File size: 23,992 Bytes
f347efc |
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 |
2023-10-13 08:55:31,689 ----------------------------------------------------------------------------------------------------
2023-10-13 08:55:31,690 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 08:55:31,690 ----------------------------------------------------------------------------------------------------
2023-10-13 08:55:31,691 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 08:55:31,691 ----------------------------------------------------------------------------------------------------
2023-10-13 08:55:31,691 Train: 1100 sentences
2023-10-13 08:55:31,691 (train_with_dev=False, train_with_test=False)
2023-10-13 08:55:31,691 ----------------------------------------------------------------------------------------------------
2023-10-13 08:55:31,691 Training Params:
2023-10-13 08:55:31,691 - learning_rate: "3e-05"
2023-10-13 08:55:31,691 - mini_batch_size: "8"
2023-10-13 08:55:31,691 - max_epochs: "10"
2023-10-13 08:55:31,691 - shuffle: "True"
2023-10-13 08:55:31,691 ----------------------------------------------------------------------------------------------------
2023-10-13 08:55:31,691 Plugins:
2023-10-13 08:55:31,691 - LinearScheduler | warmup_fraction: '0.1'
2023-10-13 08:55:31,691 ----------------------------------------------------------------------------------------------------
2023-10-13 08:55:31,691 Final evaluation on model from best epoch (best-model.pt)
2023-10-13 08:55:31,691 - metric: "('micro avg', 'f1-score')"
2023-10-13 08:55:31,691 ----------------------------------------------------------------------------------------------------
2023-10-13 08:55:31,691 Computation:
2023-10-13 08:55:31,691 - compute on device: cuda:0
2023-10-13 08:55:31,691 - embedding storage: none
2023-10-13 08:55:31,691 ----------------------------------------------------------------------------------------------------
2023-10-13 08:55:31,691 Model training base path: "hmbench-ajmc/de-dbmdz/bert-base-historic-multilingual-cased-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-5"
2023-10-13 08:55:31,691 ----------------------------------------------------------------------------------------------------
2023-10-13 08:55:31,691 ----------------------------------------------------------------------------------------------------
2023-10-13 08:55:32,450 epoch 1 - iter 13/138 - loss 3.23164083 - time (sec): 0.76 - samples/sec: 2657.16 - lr: 0.000003 - momentum: 0.000000
2023-10-13 08:55:33,231 epoch 1 - iter 26/138 - loss 3.07083348 - time (sec): 1.54 - samples/sec: 2592.11 - lr: 0.000005 - momentum: 0.000000
2023-10-13 08:55:34,010 epoch 1 - iter 39/138 - loss 2.71985264 - time (sec): 2.32 - samples/sec: 2587.53 - lr: 0.000008 - momentum: 0.000000
2023-10-13 08:55:34,819 epoch 1 - iter 52/138 - loss 2.24509201 - time (sec): 3.13 - samples/sec: 2663.88 - lr: 0.000011 - momentum: 0.000000
2023-10-13 08:55:35,628 epoch 1 - iter 65/138 - loss 1.93631872 - time (sec): 3.94 - samples/sec: 2694.87 - lr: 0.000014 - momentum: 0.000000
2023-10-13 08:55:36,459 epoch 1 - iter 78/138 - loss 1.73605929 - time (sec): 4.77 - samples/sec: 2683.95 - lr: 0.000017 - momentum: 0.000000
2023-10-13 08:55:37,270 epoch 1 - iter 91/138 - loss 1.57473090 - time (sec): 5.58 - samples/sec: 2672.60 - lr: 0.000020 - momentum: 0.000000
2023-10-13 08:55:38,072 epoch 1 - iter 104/138 - loss 1.43433242 - time (sec): 6.38 - samples/sec: 2698.19 - lr: 0.000022 - momentum: 0.000000
2023-10-13 08:55:38,741 epoch 1 - iter 117/138 - loss 1.33051183 - time (sec): 7.05 - samples/sec: 2752.06 - lr: 0.000025 - momentum: 0.000000
2023-10-13 08:55:39,432 epoch 1 - iter 130/138 - loss 1.24843363 - time (sec): 7.74 - samples/sec: 2756.81 - lr: 0.000028 - momentum: 0.000000
2023-10-13 08:55:39,882 ----------------------------------------------------------------------------------------------------
2023-10-13 08:55:39,883 EPOCH 1 done: loss 1.2007 - lr: 0.000028
2023-10-13 08:55:40,600 DEV : loss 0.28690722584724426 - f1-score (micro avg) 0.6651
2023-10-13 08:55:40,605 saving best model
2023-10-13 08:55:40,989 ----------------------------------------------------------------------------------------------------
2023-10-13 08:55:41,681 epoch 2 - iter 13/138 - loss 0.32963642 - time (sec): 0.69 - samples/sec: 3035.52 - lr: 0.000030 - momentum: 0.000000
2023-10-13 08:55:42,381 epoch 2 - iter 26/138 - loss 0.27672936 - time (sec): 1.39 - samples/sec: 3080.92 - lr: 0.000029 - momentum: 0.000000
2023-10-13 08:55:43,105 epoch 2 - iter 39/138 - loss 0.24838311 - time (sec): 2.11 - samples/sec: 3088.92 - lr: 0.000029 - momentum: 0.000000
2023-10-13 08:55:43,865 epoch 2 - iter 52/138 - loss 0.24825315 - time (sec): 2.87 - samples/sec: 3116.82 - lr: 0.000029 - momentum: 0.000000
2023-10-13 08:55:44,587 epoch 2 - iter 65/138 - loss 0.24919275 - time (sec): 3.60 - samples/sec: 3175.76 - lr: 0.000028 - momentum: 0.000000
2023-10-13 08:55:45,265 epoch 2 - iter 78/138 - loss 0.24381568 - time (sec): 4.27 - samples/sec: 3108.71 - lr: 0.000028 - momentum: 0.000000
2023-10-13 08:55:45,969 epoch 2 - iter 91/138 - loss 0.23905698 - time (sec): 4.98 - samples/sec: 3065.00 - lr: 0.000028 - momentum: 0.000000
2023-10-13 08:55:46,728 epoch 2 - iter 104/138 - loss 0.23512468 - time (sec): 5.74 - samples/sec: 3041.71 - lr: 0.000028 - momentum: 0.000000
2023-10-13 08:55:47,505 epoch 2 - iter 117/138 - loss 0.22772616 - time (sec): 6.51 - samples/sec: 3004.87 - lr: 0.000027 - momentum: 0.000000
2023-10-13 08:55:48,178 epoch 2 - iter 130/138 - loss 0.22406658 - time (sec): 7.19 - samples/sec: 3008.30 - lr: 0.000027 - momentum: 0.000000
2023-10-13 08:55:48,628 ----------------------------------------------------------------------------------------------------
2023-10-13 08:55:48,629 EPOCH 2 done: loss 0.2199 - lr: 0.000027
2023-10-13 08:55:49,287 DEV : loss 0.14626161754131317 - f1-score (micro avg) 0.7981
2023-10-13 08:55:49,292 saving best model
2023-10-13 08:55:49,827 ----------------------------------------------------------------------------------------------------
2023-10-13 08:55:50,567 epoch 3 - iter 13/138 - loss 0.11244007 - time (sec): 0.74 - samples/sec: 2858.37 - lr: 0.000026 - momentum: 0.000000
2023-10-13 08:55:51,350 epoch 3 - iter 26/138 - loss 0.09545981 - time (sec): 1.52 - samples/sec: 2944.50 - lr: 0.000026 - momentum: 0.000000
2023-10-13 08:55:52,041 epoch 3 - iter 39/138 - loss 0.11179342 - time (sec): 2.21 - samples/sec: 2936.77 - lr: 0.000026 - momentum: 0.000000
2023-10-13 08:55:52,722 epoch 3 - iter 52/138 - loss 0.10944003 - time (sec): 2.89 - samples/sec: 2886.44 - lr: 0.000025 - momentum: 0.000000
2023-10-13 08:55:53,434 epoch 3 - iter 65/138 - loss 0.10962130 - time (sec): 3.60 - samples/sec: 2950.32 - lr: 0.000025 - momentum: 0.000000
2023-10-13 08:55:54,143 epoch 3 - iter 78/138 - loss 0.10776067 - time (sec): 4.31 - samples/sec: 2949.96 - lr: 0.000025 - momentum: 0.000000
2023-10-13 08:55:54,863 epoch 3 - iter 91/138 - loss 0.11389656 - time (sec): 5.03 - samples/sec: 3003.46 - lr: 0.000025 - momentum: 0.000000
2023-10-13 08:55:55,586 epoch 3 - iter 104/138 - loss 0.11174026 - time (sec): 5.75 - samples/sec: 3002.22 - lr: 0.000024 - momentum: 0.000000
2023-10-13 08:55:56,327 epoch 3 - iter 117/138 - loss 0.11375118 - time (sec): 6.50 - samples/sec: 3004.08 - lr: 0.000024 - momentum: 0.000000
2023-10-13 08:55:57,034 epoch 3 - iter 130/138 - loss 0.11290363 - time (sec): 7.20 - samples/sec: 2992.35 - lr: 0.000024 - momentum: 0.000000
2023-10-13 08:55:57,479 ----------------------------------------------------------------------------------------------------
2023-10-13 08:55:57,480 EPOCH 3 done: loss 0.1138 - lr: 0.000024
2023-10-13 08:55:58,116 DEV : loss 0.13030706346035004 - f1-score (micro avg) 0.8351
2023-10-13 08:55:58,122 saving best model
2023-10-13 08:55:58,619 ----------------------------------------------------------------------------------------------------
2023-10-13 08:55:59,326 epoch 4 - iter 13/138 - loss 0.05847433 - time (sec): 0.70 - samples/sec: 3245.80 - lr: 0.000023 - momentum: 0.000000
2023-10-13 08:56:00,013 epoch 4 - iter 26/138 - loss 0.07319254 - time (sec): 1.39 - samples/sec: 3278.88 - lr: 0.000023 - momentum: 0.000000
2023-10-13 08:56:00,755 epoch 4 - iter 39/138 - loss 0.06410931 - time (sec): 2.13 - samples/sec: 3160.08 - lr: 0.000022 - momentum: 0.000000
2023-10-13 08:56:01,417 epoch 4 - iter 52/138 - loss 0.06694482 - time (sec): 2.79 - samples/sec: 3055.12 - lr: 0.000022 - momentum: 0.000000
2023-10-13 08:56:02,160 epoch 4 - iter 65/138 - loss 0.07449259 - time (sec): 3.54 - samples/sec: 2987.09 - lr: 0.000022 - momentum: 0.000000
2023-10-13 08:56:02,857 epoch 4 - iter 78/138 - loss 0.07065528 - time (sec): 4.23 - samples/sec: 2995.98 - lr: 0.000021 - momentum: 0.000000
2023-10-13 08:56:03,582 epoch 4 - iter 91/138 - loss 0.07272318 - time (sec): 4.96 - samples/sec: 2970.50 - lr: 0.000021 - momentum: 0.000000
2023-10-13 08:56:04,384 epoch 4 - iter 104/138 - loss 0.07925456 - time (sec): 5.76 - samples/sec: 2982.04 - lr: 0.000021 - momentum: 0.000000
2023-10-13 08:56:05,116 epoch 4 - iter 117/138 - loss 0.07849385 - time (sec): 6.49 - samples/sec: 2969.64 - lr: 0.000021 - momentum: 0.000000
2023-10-13 08:56:05,816 epoch 4 - iter 130/138 - loss 0.08000911 - time (sec): 7.19 - samples/sec: 2974.95 - lr: 0.000020 - momentum: 0.000000
2023-10-13 08:56:06,282 ----------------------------------------------------------------------------------------------------
2023-10-13 08:56:06,283 EPOCH 4 done: loss 0.0806 - lr: 0.000020
2023-10-13 08:56:06,929 DEV : loss 0.11944959312677383 - f1-score (micro avg) 0.8647
2023-10-13 08:56:06,934 saving best model
2023-10-13 08:56:07,410 ----------------------------------------------------------------------------------------------------
2023-10-13 08:56:08,184 epoch 5 - iter 13/138 - loss 0.04455185 - time (sec): 0.77 - samples/sec: 2978.01 - lr: 0.000020 - momentum: 0.000000
2023-10-13 08:56:08,883 epoch 5 - iter 26/138 - loss 0.05992813 - time (sec): 1.47 - samples/sec: 2969.05 - lr: 0.000019 - momentum: 0.000000
2023-10-13 08:56:09,579 epoch 5 - iter 39/138 - loss 0.06847538 - time (sec): 2.17 - samples/sec: 2968.84 - lr: 0.000019 - momentum: 0.000000
2023-10-13 08:56:10,312 epoch 5 - iter 52/138 - loss 0.06921582 - time (sec): 2.90 - samples/sec: 3006.87 - lr: 0.000019 - momentum: 0.000000
2023-10-13 08:56:11,043 epoch 5 - iter 65/138 - loss 0.06454351 - time (sec): 3.63 - samples/sec: 3037.21 - lr: 0.000018 - momentum: 0.000000
2023-10-13 08:56:11,741 epoch 5 - iter 78/138 - loss 0.06104228 - time (sec): 4.33 - samples/sec: 2991.88 - lr: 0.000018 - momentum: 0.000000
2023-10-13 08:56:12,423 epoch 5 - iter 91/138 - loss 0.06053616 - time (sec): 5.01 - samples/sec: 3012.00 - lr: 0.000018 - momentum: 0.000000
2023-10-13 08:56:13,136 epoch 5 - iter 104/138 - loss 0.05702704 - time (sec): 5.72 - samples/sec: 3007.00 - lr: 0.000018 - momentum: 0.000000
2023-10-13 08:56:13,889 epoch 5 - iter 117/138 - loss 0.06057969 - time (sec): 6.48 - samples/sec: 3008.34 - lr: 0.000017 - momentum: 0.000000
2023-10-13 08:56:14,585 epoch 5 - iter 130/138 - loss 0.05796395 - time (sec): 7.17 - samples/sec: 2983.23 - lr: 0.000017 - momentum: 0.000000
2023-10-13 08:56:15,051 ----------------------------------------------------------------------------------------------------
2023-10-13 08:56:15,051 EPOCH 5 done: loss 0.0560 - lr: 0.000017
2023-10-13 08:56:15,712 DEV : loss 0.14307376742362976 - f1-score (micro avg) 0.8645
2023-10-13 08:56:15,718 ----------------------------------------------------------------------------------------------------
2023-10-13 08:56:16,409 epoch 6 - iter 13/138 - loss 0.05583016 - time (sec): 0.69 - samples/sec: 3077.69 - lr: 0.000016 - momentum: 0.000000
2023-10-13 08:56:17,103 epoch 6 - iter 26/138 - loss 0.05558456 - time (sec): 1.38 - samples/sec: 2980.23 - lr: 0.000016 - momentum: 0.000000
2023-10-13 08:56:17,794 epoch 6 - iter 39/138 - loss 0.06217709 - time (sec): 2.08 - samples/sec: 2932.93 - lr: 0.000016 - momentum: 0.000000
2023-10-13 08:56:18,592 epoch 6 - iter 52/138 - loss 0.05287790 - time (sec): 2.87 - samples/sec: 2886.05 - lr: 0.000015 - momentum: 0.000000
2023-10-13 08:56:19,321 epoch 6 - iter 65/138 - loss 0.05224642 - time (sec): 3.60 - samples/sec: 2871.09 - lr: 0.000015 - momentum: 0.000000
2023-10-13 08:56:20,086 epoch 6 - iter 78/138 - loss 0.05314834 - time (sec): 4.37 - samples/sec: 2851.29 - lr: 0.000015 - momentum: 0.000000
2023-10-13 08:56:20,850 epoch 6 - iter 91/138 - loss 0.05054307 - time (sec): 5.13 - samples/sec: 2839.96 - lr: 0.000015 - momentum: 0.000000
2023-10-13 08:56:21,583 epoch 6 - iter 104/138 - loss 0.04958115 - time (sec): 5.86 - samples/sec: 2864.84 - lr: 0.000014 - momentum: 0.000000
2023-10-13 08:56:22,358 epoch 6 - iter 117/138 - loss 0.04491562 - time (sec): 6.64 - samples/sec: 2897.75 - lr: 0.000014 - momentum: 0.000000
2023-10-13 08:56:23,124 epoch 6 - iter 130/138 - loss 0.04152667 - time (sec): 7.41 - samples/sec: 2908.51 - lr: 0.000014 - momentum: 0.000000
2023-10-13 08:56:23,611 ----------------------------------------------------------------------------------------------------
2023-10-13 08:56:23,612 EPOCH 6 done: loss 0.0423 - lr: 0.000014
2023-10-13 08:56:24,252 DEV : loss 0.15504451096057892 - f1-score (micro avg) 0.8682
2023-10-13 08:56:24,258 saving best model
2023-10-13 08:56:24,735 ----------------------------------------------------------------------------------------------------
2023-10-13 08:56:25,476 epoch 7 - iter 13/138 - loss 0.03028371 - time (sec): 0.73 - samples/sec: 2893.04 - lr: 0.000013 - momentum: 0.000000
2023-10-13 08:56:26,194 epoch 7 - iter 26/138 - loss 0.03618885 - time (sec): 1.45 - samples/sec: 3014.09 - lr: 0.000013 - momentum: 0.000000
2023-10-13 08:56:26,878 epoch 7 - iter 39/138 - loss 0.03348011 - time (sec): 2.14 - samples/sec: 2976.79 - lr: 0.000012 - momentum: 0.000000
2023-10-13 08:56:27,628 epoch 7 - iter 52/138 - loss 0.03659548 - time (sec): 2.89 - samples/sec: 3003.87 - lr: 0.000012 - momentum: 0.000000
2023-10-13 08:56:28,349 epoch 7 - iter 65/138 - loss 0.03402997 - time (sec): 3.61 - samples/sec: 2985.22 - lr: 0.000012 - momentum: 0.000000
2023-10-13 08:56:29,048 epoch 7 - iter 78/138 - loss 0.03277293 - time (sec): 4.31 - samples/sec: 2950.15 - lr: 0.000012 - momentum: 0.000000
2023-10-13 08:56:29,733 epoch 7 - iter 91/138 - loss 0.02973205 - time (sec): 4.99 - samples/sec: 2960.37 - lr: 0.000011 - momentum: 0.000000
2023-10-13 08:56:30,424 epoch 7 - iter 104/138 - loss 0.03177800 - time (sec): 5.68 - samples/sec: 2961.55 - lr: 0.000011 - momentum: 0.000000
2023-10-13 08:56:31,156 epoch 7 - iter 117/138 - loss 0.03571944 - time (sec): 6.41 - samples/sec: 2945.82 - lr: 0.000011 - momentum: 0.000000
2023-10-13 08:56:31,861 epoch 7 - iter 130/138 - loss 0.03317145 - time (sec): 7.12 - samples/sec: 2995.83 - lr: 0.000010 - momentum: 0.000000
2023-10-13 08:56:32,346 ----------------------------------------------------------------------------------------------------
2023-10-13 08:56:32,347 EPOCH 7 done: loss 0.0344 - lr: 0.000010
2023-10-13 08:56:33,038 DEV : loss 0.16303503513336182 - f1-score (micro avg) 0.869
2023-10-13 08:56:33,043 saving best model
2023-10-13 08:56:33,506 ----------------------------------------------------------------------------------------------------
2023-10-13 08:56:34,208 epoch 8 - iter 13/138 - loss 0.04174621 - time (sec): 0.70 - samples/sec: 3112.07 - lr: 0.000010 - momentum: 0.000000
2023-10-13 08:56:34,943 epoch 8 - iter 26/138 - loss 0.03564655 - time (sec): 1.44 - samples/sec: 3135.33 - lr: 0.000009 - momentum: 0.000000
2023-10-13 08:56:35,690 epoch 8 - iter 39/138 - loss 0.03078857 - time (sec): 2.18 - samples/sec: 3087.36 - lr: 0.000009 - momentum: 0.000000
2023-10-13 08:56:36,438 epoch 8 - iter 52/138 - loss 0.02824037 - time (sec): 2.93 - samples/sec: 2999.92 - lr: 0.000009 - momentum: 0.000000
2023-10-13 08:56:37,114 epoch 8 - iter 65/138 - loss 0.03416201 - time (sec): 3.61 - samples/sec: 2976.52 - lr: 0.000009 - momentum: 0.000000
2023-10-13 08:56:37,809 epoch 8 - iter 78/138 - loss 0.03021071 - time (sec): 4.30 - samples/sec: 3001.05 - lr: 0.000008 - momentum: 0.000000
2023-10-13 08:56:38,523 epoch 8 - iter 91/138 - loss 0.02979412 - time (sec): 5.02 - samples/sec: 2951.07 - lr: 0.000008 - momentum: 0.000000
2023-10-13 08:56:39,292 epoch 8 - iter 104/138 - loss 0.03054127 - time (sec): 5.78 - samples/sec: 2948.92 - lr: 0.000008 - momentum: 0.000000
2023-10-13 08:56:39,952 epoch 8 - iter 117/138 - loss 0.02900375 - time (sec): 6.44 - samples/sec: 2961.97 - lr: 0.000007 - momentum: 0.000000
2023-10-13 08:56:40,794 epoch 8 - iter 130/138 - loss 0.02735633 - time (sec): 7.29 - samples/sec: 2942.14 - lr: 0.000007 - momentum: 0.000000
2023-10-13 08:56:41,285 ----------------------------------------------------------------------------------------------------
2023-10-13 08:56:41,285 EPOCH 8 done: loss 0.0269 - lr: 0.000007
2023-10-13 08:56:41,985 DEV : loss 0.16464966535568237 - f1-score (micro avg) 0.8653
2023-10-13 08:56:41,992 ----------------------------------------------------------------------------------------------------
2023-10-13 08:56:42,836 epoch 9 - iter 13/138 - loss 0.01084989 - time (sec): 0.84 - samples/sec: 2749.55 - lr: 0.000006 - momentum: 0.000000
2023-10-13 08:56:43,716 epoch 9 - iter 26/138 - loss 0.01287242 - time (sec): 1.72 - samples/sec: 2578.66 - lr: 0.000006 - momentum: 0.000000
2023-10-13 08:56:44,506 epoch 9 - iter 39/138 - loss 0.00954964 - time (sec): 2.51 - samples/sec: 2576.09 - lr: 0.000006 - momentum: 0.000000
2023-10-13 08:56:45,286 epoch 9 - iter 52/138 - loss 0.01496470 - time (sec): 3.29 - samples/sec: 2656.45 - lr: 0.000005 - momentum: 0.000000
2023-10-13 08:56:46,055 epoch 9 - iter 65/138 - loss 0.02251741 - time (sec): 4.06 - samples/sec: 2712.53 - lr: 0.000005 - momentum: 0.000000
2023-10-13 08:56:46,816 epoch 9 - iter 78/138 - loss 0.02222717 - time (sec): 4.82 - samples/sec: 2748.33 - lr: 0.000005 - momentum: 0.000000
2023-10-13 08:56:47,576 epoch 9 - iter 91/138 - loss 0.02364398 - time (sec): 5.58 - samples/sec: 2750.28 - lr: 0.000005 - momentum: 0.000000
2023-10-13 08:56:48,361 epoch 9 - iter 104/138 - loss 0.02214873 - time (sec): 6.37 - samples/sec: 2747.78 - lr: 0.000004 - momentum: 0.000000
2023-10-13 08:56:49,171 epoch 9 - iter 117/138 - loss 0.02128482 - time (sec): 7.18 - samples/sec: 2744.24 - lr: 0.000004 - momentum: 0.000000
2023-10-13 08:56:49,896 epoch 9 - iter 130/138 - loss 0.01985708 - time (sec): 7.90 - samples/sec: 2736.81 - lr: 0.000004 - momentum: 0.000000
2023-10-13 08:56:50,363 ----------------------------------------------------------------------------------------------------
2023-10-13 08:56:50,364 EPOCH 9 done: loss 0.0213 - lr: 0.000004
2023-10-13 08:56:51,039 DEV : loss 0.157485231757164 - f1-score (micro avg) 0.878
2023-10-13 08:56:51,044 saving best model
2023-10-13 08:56:51,521 ----------------------------------------------------------------------------------------------------
2023-10-13 08:56:52,311 epoch 10 - iter 13/138 - loss 0.03842880 - time (sec): 0.79 - samples/sec: 2703.78 - lr: 0.000003 - momentum: 0.000000
2023-10-13 08:56:53,031 epoch 10 - iter 26/138 - loss 0.03642163 - time (sec): 1.50 - samples/sec: 2895.86 - lr: 0.000003 - momentum: 0.000000
2023-10-13 08:56:53,800 epoch 10 - iter 39/138 - loss 0.03099660 - time (sec): 2.27 - samples/sec: 2874.39 - lr: 0.000002 - momentum: 0.000000
2023-10-13 08:56:54,572 epoch 10 - iter 52/138 - loss 0.02679781 - time (sec): 3.05 - samples/sec: 2869.85 - lr: 0.000002 - momentum: 0.000000
2023-10-13 08:56:55,332 epoch 10 - iter 65/138 - loss 0.02419200 - time (sec): 3.81 - samples/sec: 2883.76 - lr: 0.000002 - momentum: 0.000000
2023-10-13 08:56:56,119 epoch 10 - iter 78/138 - loss 0.02172226 - time (sec): 4.59 - samples/sec: 2870.91 - lr: 0.000002 - momentum: 0.000000
2023-10-13 08:56:56,817 epoch 10 - iter 91/138 - loss 0.02109726 - time (sec): 5.29 - samples/sec: 2881.57 - lr: 0.000001 - momentum: 0.000000
2023-10-13 08:56:57,564 epoch 10 - iter 104/138 - loss 0.02084847 - time (sec): 6.04 - samples/sec: 2899.55 - lr: 0.000001 - momentum: 0.000000
2023-10-13 08:56:58,282 epoch 10 - iter 117/138 - loss 0.02014893 - time (sec): 6.76 - samples/sec: 2904.60 - lr: 0.000001 - momentum: 0.000000
2023-10-13 08:56:58,985 epoch 10 - iter 130/138 - loss 0.01966590 - time (sec): 7.46 - samples/sec: 2907.61 - lr: 0.000000 - momentum: 0.000000
2023-10-13 08:56:59,403 ----------------------------------------------------------------------------------------------------
2023-10-13 08:56:59,404 EPOCH 10 done: loss 0.0194 - lr: 0.000000
2023-10-13 08:57:00,100 DEV : loss 0.15935906767845154 - f1-score (micro avg) 0.8809
2023-10-13 08:57:00,105 saving best model
2023-10-13 08:57:00,891 ----------------------------------------------------------------------------------------------------
2023-10-13 08:57:00,893 Loading model from best epoch ...
2023-10-13 08:57:02,434 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 08:57:03,219
Results:
- F-score (micro) 0.9174
- F-score (macro) 0.7519
- Accuracy 0.8537
By class:
precision recall f1-score support
scope 0.8876 0.8977 0.8927 176
pers 0.9688 0.9688 0.9688 128
work 0.9041 0.8919 0.8980 74
loc 1.0000 1.0000 1.0000 2
object 0.0000 0.0000 0.0000 2
micro avg 0.9186 0.9162 0.9174 382
macro avg 0.7521 0.7517 0.7519 382
weighted avg 0.9139 0.9162 0.9151 382
2023-10-13 08:57:03,219 ----------------------------------------------------------------------------------------------------
|