File size: 23,951 Bytes
d8ab410 |
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 |
2023-10-13 08:45:51,247 ----------------------------------------------------------------------------------------------------
2023-10-13 08:45:51,248 Model: "SequenceTagger(
(embeddings): TransformerWordEmbeddings(
(model): BertModel(
(embeddings): BertEmbeddings(
(word_embeddings): Embedding(32001, 768)
(position_embeddings): Embedding(512, 768)
(token_type_embeddings): Embedding(2, 768)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(encoder): BertEncoder(
(layer): ModuleList(
(0-11): 12 x BertLayer(
(attention): BertAttention(
(self): BertSelfAttention(
(query): Linear(in_features=768, out_features=768, bias=True)
(key): Linear(in_features=768, out_features=768, bias=True)
(value): Linear(in_features=768, out_features=768, bias=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(output): BertSelfOutput(
(dense): Linear(in_features=768, out_features=768, bias=True)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
(intermediate): BertIntermediate(
(dense): Linear(in_features=768, out_features=3072, bias=True)
(intermediate_act_fn): GELUActivation()
)
(output): BertOutput(
(dense): Linear(in_features=3072, out_features=768, bias=True)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
)
(pooler): BertPooler(
(dense): Linear(in_features=768, out_features=768, bias=True)
(activation): Tanh()
)
)
)
(locked_dropout): LockedDropout(p=0.5)
(linear): Linear(in_features=768, out_features=25, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-13 08:45:51,248 ----------------------------------------------------------------------------------------------------
2023-10-13 08:45:51,248 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
2023-10-13 08:45:51,248 ----------------------------------------------------------------------------------------------------
2023-10-13 08:45:51,248 Train: 1100 sentences
2023-10-13 08:45:51,248 (train_with_dev=False, train_with_test=False)
2023-10-13 08:45:51,248 ----------------------------------------------------------------------------------------------------
2023-10-13 08:45:51,248 Training Params:
2023-10-13 08:45:51,248 - learning_rate: "3e-05"
2023-10-13 08:45:51,248 - mini_batch_size: "8"
2023-10-13 08:45:51,248 - max_epochs: "10"
2023-10-13 08:45:51,248 - shuffle: "True"
2023-10-13 08:45:51,248 ----------------------------------------------------------------------------------------------------
2023-10-13 08:45:51,248 Plugins:
2023-10-13 08:45:51,248 - LinearScheduler | warmup_fraction: '0.1'
2023-10-13 08:45:51,248 ----------------------------------------------------------------------------------------------------
2023-10-13 08:45:51,248 Final evaluation on model from best epoch (best-model.pt)
2023-10-13 08:45:51,248 - metric: "('micro avg', 'f1-score')"
2023-10-13 08:45:51,248 ----------------------------------------------------------------------------------------------------
2023-10-13 08:45:51,248 Computation:
2023-10-13 08:45:51,249 - compute on device: cuda:0
2023-10-13 08:45:51,249 - embedding storage: none
2023-10-13 08:45:51,249 ----------------------------------------------------------------------------------------------------
2023-10-13 08:45:51,249 Model training base path: "hmbench-ajmc/de-dbmdz/bert-base-historic-multilingual-cased-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-4"
2023-10-13 08:45:51,249 ----------------------------------------------------------------------------------------------------
2023-10-13 08:45:51,249 ----------------------------------------------------------------------------------------------------
2023-10-13 08:45:52,026 epoch 1 - iter 13/138 - loss 3.42478686 - time (sec): 0.78 - samples/sec: 2613.29 - lr: 0.000003 - momentum: 0.000000
2023-10-13 08:45:52,831 epoch 1 - iter 26/138 - loss 3.25697591 - time (sec): 1.58 - samples/sec: 2702.85 - lr: 0.000005 - momentum: 0.000000
2023-10-13 08:45:53,633 epoch 1 - iter 39/138 - loss 2.92896232 - time (sec): 2.38 - samples/sec: 2678.31 - lr: 0.000008 - momentum: 0.000000
2023-10-13 08:45:54,404 epoch 1 - iter 52/138 - loss 2.46822794 - time (sec): 3.15 - samples/sec: 2690.25 - lr: 0.000011 - momentum: 0.000000
2023-10-13 08:45:55,222 epoch 1 - iter 65/138 - loss 2.15028122 - time (sec): 3.97 - samples/sec: 2689.10 - lr: 0.000014 - momentum: 0.000000
2023-10-13 08:45:55,992 epoch 1 - iter 78/138 - loss 1.93536841 - time (sec): 4.74 - samples/sec: 2693.64 - lr: 0.000017 - momentum: 0.000000
2023-10-13 08:45:56,792 epoch 1 - iter 91/138 - loss 1.74412999 - time (sec): 5.54 - samples/sec: 2732.53 - lr: 0.000020 - momentum: 0.000000
2023-10-13 08:45:57,576 epoch 1 - iter 104/138 - loss 1.58222199 - time (sec): 6.33 - samples/sec: 2745.11 - lr: 0.000022 - momentum: 0.000000
2023-10-13 08:45:58,334 epoch 1 - iter 117/138 - loss 1.45813361 - time (sec): 7.08 - samples/sec: 2741.53 - lr: 0.000025 - momentum: 0.000000
2023-10-13 08:45:59,090 epoch 1 - iter 130/138 - loss 1.35021726 - time (sec): 7.84 - samples/sec: 2749.77 - lr: 0.000028 - momentum: 0.000000
2023-10-13 08:45:59,560 ----------------------------------------------------------------------------------------------------
2023-10-13 08:45:59,560 EPOCH 1 done: loss 1.2954 - lr: 0.000028
2023-10-13 08:46:00,307 DEV : loss 0.3012372851371765 - f1-score (micro avg) 0.6066
2023-10-13 08:46:00,313 saving best model
2023-10-13 08:46:00,682 ----------------------------------------------------------------------------------------------------
2023-10-13 08:46:01,409 epoch 2 - iter 13/138 - loss 0.32651210 - time (sec): 0.73 - samples/sec: 2747.35 - lr: 0.000030 - momentum: 0.000000
2023-10-13 08:46:02,186 epoch 2 - iter 26/138 - loss 0.27645443 - time (sec): 1.50 - samples/sec: 2795.41 - lr: 0.000029 - momentum: 0.000000
2023-10-13 08:46:02,955 epoch 2 - iter 39/138 - loss 0.27978688 - time (sec): 2.27 - samples/sec: 2768.62 - lr: 0.000029 - momentum: 0.000000
2023-10-13 08:46:03,754 epoch 2 - iter 52/138 - loss 0.26376866 - time (sec): 3.07 - samples/sec: 2769.56 - lr: 0.000029 - momentum: 0.000000
2023-10-13 08:46:04,518 epoch 2 - iter 65/138 - loss 0.24497349 - time (sec): 3.84 - samples/sec: 2789.53 - lr: 0.000028 - momentum: 0.000000
2023-10-13 08:46:05,305 epoch 2 - iter 78/138 - loss 0.24382508 - time (sec): 4.62 - samples/sec: 2805.76 - lr: 0.000028 - momentum: 0.000000
2023-10-13 08:46:06,080 epoch 2 - iter 91/138 - loss 0.23545494 - time (sec): 5.40 - samples/sec: 2803.92 - lr: 0.000028 - momentum: 0.000000
2023-10-13 08:46:06,876 epoch 2 - iter 104/138 - loss 0.22685556 - time (sec): 6.19 - samples/sec: 2806.32 - lr: 0.000028 - momentum: 0.000000
2023-10-13 08:46:07,655 epoch 2 - iter 117/138 - loss 0.22301589 - time (sec): 6.97 - samples/sec: 2803.41 - lr: 0.000027 - momentum: 0.000000
2023-10-13 08:46:08,451 epoch 2 - iter 130/138 - loss 0.21917924 - time (sec): 7.77 - samples/sec: 2775.45 - lr: 0.000027 - momentum: 0.000000
2023-10-13 08:46:08,931 ----------------------------------------------------------------------------------------------------
2023-10-13 08:46:08,932 EPOCH 2 done: loss 0.2203 - lr: 0.000027
2023-10-13 08:46:09,600 DEV : loss 0.15246804058551788 - f1-score (micro avg) 0.7931
2023-10-13 08:46:09,606 saving best model
2023-10-13 08:46:10,072 ----------------------------------------------------------------------------------------------------
2023-10-13 08:46:10,838 epoch 3 - iter 13/138 - loss 0.11403792 - time (sec): 0.76 - samples/sec: 2885.88 - lr: 0.000026 - momentum: 0.000000
2023-10-13 08:46:11,686 epoch 3 - iter 26/138 - loss 0.12036127 - time (sec): 1.61 - samples/sec: 2709.68 - lr: 0.000026 - momentum: 0.000000
2023-10-13 08:46:12,500 epoch 3 - iter 39/138 - loss 0.12816251 - time (sec): 2.43 - samples/sec: 2650.50 - lr: 0.000026 - momentum: 0.000000
2023-10-13 08:46:13,344 epoch 3 - iter 52/138 - loss 0.12673381 - time (sec): 3.27 - samples/sec: 2628.56 - lr: 0.000025 - momentum: 0.000000
2023-10-13 08:46:14,126 epoch 3 - iter 65/138 - loss 0.12713805 - time (sec): 4.05 - samples/sec: 2676.32 - lr: 0.000025 - momentum: 0.000000
2023-10-13 08:46:14,877 epoch 3 - iter 78/138 - loss 0.12679176 - time (sec): 4.80 - samples/sec: 2701.29 - lr: 0.000025 - momentum: 0.000000
2023-10-13 08:46:15,698 epoch 3 - iter 91/138 - loss 0.12344778 - time (sec): 5.62 - samples/sec: 2721.66 - lr: 0.000025 - momentum: 0.000000
2023-10-13 08:46:16,447 epoch 3 - iter 104/138 - loss 0.11980624 - time (sec): 6.37 - samples/sec: 2747.00 - lr: 0.000024 - momentum: 0.000000
2023-10-13 08:46:17,198 epoch 3 - iter 117/138 - loss 0.11466405 - time (sec): 7.12 - samples/sec: 2750.58 - lr: 0.000024 - momentum: 0.000000
2023-10-13 08:46:17,981 epoch 3 - iter 130/138 - loss 0.11116083 - time (sec): 7.91 - samples/sec: 2756.19 - lr: 0.000024 - momentum: 0.000000
2023-10-13 08:46:18,414 ----------------------------------------------------------------------------------------------------
2023-10-13 08:46:18,414 EPOCH 3 done: loss 0.1131 - lr: 0.000024
2023-10-13 08:46:19,095 DEV : loss 0.12917353212833405 - f1-score (micro avg) 0.8363
2023-10-13 08:46:19,100 saving best model
2023-10-13 08:46:19,550 ----------------------------------------------------------------------------------------------------
2023-10-13 08:46:20,337 epoch 4 - iter 13/138 - loss 0.05993807 - time (sec): 0.78 - samples/sec: 2718.93 - lr: 0.000023 - momentum: 0.000000
2023-10-13 08:46:21,099 epoch 4 - iter 26/138 - loss 0.06306592 - time (sec): 1.55 - samples/sec: 2684.08 - lr: 0.000023 - momentum: 0.000000
2023-10-13 08:46:21,878 epoch 4 - iter 39/138 - loss 0.07823448 - time (sec): 2.32 - samples/sec: 2666.72 - lr: 0.000022 - momentum: 0.000000
2023-10-13 08:46:22,683 epoch 4 - iter 52/138 - loss 0.08537937 - time (sec): 3.13 - samples/sec: 2606.09 - lr: 0.000022 - momentum: 0.000000
2023-10-13 08:46:23,444 epoch 4 - iter 65/138 - loss 0.08099344 - time (sec): 3.89 - samples/sec: 2616.84 - lr: 0.000022 - momentum: 0.000000
2023-10-13 08:46:24,243 epoch 4 - iter 78/138 - loss 0.07823728 - time (sec): 4.69 - samples/sec: 2670.18 - lr: 0.000021 - momentum: 0.000000
2023-10-13 08:46:24,958 epoch 4 - iter 91/138 - loss 0.07524355 - time (sec): 5.40 - samples/sec: 2655.08 - lr: 0.000021 - momentum: 0.000000
2023-10-13 08:46:25,662 epoch 4 - iter 104/138 - loss 0.07383065 - time (sec): 6.11 - samples/sec: 2705.14 - lr: 0.000021 - momentum: 0.000000
2023-10-13 08:46:26,454 epoch 4 - iter 117/138 - loss 0.07842970 - time (sec): 6.90 - samples/sec: 2738.62 - lr: 0.000021 - momentum: 0.000000
2023-10-13 08:46:27,288 epoch 4 - iter 130/138 - loss 0.07677490 - time (sec): 7.73 - samples/sec: 2775.36 - lr: 0.000020 - momentum: 0.000000
2023-10-13 08:46:27,742 ----------------------------------------------------------------------------------------------------
2023-10-13 08:46:27,742 EPOCH 4 done: loss 0.0747 - lr: 0.000020
2023-10-13 08:46:28,403 DEV : loss 0.13316737115383148 - f1-score (micro avg) 0.8571
2023-10-13 08:46:28,408 saving best model
2023-10-13 08:46:28,900 ----------------------------------------------------------------------------------------------------
2023-10-13 08:46:29,722 epoch 5 - iter 13/138 - loss 0.05668714 - time (sec): 0.81 - samples/sec: 2608.90 - lr: 0.000020 - momentum: 0.000000
2023-10-13 08:46:30,558 epoch 5 - iter 26/138 - loss 0.05073595 - time (sec): 1.65 - samples/sec: 2682.00 - lr: 0.000019 - momentum: 0.000000
2023-10-13 08:46:31,332 epoch 5 - iter 39/138 - loss 0.04877345 - time (sec): 2.42 - samples/sec: 2701.23 - lr: 0.000019 - momentum: 0.000000
2023-10-13 08:46:32,090 epoch 5 - iter 52/138 - loss 0.05670311 - time (sec): 3.18 - samples/sec: 2769.38 - lr: 0.000019 - momentum: 0.000000
2023-10-13 08:46:32,875 epoch 5 - iter 65/138 - loss 0.05551536 - time (sec): 3.97 - samples/sec: 2774.21 - lr: 0.000018 - momentum: 0.000000
2023-10-13 08:46:33,614 epoch 5 - iter 78/138 - loss 0.05335137 - time (sec): 4.71 - samples/sec: 2797.57 - lr: 0.000018 - momentum: 0.000000
2023-10-13 08:46:34,309 epoch 5 - iter 91/138 - loss 0.05219670 - time (sec): 5.40 - samples/sec: 2758.25 - lr: 0.000018 - momentum: 0.000000
2023-10-13 08:46:35,077 epoch 5 - iter 104/138 - loss 0.05411803 - time (sec): 6.17 - samples/sec: 2771.03 - lr: 0.000018 - momentum: 0.000000
2023-10-13 08:46:35,826 epoch 5 - iter 117/138 - loss 0.05510158 - time (sec): 6.92 - samples/sec: 2777.94 - lr: 0.000017 - momentum: 0.000000
2023-10-13 08:46:36,545 epoch 5 - iter 130/138 - loss 0.05336093 - time (sec): 7.64 - samples/sec: 2798.85 - lr: 0.000017 - momentum: 0.000000
2023-10-13 08:46:37,003 ----------------------------------------------------------------------------------------------------
2023-10-13 08:46:37,003 EPOCH 5 done: loss 0.0538 - lr: 0.000017
2023-10-13 08:46:37,711 DEV : loss 0.14850230515003204 - f1-score (micro avg) 0.8528
2023-10-13 08:46:37,717 ----------------------------------------------------------------------------------------------------
2023-10-13 08:46:38,500 epoch 6 - iter 13/138 - loss 0.00779610 - time (sec): 0.78 - samples/sec: 2672.02 - lr: 0.000016 - momentum: 0.000000
2023-10-13 08:46:39,310 epoch 6 - iter 26/138 - loss 0.02041797 - time (sec): 1.59 - samples/sec: 2769.61 - lr: 0.000016 - momentum: 0.000000
2023-10-13 08:46:40,119 epoch 6 - iter 39/138 - loss 0.03794121 - time (sec): 2.40 - samples/sec: 2781.08 - lr: 0.000016 - momentum: 0.000000
2023-10-13 08:46:40,846 epoch 6 - iter 52/138 - loss 0.03830060 - time (sec): 3.13 - samples/sec: 2742.92 - lr: 0.000015 - momentum: 0.000000
2023-10-13 08:46:41,612 epoch 6 - iter 65/138 - loss 0.03985147 - time (sec): 3.89 - samples/sec: 2764.55 - lr: 0.000015 - momentum: 0.000000
2023-10-13 08:46:42,415 epoch 6 - iter 78/138 - loss 0.03508542 - time (sec): 4.70 - samples/sec: 2757.98 - lr: 0.000015 - momentum: 0.000000
2023-10-13 08:46:43,244 epoch 6 - iter 91/138 - loss 0.03779644 - time (sec): 5.53 - samples/sec: 2755.05 - lr: 0.000015 - momentum: 0.000000
2023-10-13 08:46:43,983 epoch 6 - iter 104/138 - loss 0.03970915 - time (sec): 6.26 - samples/sec: 2752.57 - lr: 0.000014 - momentum: 0.000000
2023-10-13 08:46:44,755 epoch 6 - iter 117/138 - loss 0.04009455 - time (sec): 7.04 - samples/sec: 2786.24 - lr: 0.000014 - momentum: 0.000000
2023-10-13 08:46:45,551 epoch 6 - iter 130/138 - loss 0.03795431 - time (sec): 7.83 - samples/sec: 2774.63 - lr: 0.000014 - momentum: 0.000000
2023-10-13 08:46:46,004 ----------------------------------------------------------------------------------------------------
2023-10-13 08:46:46,004 EPOCH 6 done: loss 0.0367 - lr: 0.000014
2023-10-13 08:46:46,658 DEV : loss 0.15156938135623932 - f1-score (micro avg) 0.8659
2023-10-13 08:46:46,663 saving best model
2023-10-13 08:46:47,138 ----------------------------------------------------------------------------------------------------
2023-10-13 08:46:47,922 epoch 7 - iter 13/138 - loss 0.03061480 - time (sec): 0.78 - samples/sec: 2343.11 - lr: 0.000013 - momentum: 0.000000
2023-10-13 08:46:48,694 epoch 7 - iter 26/138 - loss 0.04366932 - time (sec): 1.55 - samples/sec: 2524.16 - lr: 0.000013 - momentum: 0.000000
2023-10-13 08:46:49,455 epoch 7 - iter 39/138 - loss 0.04163857 - time (sec): 2.32 - samples/sec: 2702.38 - lr: 0.000012 - momentum: 0.000000
2023-10-13 08:46:50,232 epoch 7 - iter 52/138 - loss 0.03973254 - time (sec): 3.09 - samples/sec: 2749.00 - lr: 0.000012 - momentum: 0.000000
2023-10-13 08:46:51,024 epoch 7 - iter 65/138 - loss 0.03769308 - time (sec): 3.88 - samples/sec: 2748.30 - lr: 0.000012 - momentum: 0.000000
2023-10-13 08:46:51,736 epoch 7 - iter 78/138 - loss 0.03971687 - time (sec): 4.60 - samples/sec: 2783.51 - lr: 0.000012 - momentum: 0.000000
2023-10-13 08:46:52,530 epoch 7 - iter 91/138 - loss 0.03645958 - time (sec): 5.39 - samples/sec: 2796.20 - lr: 0.000011 - momentum: 0.000000
2023-10-13 08:46:53,342 epoch 7 - iter 104/138 - loss 0.03421300 - time (sec): 6.20 - samples/sec: 2810.30 - lr: 0.000011 - momentum: 0.000000
2023-10-13 08:46:54,134 epoch 7 - iter 117/138 - loss 0.03374440 - time (sec): 6.99 - samples/sec: 2792.50 - lr: 0.000011 - momentum: 0.000000
2023-10-13 08:46:54,858 epoch 7 - iter 130/138 - loss 0.03172062 - time (sec): 7.72 - samples/sec: 2788.64 - lr: 0.000010 - momentum: 0.000000
2023-10-13 08:46:55,303 ----------------------------------------------------------------------------------------------------
2023-10-13 08:46:55,303 EPOCH 7 done: loss 0.0313 - lr: 0.000010
2023-10-13 08:46:55,993 DEV : loss 0.16039469838142395 - f1-score (micro avg) 0.8625
2023-10-13 08:46:55,999 ----------------------------------------------------------------------------------------------------
2023-10-13 08:46:56,743 epoch 8 - iter 13/138 - loss 0.01939940 - time (sec): 0.74 - samples/sec: 2741.27 - lr: 0.000010 - momentum: 0.000000
2023-10-13 08:46:57,483 epoch 8 - iter 26/138 - loss 0.01610771 - time (sec): 1.48 - samples/sec: 2759.33 - lr: 0.000009 - momentum: 0.000000
2023-10-13 08:46:58,298 epoch 8 - iter 39/138 - loss 0.01296939 - time (sec): 2.30 - samples/sec: 2663.91 - lr: 0.000009 - momentum: 0.000000
2023-10-13 08:46:59,109 epoch 8 - iter 52/138 - loss 0.02739881 - time (sec): 3.11 - samples/sec: 2715.86 - lr: 0.000009 - momentum: 0.000000
2023-10-13 08:46:59,851 epoch 8 - iter 65/138 - loss 0.02796483 - time (sec): 3.85 - samples/sec: 2723.50 - lr: 0.000009 - momentum: 0.000000
2023-10-13 08:47:00,614 epoch 8 - iter 78/138 - loss 0.02826291 - time (sec): 4.61 - samples/sec: 2716.07 - lr: 0.000008 - momentum: 0.000000
2023-10-13 08:47:01,366 epoch 8 - iter 91/138 - loss 0.02654856 - time (sec): 5.37 - samples/sec: 2751.15 - lr: 0.000008 - momentum: 0.000000
2023-10-13 08:47:02,204 epoch 8 - iter 104/138 - loss 0.02640491 - time (sec): 6.20 - samples/sec: 2753.61 - lr: 0.000008 - momentum: 0.000000
2023-10-13 08:47:03,043 epoch 8 - iter 117/138 - loss 0.02653671 - time (sec): 7.04 - samples/sec: 2741.03 - lr: 0.000007 - momentum: 0.000000
2023-10-13 08:47:03,755 epoch 8 - iter 130/138 - loss 0.02503400 - time (sec): 7.75 - samples/sec: 2763.58 - lr: 0.000007 - momentum: 0.000000
2023-10-13 08:47:04,226 ----------------------------------------------------------------------------------------------------
2023-10-13 08:47:04,227 EPOCH 8 done: loss 0.0272 - lr: 0.000007
2023-10-13 08:47:04,908 DEV : loss 0.15766263008117676 - f1-score (micro avg) 0.8729
2023-10-13 08:47:04,915 saving best model
2023-10-13 08:47:05,502 ----------------------------------------------------------------------------------------------------
2023-10-13 08:47:06,200 epoch 9 - iter 13/138 - loss 0.00078101 - time (sec): 0.69 - samples/sec: 2794.47 - lr: 0.000006 - momentum: 0.000000
2023-10-13 08:47:06,892 epoch 9 - iter 26/138 - loss 0.03214127 - time (sec): 1.39 - samples/sec: 2858.04 - lr: 0.000006 - momentum: 0.000000
2023-10-13 08:47:07,612 epoch 9 - iter 39/138 - loss 0.03065369 - time (sec): 2.11 - samples/sec: 2917.93 - lr: 0.000006 - momentum: 0.000000
2023-10-13 08:47:08,381 epoch 9 - iter 52/138 - loss 0.03039047 - time (sec): 2.88 - samples/sec: 2938.00 - lr: 0.000005 - momentum: 0.000000
2023-10-13 08:47:09,082 epoch 9 - iter 65/138 - loss 0.02685816 - time (sec): 3.58 - samples/sec: 2979.84 - lr: 0.000005 - momentum: 0.000000
2023-10-13 08:47:09,879 epoch 9 - iter 78/138 - loss 0.02302219 - time (sec): 4.37 - samples/sec: 2943.70 - lr: 0.000005 - momentum: 0.000000
2023-10-13 08:47:10,625 epoch 9 - iter 91/138 - loss 0.02039722 - time (sec): 5.12 - samples/sec: 2898.26 - lr: 0.000005 - momentum: 0.000000
2023-10-13 08:47:11,413 epoch 9 - iter 104/138 - loss 0.02021704 - time (sec): 5.91 - samples/sec: 2902.62 - lr: 0.000004 - momentum: 0.000000
2023-10-13 08:47:12,361 epoch 9 - iter 117/138 - loss 0.02018099 - time (sec): 6.86 - samples/sec: 2834.51 - lr: 0.000004 - momentum: 0.000000
2023-10-13 08:47:13,136 epoch 9 - iter 130/138 - loss 0.02039863 - time (sec): 7.63 - samples/sec: 2829.20 - lr: 0.000004 - momentum: 0.000000
2023-10-13 08:47:13,624 ----------------------------------------------------------------------------------------------------
2023-10-13 08:47:13,624 EPOCH 9 done: loss 0.0204 - lr: 0.000004
2023-10-13 08:47:14,284 DEV : loss 0.1462574005126953 - f1-score (micro avg) 0.8932
2023-10-13 08:47:14,289 saving best model
2023-10-13 08:47:14,747 ----------------------------------------------------------------------------------------------------
2023-10-13 08:47:15,508 epoch 10 - iter 13/138 - loss 0.00149311 - time (sec): 0.75 - samples/sec: 2621.90 - lr: 0.000003 - momentum: 0.000000
2023-10-13 08:47:16,285 epoch 10 - iter 26/138 - loss 0.01028203 - time (sec): 1.53 - samples/sec: 2753.93 - lr: 0.000003 - momentum: 0.000000
2023-10-13 08:47:17,047 epoch 10 - iter 39/138 - loss 0.01276969 - time (sec): 2.29 - samples/sec: 2737.87 - lr: 0.000002 - momentum: 0.000000
2023-10-13 08:47:17,814 epoch 10 - iter 52/138 - loss 0.01643193 - time (sec): 3.06 - samples/sec: 2749.39 - lr: 0.000002 - momentum: 0.000000
2023-10-13 08:47:18,611 epoch 10 - iter 65/138 - loss 0.01821951 - time (sec): 3.86 - samples/sec: 2757.10 - lr: 0.000002 - momentum: 0.000000
2023-10-13 08:47:19,411 epoch 10 - iter 78/138 - loss 0.01906226 - time (sec): 4.66 - samples/sec: 2765.54 - lr: 0.000002 - momentum: 0.000000
2023-10-13 08:47:20,157 epoch 10 - iter 91/138 - loss 0.01808510 - time (sec): 5.40 - samples/sec: 2788.91 - lr: 0.000001 - momentum: 0.000000
2023-10-13 08:47:20,927 epoch 10 - iter 104/138 - loss 0.02012840 - time (sec): 6.17 - samples/sec: 2783.85 - lr: 0.000001 - momentum: 0.000000
2023-10-13 08:47:21,683 epoch 10 - iter 117/138 - loss 0.01893462 - time (sec): 6.93 - samples/sec: 2784.50 - lr: 0.000001 - momentum: 0.000000
2023-10-13 08:47:22,501 epoch 10 - iter 130/138 - loss 0.01782572 - time (sec): 7.75 - samples/sec: 2779.88 - lr: 0.000000 - momentum: 0.000000
2023-10-13 08:47:22,971 ----------------------------------------------------------------------------------------------------
2023-10-13 08:47:22,971 EPOCH 10 done: loss 0.0188 - lr: 0.000000
2023-10-13 08:47:23,630 DEV : loss 0.1496874839067459 - f1-score (micro avg) 0.8854
2023-10-13 08:47:23,990 ----------------------------------------------------------------------------------------------------
2023-10-13 08:47:23,991 Loading model from best epoch ...
2023-10-13 08:47:25,655 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
2023-10-13 08:47:26,339
Results:
- F-score (micro) 0.9027
- F-score (macro) 0.6411
- Accuracy 0.8365
By class:
precision recall f1-score support
scope 0.8757 0.9205 0.8975 176
pers 0.9528 0.9453 0.9490 128
work 0.8533 0.8649 0.8591 74
loc 0.5000 0.5000 0.5000 2
object 0.0000 0.0000 0.0000 2
micro avg 0.8946 0.9110 0.9027 382
macro avg 0.6364 0.6461 0.6411 382
weighted avg 0.8906 0.9110 0.9005 382
2023-10-13 08:47:26,339 ----------------------------------------------------------------------------------------------------
|