File size: 24,035 Bytes
31275b4 |
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 11:44:15,758 ----------------------------------------------------------------------------------------------------
2023-10-13 11:44:15,759 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=21, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-13 11:44:15,759 ----------------------------------------------------------------------------------------------------
2023-10-13 11:44:15,759 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
2023-10-13 11:44:15,759 ----------------------------------------------------------------------------------------------------
2023-10-13 11:44:15,759 Train: 3575 sentences
2023-10-13 11:44:15,759 (train_with_dev=False, train_with_test=False)
2023-10-13 11:44:15,759 ----------------------------------------------------------------------------------------------------
2023-10-13 11:44:15,759 Training Params:
2023-10-13 11:44:15,760 - learning_rate: "5e-05"
2023-10-13 11:44:15,760 - mini_batch_size: "8"
2023-10-13 11:44:15,760 - max_epochs: "10"
2023-10-13 11:44:15,760 - shuffle: "True"
2023-10-13 11:44:15,760 ----------------------------------------------------------------------------------------------------
2023-10-13 11:44:15,760 Plugins:
2023-10-13 11:44:15,760 - LinearScheduler | warmup_fraction: '0.1'
2023-10-13 11:44:15,760 ----------------------------------------------------------------------------------------------------
2023-10-13 11:44:15,760 Final evaluation on model from best epoch (best-model.pt)
2023-10-13 11:44:15,760 - metric: "('micro avg', 'f1-score')"
2023-10-13 11:44:15,760 ----------------------------------------------------------------------------------------------------
2023-10-13 11:44:15,760 Computation:
2023-10-13 11:44:15,760 - compute on device: cuda:0
2023-10-13 11:44:15,760 - embedding storage: none
2023-10-13 11:44:15,760 ----------------------------------------------------------------------------------------------------
2023-10-13 11:44:15,760 Model training base path: "hmbench-hipe2020/de-dbmdz/bert-base-historic-multilingual-cased-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-1"
2023-10-13 11:44:15,760 ----------------------------------------------------------------------------------------------------
2023-10-13 11:44:15,760 ----------------------------------------------------------------------------------------------------
2023-10-13 11:44:18,820 epoch 1 - iter 44/447 - loss 3.00192754 - time (sec): 3.06 - samples/sec: 3107.41 - lr: 0.000005 - momentum: 0.000000
2023-10-13 11:44:21,453 epoch 1 - iter 88/447 - loss 2.12561225 - time (sec): 5.69 - samples/sec: 3053.75 - lr: 0.000010 - momentum: 0.000000
2023-10-13 11:44:24,069 epoch 1 - iter 132/447 - loss 1.62736701 - time (sec): 8.31 - samples/sec: 3012.86 - lr: 0.000015 - momentum: 0.000000
2023-10-13 11:44:26,817 epoch 1 - iter 176/447 - loss 1.31615057 - time (sec): 11.06 - samples/sec: 3025.58 - lr: 0.000020 - momentum: 0.000000
2023-10-13 11:44:29,573 epoch 1 - iter 220/447 - loss 1.13111058 - time (sec): 13.81 - samples/sec: 3009.28 - lr: 0.000024 - momentum: 0.000000
2023-10-13 11:44:32,463 epoch 1 - iter 264/447 - loss 0.98651046 - time (sec): 16.70 - samples/sec: 3017.43 - lr: 0.000029 - momentum: 0.000000
2023-10-13 11:44:35,214 epoch 1 - iter 308/447 - loss 0.88523112 - time (sec): 19.45 - samples/sec: 3032.23 - lr: 0.000034 - momentum: 0.000000
2023-10-13 11:44:38,573 epoch 1 - iter 352/447 - loss 0.79933006 - time (sec): 22.81 - samples/sec: 2982.26 - lr: 0.000039 - momentum: 0.000000
2023-10-13 11:44:41,327 epoch 1 - iter 396/447 - loss 0.73789913 - time (sec): 25.57 - samples/sec: 2984.81 - lr: 0.000044 - momentum: 0.000000
2023-10-13 11:44:44,373 epoch 1 - iter 440/447 - loss 0.68906501 - time (sec): 28.61 - samples/sec: 2985.26 - lr: 0.000049 - momentum: 0.000000
2023-10-13 11:44:44,800 ----------------------------------------------------------------------------------------------------
2023-10-13 11:44:44,800 EPOCH 1 done: loss 0.6828 - lr: 0.000049
2023-10-13 11:44:49,329 DEV : loss 0.17432522773742676 - f1-score (micro avg) 0.6361
2023-10-13 11:44:49,359 saving best model
2023-10-13 11:44:49,798 ----------------------------------------------------------------------------------------------------
2023-10-13 11:44:52,910 epoch 2 - iter 44/447 - loss 0.16099947 - time (sec): 3.11 - samples/sec: 2748.55 - lr: 0.000049 - momentum: 0.000000
2023-10-13 11:44:55,787 epoch 2 - iter 88/447 - loss 0.18473236 - time (sec): 5.99 - samples/sec: 2846.17 - lr: 0.000049 - momentum: 0.000000
2023-10-13 11:44:58,532 epoch 2 - iter 132/447 - loss 0.17741855 - time (sec): 8.73 - samples/sec: 2942.70 - lr: 0.000048 - momentum: 0.000000
2023-10-13 11:45:01,432 epoch 2 - iter 176/447 - loss 0.17120620 - time (sec): 11.63 - samples/sec: 2930.41 - lr: 0.000048 - momentum: 0.000000
2023-10-13 11:45:04,076 epoch 2 - iter 220/447 - loss 0.16623014 - time (sec): 14.28 - samples/sec: 2939.48 - lr: 0.000047 - momentum: 0.000000
2023-10-13 11:45:07,067 epoch 2 - iter 264/447 - loss 0.15782578 - time (sec): 17.27 - samples/sec: 2928.18 - lr: 0.000047 - momentum: 0.000000
2023-10-13 11:45:10,075 epoch 2 - iter 308/447 - loss 0.15657926 - time (sec): 20.28 - samples/sec: 2950.92 - lr: 0.000046 - momentum: 0.000000
2023-10-13 11:45:12,864 epoch 2 - iter 352/447 - loss 0.15440721 - time (sec): 23.06 - samples/sec: 2941.58 - lr: 0.000046 - momentum: 0.000000
2023-10-13 11:45:15,607 epoch 2 - iter 396/447 - loss 0.15306937 - time (sec): 25.81 - samples/sec: 2941.23 - lr: 0.000045 - momentum: 0.000000
2023-10-13 11:45:18,675 epoch 2 - iter 440/447 - loss 0.15097223 - time (sec): 28.88 - samples/sec: 2955.45 - lr: 0.000045 - momentum: 0.000000
2023-10-13 11:45:19,094 ----------------------------------------------------------------------------------------------------
2023-10-13 11:45:19,094 EPOCH 2 done: loss 0.1500 - lr: 0.000045
2023-10-13 11:45:27,170 DEV : loss 0.128895103931427 - f1-score (micro avg) 0.7135
2023-10-13 11:45:27,197 saving best model
2023-10-13 11:45:27,641 ----------------------------------------------------------------------------------------------------
2023-10-13 11:45:30,333 epoch 3 - iter 44/447 - loss 0.08954463 - time (sec): 2.69 - samples/sec: 2870.97 - lr: 0.000044 - momentum: 0.000000
2023-10-13 11:45:32,983 epoch 3 - iter 88/447 - loss 0.07907098 - time (sec): 5.34 - samples/sec: 2990.59 - lr: 0.000043 - momentum: 0.000000
2023-10-13 11:45:35,676 epoch 3 - iter 132/447 - loss 0.08762480 - time (sec): 8.03 - samples/sec: 2991.05 - lr: 0.000043 - momentum: 0.000000
2023-10-13 11:45:39,076 epoch 3 - iter 176/447 - loss 0.08018442 - time (sec): 11.43 - samples/sec: 2875.98 - lr: 0.000042 - momentum: 0.000000
2023-10-13 11:45:42,286 epoch 3 - iter 220/447 - loss 0.08160036 - time (sec): 14.64 - samples/sec: 2864.74 - lr: 0.000042 - momentum: 0.000000
2023-10-13 11:45:45,033 epoch 3 - iter 264/447 - loss 0.07929525 - time (sec): 17.39 - samples/sec: 2906.70 - lr: 0.000041 - momentum: 0.000000
2023-10-13 11:45:47,994 epoch 3 - iter 308/447 - loss 0.08305282 - time (sec): 20.35 - samples/sec: 2906.96 - lr: 0.000041 - momentum: 0.000000
2023-10-13 11:45:51,011 epoch 3 - iter 352/447 - loss 0.08280962 - time (sec): 23.36 - samples/sec: 2901.89 - lr: 0.000040 - momentum: 0.000000
2023-10-13 11:45:53,745 epoch 3 - iter 396/447 - loss 0.08300649 - time (sec): 26.10 - samples/sec: 2916.83 - lr: 0.000040 - momentum: 0.000000
2023-10-13 11:45:57,008 epoch 3 - iter 440/447 - loss 0.08157697 - time (sec): 29.36 - samples/sec: 2910.52 - lr: 0.000039 - momentum: 0.000000
2023-10-13 11:45:57,419 ----------------------------------------------------------------------------------------------------
2023-10-13 11:45:57,419 EPOCH 3 done: loss 0.0817 - lr: 0.000039
2023-10-13 11:46:05,551 DEV : loss 0.12530890107154846 - f1-score (micro avg) 0.7312
2023-10-13 11:46:05,579 saving best model
2023-10-13 11:46:06,046 ----------------------------------------------------------------------------------------------------
2023-10-13 11:46:08,859 epoch 4 - iter 44/447 - loss 0.06395096 - time (sec): 2.81 - samples/sec: 3188.81 - lr: 0.000038 - momentum: 0.000000
2023-10-13 11:46:11,496 epoch 4 - iter 88/447 - loss 0.05609227 - time (sec): 5.45 - samples/sec: 3121.01 - lr: 0.000038 - momentum: 0.000000
2023-10-13 11:46:14,489 epoch 4 - iter 132/447 - loss 0.05254098 - time (sec): 8.44 - samples/sec: 3085.83 - lr: 0.000037 - momentum: 0.000000
2023-10-13 11:46:17,596 epoch 4 - iter 176/447 - loss 0.05434937 - time (sec): 11.55 - samples/sec: 3089.85 - lr: 0.000037 - momentum: 0.000000
2023-10-13 11:46:20,600 epoch 4 - iter 220/447 - loss 0.05216684 - time (sec): 14.55 - samples/sec: 3047.80 - lr: 0.000036 - momentum: 0.000000
2023-10-13 11:46:23,946 epoch 4 - iter 264/447 - loss 0.05292772 - time (sec): 17.90 - samples/sec: 2960.39 - lr: 0.000036 - momentum: 0.000000
2023-10-13 11:46:26,646 epoch 4 - iter 308/447 - loss 0.05273653 - time (sec): 20.60 - samples/sec: 2976.45 - lr: 0.000035 - momentum: 0.000000
2023-10-13 11:46:29,376 epoch 4 - iter 352/447 - loss 0.05196364 - time (sec): 23.33 - samples/sec: 2987.17 - lr: 0.000035 - momentum: 0.000000
2023-10-13 11:46:31,872 epoch 4 - iter 396/447 - loss 0.04948670 - time (sec): 25.82 - samples/sec: 2976.55 - lr: 0.000034 - momentum: 0.000000
2023-10-13 11:46:34,756 epoch 4 - iter 440/447 - loss 0.04973611 - time (sec): 28.71 - samples/sec: 2973.55 - lr: 0.000033 - momentum: 0.000000
2023-10-13 11:46:35,161 ----------------------------------------------------------------------------------------------------
2023-10-13 11:46:35,162 EPOCH 4 done: loss 0.0506 - lr: 0.000033
2023-10-13 11:46:43,220 DEV : loss 0.16665692627429962 - f1-score (micro avg) 0.7594
2023-10-13 11:46:43,248 saving best model
2023-10-13 11:46:43,743 ----------------------------------------------------------------------------------------------------
2023-10-13 11:46:47,190 epoch 5 - iter 44/447 - loss 0.04162747 - time (sec): 3.44 - samples/sec: 2799.52 - lr: 0.000033 - momentum: 0.000000
2023-10-13 11:46:49,839 epoch 5 - iter 88/447 - loss 0.03504743 - time (sec): 6.09 - samples/sec: 2894.26 - lr: 0.000032 - momentum: 0.000000
2023-10-13 11:46:52,807 epoch 5 - iter 132/447 - loss 0.03495945 - time (sec): 9.06 - samples/sec: 2897.46 - lr: 0.000032 - momentum: 0.000000
2023-10-13 11:46:55,586 epoch 5 - iter 176/447 - loss 0.03617387 - time (sec): 11.84 - samples/sec: 2902.83 - lr: 0.000031 - momentum: 0.000000
2023-10-13 11:46:58,575 epoch 5 - iter 220/447 - loss 0.03429710 - time (sec): 14.83 - samples/sec: 2925.74 - lr: 0.000031 - momentum: 0.000000
2023-10-13 11:47:01,324 epoch 5 - iter 264/447 - loss 0.03544567 - time (sec): 17.58 - samples/sec: 2959.14 - lr: 0.000030 - momentum: 0.000000
2023-10-13 11:47:04,094 epoch 5 - iter 308/447 - loss 0.03369941 - time (sec): 20.35 - samples/sec: 2952.63 - lr: 0.000030 - momentum: 0.000000
2023-10-13 11:47:06,957 epoch 5 - iter 352/447 - loss 0.03451202 - time (sec): 23.21 - samples/sec: 2962.77 - lr: 0.000029 - momentum: 0.000000
2023-10-13 11:47:10,024 epoch 5 - iter 396/447 - loss 0.03499281 - time (sec): 26.28 - samples/sec: 2917.55 - lr: 0.000028 - momentum: 0.000000
2023-10-13 11:47:12,877 epoch 5 - iter 440/447 - loss 0.03611396 - time (sec): 29.13 - samples/sec: 2927.39 - lr: 0.000028 - momentum: 0.000000
2023-10-13 11:47:13,320 ----------------------------------------------------------------------------------------------------
2023-10-13 11:47:13,321 EPOCH 5 done: loss 0.0361 - lr: 0.000028
2023-10-13 11:47:21,826 DEV : loss 0.19516603648662567 - f1-score (micro avg) 0.7573
2023-10-13 11:47:21,857 ----------------------------------------------------------------------------------------------------
2023-10-13 11:47:24,788 epoch 6 - iter 44/447 - loss 0.02058854 - time (sec): 2.93 - samples/sec: 2932.43 - lr: 0.000027 - momentum: 0.000000
2023-10-13 11:47:27,413 epoch 6 - iter 88/447 - loss 0.02229072 - time (sec): 5.55 - samples/sec: 2914.52 - lr: 0.000027 - momentum: 0.000000
2023-10-13 11:47:30,610 epoch 6 - iter 132/447 - loss 0.01983331 - time (sec): 8.75 - samples/sec: 2861.48 - lr: 0.000026 - momentum: 0.000000
2023-10-13 11:47:33,740 epoch 6 - iter 176/447 - loss 0.02048267 - time (sec): 11.88 - samples/sec: 2858.33 - lr: 0.000026 - momentum: 0.000000
2023-10-13 11:47:36,507 epoch 6 - iter 220/447 - loss 0.01960330 - time (sec): 14.65 - samples/sec: 2836.30 - lr: 0.000025 - momentum: 0.000000
2023-10-13 11:47:39,303 epoch 6 - iter 264/447 - loss 0.01907838 - time (sec): 17.45 - samples/sec: 2843.15 - lr: 0.000025 - momentum: 0.000000
2023-10-13 11:47:41,984 epoch 6 - iter 308/447 - loss 0.02150005 - time (sec): 20.13 - samples/sec: 2851.58 - lr: 0.000024 - momentum: 0.000000
2023-10-13 11:47:44,621 epoch 6 - iter 352/447 - loss 0.02176467 - time (sec): 22.76 - samples/sec: 2898.74 - lr: 0.000023 - momentum: 0.000000
2023-10-13 11:47:47,920 epoch 6 - iter 396/447 - loss 0.02302182 - time (sec): 26.06 - samples/sec: 2922.86 - lr: 0.000023 - momentum: 0.000000
2023-10-13 11:47:50,941 epoch 6 - iter 440/447 - loss 0.02339372 - time (sec): 29.08 - samples/sec: 2930.41 - lr: 0.000022 - momentum: 0.000000
2023-10-13 11:47:51,371 ----------------------------------------------------------------------------------------------------
2023-10-13 11:47:51,372 EPOCH 6 done: loss 0.0233 - lr: 0.000022
2023-10-13 11:47:59,864 DEV : loss 0.20644259452819824 - f1-score (micro avg) 0.7515
2023-10-13 11:47:59,893 ----------------------------------------------------------------------------------------------------
2023-10-13 11:48:02,613 epoch 7 - iter 44/447 - loss 0.01706592 - time (sec): 2.72 - samples/sec: 3215.33 - lr: 0.000022 - momentum: 0.000000
2023-10-13 11:48:05,274 epoch 7 - iter 88/447 - loss 0.01709019 - time (sec): 5.38 - samples/sec: 3137.91 - lr: 0.000021 - momentum: 0.000000
2023-10-13 11:48:08,702 epoch 7 - iter 132/447 - loss 0.01581202 - time (sec): 8.81 - samples/sec: 3063.55 - lr: 0.000021 - momentum: 0.000000
2023-10-13 11:48:11,559 epoch 7 - iter 176/447 - loss 0.01433771 - time (sec): 11.66 - samples/sec: 3028.44 - lr: 0.000020 - momentum: 0.000000
2023-10-13 11:48:14,480 epoch 7 - iter 220/447 - loss 0.01468541 - time (sec): 14.59 - samples/sec: 3021.21 - lr: 0.000020 - momentum: 0.000000
2023-10-13 11:48:17,059 epoch 7 - iter 264/447 - loss 0.01493250 - time (sec): 17.16 - samples/sec: 3031.79 - lr: 0.000019 - momentum: 0.000000
2023-10-13 11:48:19,824 epoch 7 - iter 308/447 - loss 0.01340649 - time (sec): 19.93 - samples/sec: 3018.85 - lr: 0.000018 - momentum: 0.000000
2023-10-13 11:48:22,745 epoch 7 - iter 352/447 - loss 0.01446296 - time (sec): 22.85 - samples/sec: 2997.46 - lr: 0.000018 - momentum: 0.000000
2023-10-13 11:48:25,409 epoch 7 - iter 396/447 - loss 0.01498454 - time (sec): 25.51 - samples/sec: 2984.77 - lr: 0.000017 - momentum: 0.000000
2023-10-13 11:48:28,022 epoch 7 - iter 440/447 - loss 0.01502147 - time (sec): 28.13 - samples/sec: 2996.49 - lr: 0.000017 - momentum: 0.000000
2023-10-13 11:48:28,735 ----------------------------------------------------------------------------------------------------
2023-10-13 11:48:28,735 EPOCH 7 done: loss 0.0149 - lr: 0.000017
2023-10-13 11:48:37,330 DEV : loss 0.19954054057598114 - f1-score (micro avg) 0.7814
2023-10-13 11:48:37,360 saving best model
2023-10-13 11:48:37,781 ----------------------------------------------------------------------------------------------------
2023-10-13 11:48:40,889 epoch 8 - iter 44/447 - loss 0.00534479 - time (sec): 3.11 - samples/sec: 2758.32 - lr: 0.000016 - momentum: 0.000000
2023-10-13 11:48:43,944 epoch 8 - iter 88/447 - loss 0.00779708 - time (sec): 6.16 - samples/sec: 2854.83 - lr: 0.000016 - momentum: 0.000000
2023-10-13 11:48:46,751 epoch 8 - iter 132/447 - loss 0.00659972 - time (sec): 8.97 - samples/sec: 2973.02 - lr: 0.000015 - momentum: 0.000000
2023-10-13 11:48:49,824 epoch 8 - iter 176/447 - loss 0.00669648 - time (sec): 12.04 - samples/sec: 2999.66 - lr: 0.000015 - momentum: 0.000000
2023-10-13 11:48:52,453 epoch 8 - iter 220/447 - loss 0.00843461 - time (sec): 14.67 - samples/sec: 2992.45 - lr: 0.000014 - momentum: 0.000000
2023-10-13 11:48:55,276 epoch 8 - iter 264/447 - loss 0.00900303 - time (sec): 17.49 - samples/sec: 2971.82 - lr: 0.000013 - momentum: 0.000000
2023-10-13 11:48:58,051 epoch 8 - iter 308/447 - loss 0.00962997 - time (sec): 20.27 - samples/sec: 3006.72 - lr: 0.000013 - momentum: 0.000000
2023-10-13 11:49:00,666 epoch 8 - iter 352/447 - loss 0.00959478 - time (sec): 22.88 - samples/sec: 3026.20 - lr: 0.000012 - momentum: 0.000000
2023-10-13 11:49:03,344 epoch 8 - iter 396/447 - loss 0.00940630 - time (sec): 25.56 - samples/sec: 3027.67 - lr: 0.000012 - momentum: 0.000000
2023-10-13 11:49:06,125 epoch 8 - iter 440/447 - loss 0.00959920 - time (sec): 28.34 - samples/sec: 3009.55 - lr: 0.000011 - momentum: 0.000000
2023-10-13 11:49:06,520 ----------------------------------------------------------------------------------------------------
2023-10-13 11:49:06,521 EPOCH 8 done: loss 0.0098 - lr: 0.000011
2023-10-13 11:49:14,950 DEV : loss 0.22160013020038605 - f1-score (micro avg) 0.7869
2023-10-13 11:49:14,980 saving best model
2023-10-13 11:49:15,445 ----------------------------------------------------------------------------------------------------
2023-10-13 11:49:18,252 epoch 9 - iter 44/447 - loss 0.00961171 - time (sec): 2.80 - samples/sec: 2905.04 - lr: 0.000011 - momentum: 0.000000
2023-10-13 11:49:21,188 epoch 9 - iter 88/447 - loss 0.00795760 - time (sec): 5.74 - samples/sec: 3027.22 - lr: 0.000010 - momentum: 0.000000
2023-10-13 11:49:24,129 epoch 9 - iter 132/447 - loss 0.00787975 - time (sec): 8.68 - samples/sec: 2963.45 - lr: 0.000010 - momentum: 0.000000
2023-10-13 11:49:27,046 epoch 9 - iter 176/447 - loss 0.00643769 - time (sec): 11.60 - samples/sec: 3004.95 - lr: 0.000009 - momentum: 0.000000
2023-10-13 11:49:30,207 epoch 9 - iter 220/447 - loss 0.00550591 - time (sec): 14.76 - samples/sec: 2958.57 - lr: 0.000008 - momentum: 0.000000
2023-10-13 11:49:32,877 epoch 9 - iter 264/447 - loss 0.00628025 - time (sec): 17.43 - samples/sec: 2976.74 - lr: 0.000008 - momentum: 0.000000
2023-10-13 11:49:35,917 epoch 9 - iter 308/447 - loss 0.00568075 - time (sec): 20.47 - samples/sec: 3002.41 - lr: 0.000007 - momentum: 0.000000
2023-10-13 11:49:38,518 epoch 9 - iter 352/447 - loss 0.00547781 - time (sec): 23.07 - samples/sec: 3011.57 - lr: 0.000007 - momentum: 0.000000
2023-10-13 11:49:41,118 epoch 9 - iter 396/447 - loss 0.00510707 - time (sec): 25.67 - samples/sec: 3015.03 - lr: 0.000006 - momentum: 0.000000
2023-10-13 11:49:43,957 epoch 9 - iter 440/447 - loss 0.00579440 - time (sec): 28.51 - samples/sec: 2993.00 - lr: 0.000006 - momentum: 0.000000
2023-10-13 11:49:44,368 ----------------------------------------------------------------------------------------------------
2023-10-13 11:49:44,369 EPOCH 9 done: loss 0.0058 - lr: 0.000006
2023-10-13 11:49:52,737 DEV : loss 0.22420544922351837 - f1-score (micro avg) 0.7939
2023-10-13 11:49:52,765 saving best model
2023-10-13 11:49:53,207 ----------------------------------------------------------------------------------------------------
2023-10-13 11:49:56,079 epoch 10 - iter 44/447 - loss 0.00503684 - time (sec): 2.87 - samples/sec: 3031.57 - lr: 0.000005 - momentum: 0.000000
2023-10-13 11:49:58,702 epoch 10 - iter 88/447 - loss 0.00356275 - time (sec): 5.49 - samples/sec: 3012.23 - lr: 0.000005 - momentum: 0.000000
2023-10-13 11:50:01,395 epoch 10 - iter 132/447 - loss 0.00318696 - time (sec): 8.18 - samples/sec: 3059.58 - lr: 0.000004 - momentum: 0.000000
2023-10-13 11:50:04,264 epoch 10 - iter 176/447 - loss 0.00359811 - time (sec): 11.05 - samples/sec: 3046.15 - lr: 0.000003 - momentum: 0.000000
2023-10-13 11:50:07,363 epoch 10 - iter 220/447 - loss 0.00385787 - time (sec): 14.15 - samples/sec: 3025.06 - lr: 0.000003 - momentum: 0.000000
2023-10-13 11:50:10,306 epoch 10 - iter 264/447 - loss 0.00365876 - time (sec): 17.10 - samples/sec: 3014.61 - lr: 0.000002 - momentum: 0.000000
2023-10-13 11:50:13,281 epoch 10 - iter 308/447 - loss 0.00370964 - time (sec): 20.07 - samples/sec: 3007.91 - lr: 0.000002 - momentum: 0.000000
2023-10-13 11:50:15,872 epoch 10 - iter 352/447 - loss 0.00391637 - time (sec): 22.66 - samples/sec: 3016.13 - lr: 0.000001 - momentum: 0.000000
2023-10-13 11:50:18,576 epoch 10 - iter 396/447 - loss 0.00392657 - time (sec): 25.36 - samples/sec: 3013.15 - lr: 0.000001 - momentum: 0.000000
2023-10-13 11:50:21,544 epoch 10 - iter 440/447 - loss 0.00407881 - time (sec): 28.33 - samples/sec: 2997.94 - lr: 0.000000 - momentum: 0.000000
2023-10-13 11:50:22,045 ----------------------------------------------------------------------------------------------------
2023-10-13 11:50:22,045 EPOCH 10 done: loss 0.0040 - lr: 0.000000
2023-10-13 11:50:30,395 DEV : loss 0.2235824018716812 - f1-score (micro avg) 0.7934
2023-10-13 11:50:30,794 ----------------------------------------------------------------------------------------------------
2023-10-13 11:50:30,796 Loading model from best epoch ...
2023-10-13 11:50:32,326 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
2023-10-13 11:50:36,754
Results:
- F-score (micro) 0.7481
- F-score (macro) 0.6626
- Accuracy 0.6183
By class:
precision recall f1-score support
loc 0.8471 0.8456 0.8463 596
pers 0.6702 0.7568 0.7109 333
org 0.5227 0.5227 0.5227 132
prod 0.5714 0.4848 0.5246 66
time 0.7234 0.6939 0.7083 49
micro avg 0.7388 0.7577 0.7481 1176
macro avg 0.6670 0.6608 0.6626 1176
weighted avg 0.7400 0.7577 0.7478 1176
2023-10-13 11:50:36,754 ----------------------------------------------------------------------------------------------------
|