-18 18:11:15,573 ----------------------------------------------------------------------------------------------------
2023-10-18 18:11:16,597 epoch 7 - iter 44/447 - loss 0.25287805 - time (sec): 1.02 - samples/sec: 7909.44 - lr: 0.000022 - momentum: 0.000000
2023-10-18 18:11:17,643 epoch 7 - iter 88/447 - loss 0.28977731 - time (sec): 2.07 - samples/sec: 8401.57 - lr: 0.000021 - momentum: 0.000000
2023-10-18 18:11:18,644 epoch 7 - iter 132/447 - loss 0.28988167 - time (sec): 3.07 - samples/sec: 8379.93 - lr: 0.000021 - momentum: 0.000000
2023-10-18 18:11:19,673 epoch 7 - iter 176/447 - loss 0.29009755 - time (sec): 4.10 - samples/sec: 8671.40 - lr: 0.000020 - momentum: 0.000000
2023-10-18 18:11:20,674 epoch 7 - iter 220/447 - loss 0.29294307 - time (sec): 5.10 - samples/sec: 8638.88 - lr: 0.000020 - momentum: 0.000000
2023-10-18 18:11:22,035 epoch 7 - iter 264/447 - loss 0.29453891 - time (sec): 6.46 - samples/sec: 8222.76 - lr: 0.000019 - momentum: 0.000000
2023-10-18 18:11:23,013 epoch 7 - iter 308/447 - loss 0.29521798 - time (sec): 7.44 - samples/sec: 8191.12 - lr: 0.000018 - momentum: 0.000000
2023-10-18 18:11:24,050 epoch 7 - iter 352/447 - loss 0.29903791 - time (sec): 8.48 - samples/sec: 8180.70 - lr: 0.000018 - momentum: 0.000000
2023-10-18 18:11:25,047 epoch 7 - iter 396/447 - loss 0.30193005 - time (sec): 9.47 - samples/sec: 8173.12 - lr: 0.000017 - momentum: 0.000000
2023-10-18 18:11:26,053 epoch 7 - iter 440/447 - loss 0.29772260 - time (sec): 10.48 - samples/sec: 8138.11 - lr: 0.000017 - momentum: 0.000000
2023-10-18 18:11:26,218 ----------------------------------------------------------------------------------------------------
2023-10-18 18:11:26,218 EPOCH 7 done: loss 0.2996 - lr: 0.000017
2023-10-18 18:11:31,159 DEV : loss 0.2995185852050781 - f1-score (micro avg) 0.3428
2023-10-18 18:11:31,184 saving best model
2023-10-18 18:11:31,223 ----------------------------------------------------------------------------------------------------
2023-10-18 18:11:32,193 epoch 8 - iter 44/447 - loss 0.27978168 - time (sec): 0.97 - samples/sec: 8119.90 - lr: 0.000016 - momentum: 0.000000
2023-10-18 18:11:33,151 epoch 8 - iter 88/447 - loss 0.28824429 - time (sec): 1.93 - samples/sec: 7853.50 - lr: 0.000016 - momentum: 0.000000
2023-10-18 18:11:34,148 epoch 8 - iter 132/447 - loss 0.28662310 - time (sec): 2.92 - samples/sec: 8149.03 - lr: 0.000015 - momentum: 0.000000
2023-10-18 18:11:35,134 epoch 8 - iter 176/447 - loss 0.29888598 - time (sec): 3.91 - samples/sec: 8208.25 - lr: 0.000015 - momentum: 0.000000
2023-10-18 18:11:36,084 epoch 8 - iter 220/447 - loss 0.29137520 - time (sec): 4.86 - samples/sec: 8235.72 - lr: 0.000014 - momentum: 0.000000
2023-10-18 18:11:37,086 epoch 8 - iter 264/447 - loss 0.28874586 - time (sec): 5.86 - samples/sec: 8191.30 - lr: 0.000013 - momentum: 0.000000
2023-10-18 18:11:38,115 epoch 8 - iter 308/447 - loss 0.28559013 - time (sec): 6.89 - samples/sec: 8345.14 - lr: 0.000013 - momentum: 0.000000
2023-10-18 18:11:39,161 epoch 8 - iter 352/447 - loss 0.29125430 - time (sec): 7.94 - samples/sec: 8419.16 - lr: 0.000012 - momentum: 0.000000
2023-10-18 18:11:40,002 epoch 8 - iter 396/447 - loss 0.29049800 - time (sec): 8.78 - samples/sec: 8510.69 - lr: 0.000012 - momentum: 0.000000
2023-10-18 18:11:40,885 epoch 8 - iter 440/447 - loss 0.28778174 - time (sec): 9.66 - samples/sec: 8688.77 - lr: 0.000011 - momentum: 0.000000
2023-10-18 18:11:41,101 ----------------------------------------------------------------------------------------------------
2023-10-18 18:11:41,101 EPOCH 8 done: loss 0.2870 - lr: 0.000011
2023-10-18 18:11:46,365 DEV : loss 0.29179754853248596 - f1-score (micro avg) 0.3475
2023-10-18 18:11:46,390 saving best model
2023-10-18 18:11:46,425 ----------------------------------------------------------------------------------------------------
2023-10-18 18:11:47,450 epoch 9 - iter 44/447 - loss 0.30150118 - time (sec): 1.02 - samples/sec: 9520.13 - lr: 0.000011 - momentum: 0.000000
2023-10-18 18:11:48,460 epoch 9 - iter 88/447 - loss 0.30631108 - time (sec): 2.04 - samples/sec: 8952.24 - lr: 0.000010 - momentum: 0.000000
2023-10-18 18:11:49,498 epoch 9 - iter 132/447 - loss 0.29415977 - time (sec): 3.07 - samples/sec: 8877.64 - lr: 0.000010 - momentum: 0.000000
2023-10-18 18:11:50,475 epoch 9 - iter 176/447 - loss 0.29426276 - time (sec): 4.05 - samples/sec: 8537.82 - lr: 0.000009 - momentum: 0.000000
2023-10-18 18:11:51,454 epoch 9 - iter 220/447 - loss 0.27981800 - time (sec): 5.03 - samples/sec: 8555.77 - lr: 0.000008 - momentum: 0.000000
2023-10-18 18:11:52,451 epoch 9 - iter 264/447 - loss 0.28872070 - time (sec): 6.03 - samples/sec: 8511.27 - lr: 0.000008 - momentum: 0.000000
2023-10-18 18:11:53,450 epoch 9 - iter 308/447 - loss 0.28551834 - time (sec): 7.03 - samples/sec: 8505.13 - lr: 0.000007 - momentum: 0.000000
2023-10-18 18:11:54,461 epoch 9 - iter 352/447 - loss 0.28086491 - time (sec): 8.04 - samples/sec: 8503.20 - lr: 0.000007 - momentum: 0.000000
2023-10-18 18:11:55,447 epoch 9 - iter 396/447 - loss 0.27926325 - time (sec): 9.02 - samples/sec: 8513.26 - lr: 0.000006 - momentum: 0.000000
2023-10-18 18:11:56,409 epoch 9 - iter 440/447 - loss 0.27899408 - time (sec): 9.98 - samples/sec: 8476.82 - lr: 0.000006 - momentum: 0.000000
2023-10-18 18:11:56,600 ----------------------------------------------------------------------------------------------------
2023-10-18 18:11:56,600 EPOCH 9 done: loss 0.2788 - lr: 0.000006
2023-10-18 18:12:01,866 DEV : loss 0.29360440373420715 - f1-score (micro avg) 0.3437
2023-10-18 18:12:01,891 ----------------------------------------------------------------------------------------------------
2023-10-18 18:12:02,845 epoch 10 - iter 44/447 - loss 0.29982169 - time (sec): 0.95 - samples/sec: 9260.58 - lr: 0.000005 - momentum: 0.000000
2023-10-18 18:12:03,842 epoch 10 - iter 88/447 - loss 0.29743769 - time (sec): 1.95 - samples/sec: 8975.17 - lr: 0.000005 - momentum: 0.000000
2023-10-18 18:12:04,837 epoch 10 - iter 132/447 - loss 0.28765785 - time (sec): 2.95 - samples/sec: 8664.41 - lr: 0.000004 - momentum: 0.000000
2023-10-18 18:12:05,825 epoch 10 - iter 176/447 - loss 0.29160770 - time (sec): 3.93 - samples/sec: 8453.40 - lr: 0.000003 - momentum: 0.000000
2023-10-18 18:12:06,779 epoch 10 - iter 220/447 - loss 0.28993886 - time (sec): 4.89 - samples/sec: 8449.64 - lr: 0.000003 - momentum: 0.000000
2023-10-18 18:12:07,819 epoch 10 - iter 264/447 - loss 0.28561611 - time (sec): 5.93 - samples/sec: 8473.17 - lr: 0.000002 - momentum: 0.000000
2023-10-18 18:12:08,861 epoch 10 - iter 308/447 - loss 0.27784141 - time (sec): 6.97 - samples/sec: 8501.64 - lr: 0.000002 - momentum: 0.000000
2023-10-18 18:12:09,891 epoch 10 - iter 352/447 - loss 0.27407641 - time (sec): 8.00 - samples/sec: 8474.39 - lr: 0.000001 - momentum: 0.000000
2023-10-18 18:12:10,888 epoch 10 - iter 396/447 - loss 0.27298519 - time (sec): 9.00 - samples/sec: 8489.94 - lr: 0.000001 - momentum: 0.000000
2023-10-18 18:12:11,937 epoch 10 - iter 440/447 - loss 0.27094189 - time (sec): 10.05 - samples/sec: 8479.74 - lr: 0.000000 - momentum: 0.000000
2023-10-18 18:12:12,101 ----------------------------------------------------------------------------------------------------
2023-10-18 18:12:12,101 EPOCH 10 done: loss 0.2713 - lr: 0.000000
2023-10-18 18:12:17,372 DEV : loss 0.2932772934436798 - f1-score (micro avg) 0.3486
2023-10-18 18:12:17,396 saving best model
2023-10-18 18:12:17,467 ----------------------------------------------------------------------------------------------------
2023-10-18 18:12:17,467 Loading model from best epoch ...
2023-10-18 18:12:17,542 SequenceTagger predicts: Dictionary with 21 tags: O, S-loc, B-loc, E-loc, I-loc, S-pers, B-pers, E-pers, I-pers, S-org, B-org, E-org, I-org, S-prod, B-prod, E-prod, I-prod, S-time, B-time, E-time, I-time
2023-10-18 18:12:19,483
Results:
- F-score (micro) 0.3408
- F-score (macro) 0.1668
- Accuracy 0.216
By class:
precision recall f1-score support
loc 0.4727 0.5369 0.5027 596
pers 0.1563 0.1892 0.1712 333
org 0.0000 0.0000 0.0000 132
time 0.2308 0.1224 0.1600 49
prod 0.0000 0.0000 0.0000 66
micro avg 0.3514 0.3308 0.3408 1176
macro avg 0.1720 0.1697 0.1668 1176
weighted avg 0.2934 0.3308 0.3099 1176
2023-10-18 18:12:19,483 ----------------------------------------------------------------------------------------------------
|