File size: 23,843 Bytes
43caa5d |
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 |
2023-10-17 17:51:15,725 ----------------------------------------------------------------------------------------------------
2023-10-17 17:51:15,726 Model: "SequenceTagger(
(embeddings): TransformerWordEmbeddings(
(model): ElectraModel(
(embeddings): ElectraEmbeddings(
(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): ElectraEncoder(
(layer): ModuleList(
(0-11): 12 x ElectraLayer(
(attention): ElectraAttention(
(self): ElectraSelfAttention(
(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): ElectraSelfOutput(
(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): ElectraIntermediate(
(dense): Linear(in_features=768, out_features=3072, bias=True)
(intermediate_act_fn): GELUActivation()
)
(output): ElectraOutput(
(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)
)
)
)
)
)
)
(locked_dropout): LockedDropout(p=0.5)
(linear): Linear(in_features=768, out_features=17, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-17 17:51:15,726 ----------------------------------------------------------------------------------------------------
2023-10-17 17:51:15,727 MultiCorpus: 1166 train + 165 dev + 415 test sentences
- NER_HIPE_2022 Corpus: 1166 train + 165 dev + 415 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/newseye/fi/with_doc_seperator
2023-10-17 17:51:15,727 ----------------------------------------------------------------------------------------------------
2023-10-17 17:51:15,727 Train: 1166 sentences
2023-10-17 17:51:15,727 (train_with_dev=False, train_with_test=False)
2023-10-17 17:51:15,727 ----------------------------------------------------------------------------------------------------
2023-10-17 17:51:15,727 Training Params:
2023-10-17 17:51:15,727 - learning_rate: "5e-05"
2023-10-17 17:51:15,727 - mini_batch_size: "8"
2023-10-17 17:51:15,727 - max_epochs: "10"
2023-10-17 17:51:15,727 - shuffle: "True"
2023-10-17 17:51:15,727 ----------------------------------------------------------------------------------------------------
2023-10-17 17:51:15,727 Plugins:
2023-10-17 17:51:15,727 - TensorboardLogger
2023-10-17 17:51:15,727 - LinearScheduler | warmup_fraction: '0.1'
2023-10-17 17:51:15,727 ----------------------------------------------------------------------------------------------------
2023-10-17 17:51:15,727 Final evaluation on model from best epoch (best-model.pt)
2023-10-17 17:51:15,727 - metric: "('micro avg', 'f1-score')"
2023-10-17 17:51:15,727 ----------------------------------------------------------------------------------------------------
2023-10-17 17:51:15,727 Computation:
2023-10-17 17:51:15,727 - compute on device: cuda:0
2023-10-17 17:51:15,727 - embedding storage: none
2023-10-17 17:51:15,727 ----------------------------------------------------------------------------------------------------
2023-10-17 17:51:15,727 Model training base path: "hmbench-newseye/fi-hmteams/teams-base-historic-multilingual-discriminator-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-2"
2023-10-17 17:51:15,727 ----------------------------------------------------------------------------------------------------
2023-10-17 17:51:15,727 ----------------------------------------------------------------------------------------------------
2023-10-17 17:51:15,728 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-17 17:51:17,221 epoch 1 - iter 14/146 - loss 3.64932727 - time (sec): 1.49 - samples/sec: 2867.39 - lr: 0.000004 - momentum: 0.000000
2023-10-17 17:51:18,616 epoch 1 - iter 28/146 - loss 3.24147358 - time (sec): 2.89 - samples/sec: 3050.04 - lr: 0.000009 - momentum: 0.000000
2023-10-17 17:51:20,366 epoch 1 - iter 42/146 - loss 2.52637762 - time (sec): 4.64 - samples/sec: 2820.99 - lr: 0.000014 - momentum: 0.000000
2023-10-17 17:51:21,662 epoch 1 - iter 56/146 - loss 2.07805214 - time (sec): 5.93 - samples/sec: 2843.80 - lr: 0.000019 - momentum: 0.000000
2023-10-17 17:51:22,860 epoch 1 - iter 70/146 - loss 1.82936614 - time (sec): 7.13 - samples/sec: 2864.60 - lr: 0.000024 - momentum: 0.000000
2023-10-17 17:51:24,170 epoch 1 - iter 84/146 - loss 1.60733726 - time (sec): 8.44 - samples/sec: 2875.49 - lr: 0.000028 - momentum: 0.000000
2023-10-17 17:51:25,653 epoch 1 - iter 98/146 - loss 1.41652938 - time (sec): 9.92 - samples/sec: 2901.33 - lr: 0.000033 - momentum: 0.000000
2023-10-17 17:51:26,915 epoch 1 - iter 112/146 - loss 1.29603818 - time (sec): 11.19 - samples/sec: 2910.85 - lr: 0.000038 - momentum: 0.000000
2023-10-17 17:51:28,301 epoch 1 - iter 126/146 - loss 1.18560723 - time (sec): 12.57 - samples/sec: 2914.11 - lr: 0.000043 - momentum: 0.000000
2023-10-17 17:51:30,183 epoch 1 - iter 140/146 - loss 1.07325034 - time (sec): 14.45 - samples/sec: 2922.08 - lr: 0.000048 - momentum: 0.000000
2023-10-17 17:51:31,007 ----------------------------------------------------------------------------------------------------
2023-10-17 17:51:31,008 EPOCH 1 done: loss 1.0409 - lr: 0.000048
2023-10-17 17:51:31,833 DEV : loss 0.19442486763000488 - f1-score (micro avg) 0.4893
2023-10-17 17:51:31,838 saving best model
2023-10-17 17:51:32,169 ----------------------------------------------------------------------------------------------------
2023-10-17 17:51:33,822 epoch 2 - iter 14/146 - loss 0.28292930 - time (sec): 1.65 - samples/sec: 2889.28 - lr: 0.000050 - momentum: 0.000000
2023-10-17 17:51:35,067 epoch 2 - iter 28/146 - loss 0.25593623 - time (sec): 2.90 - samples/sec: 2924.31 - lr: 0.000049 - momentum: 0.000000
2023-10-17 17:51:36,271 epoch 2 - iter 42/146 - loss 0.24056504 - time (sec): 4.10 - samples/sec: 2973.51 - lr: 0.000048 - momentum: 0.000000
2023-10-17 17:51:37,400 epoch 2 - iter 56/146 - loss 0.24098347 - time (sec): 5.23 - samples/sec: 3023.55 - lr: 0.000048 - momentum: 0.000000
2023-10-17 17:51:39,011 epoch 2 - iter 70/146 - loss 0.23990925 - time (sec): 6.84 - samples/sec: 3004.42 - lr: 0.000047 - momentum: 0.000000
2023-10-17 17:51:40,625 epoch 2 - iter 84/146 - loss 0.22035000 - time (sec): 8.46 - samples/sec: 2931.59 - lr: 0.000047 - momentum: 0.000000
2023-10-17 17:51:42,021 epoch 2 - iter 98/146 - loss 0.20717226 - time (sec): 9.85 - samples/sec: 2902.48 - lr: 0.000046 - momentum: 0.000000
2023-10-17 17:51:43,290 epoch 2 - iter 112/146 - loss 0.20187917 - time (sec): 11.12 - samples/sec: 2913.22 - lr: 0.000046 - momentum: 0.000000
2023-10-17 17:51:44,685 epoch 2 - iter 126/146 - loss 0.19646511 - time (sec): 12.52 - samples/sec: 2935.37 - lr: 0.000045 - momentum: 0.000000
2023-10-17 17:51:46,635 epoch 2 - iter 140/146 - loss 0.18970816 - time (sec): 14.46 - samples/sec: 2944.06 - lr: 0.000045 - momentum: 0.000000
2023-10-17 17:51:47,195 ----------------------------------------------------------------------------------------------------
2023-10-17 17:51:47,195 EPOCH 2 done: loss 0.1871 - lr: 0.000045
2023-10-17 17:51:48,684 DEV : loss 0.12840227782726288 - f1-score (micro avg) 0.6597
2023-10-17 17:51:48,690 saving best model
2023-10-17 17:51:49,150 ----------------------------------------------------------------------------------------------------
2023-10-17 17:51:50,352 epoch 3 - iter 14/146 - loss 0.12172571 - time (sec): 1.20 - samples/sec: 2897.39 - lr: 0.000044 - momentum: 0.000000
2023-10-17 17:51:52,174 epoch 3 - iter 28/146 - loss 0.09513845 - time (sec): 3.02 - samples/sec: 2796.32 - lr: 0.000043 - momentum: 0.000000
2023-10-17 17:51:53,613 epoch 3 - iter 42/146 - loss 0.09306659 - time (sec): 4.46 - samples/sec: 2847.48 - lr: 0.000043 - momentum: 0.000000
2023-10-17 17:51:55,371 epoch 3 - iter 56/146 - loss 0.08892443 - time (sec): 6.22 - samples/sec: 2801.67 - lr: 0.000042 - momentum: 0.000000
2023-10-17 17:51:56,653 epoch 3 - iter 70/146 - loss 0.09802952 - time (sec): 7.50 - samples/sec: 2817.17 - lr: 0.000042 - momentum: 0.000000
2023-10-17 17:51:58,035 epoch 3 - iter 84/146 - loss 0.09938552 - time (sec): 8.88 - samples/sec: 2849.02 - lr: 0.000041 - momentum: 0.000000
2023-10-17 17:51:59,244 epoch 3 - iter 98/146 - loss 0.09932542 - time (sec): 10.09 - samples/sec: 2830.08 - lr: 0.000041 - momentum: 0.000000
2023-10-17 17:52:00,940 epoch 3 - iter 112/146 - loss 0.09988457 - time (sec): 11.79 - samples/sec: 2851.55 - lr: 0.000040 - momentum: 0.000000
2023-10-17 17:52:02,529 epoch 3 - iter 126/146 - loss 0.09937404 - time (sec): 13.38 - samples/sec: 2840.97 - lr: 0.000040 - momentum: 0.000000
2023-10-17 17:52:04,197 epoch 3 - iter 140/146 - loss 0.10083464 - time (sec): 15.04 - samples/sec: 2855.29 - lr: 0.000039 - momentum: 0.000000
2023-10-17 17:52:04,661 ----------------------------------------------------------------------------------------------------
2023-10-17 17:52:04,661 EPOCH 3 done: loss 0.0986 - lr: 0.000039
2023-10-17 17:52:05,924 DEV : loss 0.11469055712223053 - f1-score (micro avg) 0.7558
2023-10-17 17:52:05,929 saving best model
2023-10-17 17:52:06,381 ----------------------------------------------------------------------------------------------------
2023-10-17 17:52:07,886 epoch 4 - iter 14/146 - loss 0.08001683 - time (sec): 1.50 - samples/sec: 3162.98 - lr: 0.000038 - momentum: 0.000000
2023-10-17 17:52:09,311 epoch 4 - iter 28/146 - loss 0.08313806 - time (sec): 2.92 - samples/sec: 3113.44 - lr: 0.000038 - momentum: 0.000000
2023-10-17 17:52:10,918 epoch 4 - iter 42/146 - loss 0.08932962 - time (sec): 4.53 - samples/sec: 2955.43 - lr: 0.000037 - momentum: 0.000000
2023-10-17 17:52:12,208 epoch 4 - iter 56/146 - loss 0.08311168 - time (sec): 5.82 - samples/sec: 2896.03 - lr: 0.000037 - momentum: 0.000000
2023-10-17 17:52:13,719 epoch 4 - iter 70/146 - loss 0.07749470 - time (sec): 7.33 - samples/sec: 2885.81 - lr: 0.000036 - momentum: 0.000000
2023-10-17 17:52:15,338 epoch 4 - iter 84/146 - loss 0.07439762 - time (sec): 8.95 - samples/sec: 2898.06 - lr: 0.000036 - momentum: 0.000000
2023-10-17 17:52:16,556 epoch 4 - iter 98/146 - loss 0.07128648 - time (sec): 10.17 - samples/sec: 2885.30 - lr: 0.000035 - momentum: 0.000000
2023-10-17 17:52:17,985 epoch 4 - iter 112/146 - loss 0.07040154 - time (sec): 11.60 - samples/sec: 2874.41 - lr: 0.000035 - momentum: 0.000000
2023-10-17 17:52:19,504 epoch 4 - iter 126/146 - loss 0.06725890 - time (sec): 13.12 - samples/sec: 2895.24 - lr: 0.000034 - momentum: 0.000000
2023-10-17 17:52:21,050 epoch 4 - iter 140/146 - loss 0.06467563 - time (sec): 14.66 - samples/sec: 2904.48 - lr: 0.000034 - momentum: 0.000000
2023-10-17 17:52:21,683 ----------------------------------------------------------------------------------------------------
2023-10-17 17:52:21,684 EPOCH 4 done: loss 0.0643 - lr: 0.000034
2023-10-17 17:52:22,973 DEV : loss 0.12082179635763168 - f1-score (micro avg) 0.7424
2023-10-17 17:52:22,978 ----------------------------------------------------------------------------------------------------
2023-10-17 17:52:24,381 epoch 5 - iter 14/146 - loss 0.06422344 - time (sec): 1.40 - samples/sec: 2739.89 - lr: 0.000033 - momentum: 0.000000
2023-10-17 17:52:25,958 epoch 5 - iter 28/146 - loss 0.05442231 - time (sec): 2.98 - samples/sec: 2842.57 - lr: 0.000032 - momentum: 0.000000
2023-10-17 17:52:27,408 epoch 5 - iter 42/146 - loss 0.04507790 - time (sec): 4.43 - samples/sec: 2940.79 - lr: 0.000032 - momentum: 0.000000
2023-10-17 17:52:29,131 epoch 5 - iter 56/146 - loss 0.05002005 - time (sec): 6.15 - samples/sec: 2884.12 - lr: 0.000031 - momentum: 0.000000
2023-10-17 17:52:30,452 epoch 5 - iter 70/146 - loss 0.04635323 - time (sec): 7.47 - samples/sec: 2896.08 - lr: 0.000031 - momentum: 0.000000
2023-10-17 17:52:31,711 epoch 5 - iter 84/146 - loss 0.04594857 - time (sec): 8.73 - samples/sec: 2876.47 - lr: 0.000030 - momentum: 0.000000
2023-10-17 17:52:33,300 epoch 5 - iter 98/146 - loss 0.04185253 - time (sec): 10.32 - samples/sec: 2899.36 - lr: 0.000030 - momentum: 0.000000
2023-10-17 17:52:34,569 epoch 5 - iter 112/146 - loss 0.04051975 - time (sec): 11.59 - samples/sec: 2904.58 - lr: 0.000029 - momentum: 0.000000
2023-10-17 17:52:36,442 epoch 5 - iter 126/146 - loss 0.04226250 - time (sec): 13.46 - samples/sec: 2874.79 - lr: 0.000029 - momentum: 0.000000
2023-10-17 17:52:37,925 epoch 5 - iter 140/146 - loss 0.04401633 - time (sec): 14.95 - samples/sec: 2857.55 - lr: 0.000028 - momentum: 0.000000
2023-10-17 17:52:38,457 ----------------------------------------------------------------------------------------------------
2023-10-17 17:52:38,458 EPOCH 5 done: loss 0.0431 - lr: 0.000028
2023-10-17 17:52:39,713 DEV : loss 0.12378506362438202 - f1-score (micro avg) 0.7638
2023-10-17 17:52:39,718 saving best model
2023-10-17 17:52:40,156 ----------------------------------------------------------------------------------------------------
2023-10-17 17:52:41,483 epoch 6 - iter 14/146 - loss 0.02821174 - time (sec): 1.32 - samples/sec: 2970.80 - lr: 0.000027 - momentum: 0.000000
2023-10-17 17:52:43,424 epoch 6 - iter 28/146 - loss 0.02719467 - time (sec): 3.27 - samples/sec: 2716.97 - lr: 0.000027 - momentum: 0.000000
2023-10-17 17:52:44,685 epoch 6 - iter 42/146 - loss 0.02593663 - time (sec): 4.53 - samples/sec: 2689.02 - lr: 0.000026 - momentum: 0.000000
2023-10-17 17:52:46,264 epoch 6 - iter 56/146 - loss 0.02530266 - time (sec): 6.11 - samples/sec: 2759.35 - lr: 0.000026 - momentum: 0.000000
2023-10-17 17:52:48,033 epoch 6 - iter 70/146 - loss 0.02607955 - time (sec): 7.87 - samples/sec: 2739.94 - lr: 0.000025 - momentum: 0.000000
2023-10-17 17:52:49,466 epoch 6 - iter 84/146 - loss 0.02856308 - time (sec): 9.31 - samples/sec: 2799.68 - lr: 0.000025 - momentum: 0.000000
2023-10-17 17:52:50,941 epoch 6 - iter 98/146 - loss 0.03007988 - time (sec): 10.78 - samples/sec: 2810.94 - lr: 0.000024 - momentum: 0.000000
2023-10-17 17:52:52,315 epoch 6 - iter 112/146 - loss 0.02972773 - time (sec): 12.16 - samples/sec: 2807.81 - lr: 0.000024 - momentum: 0.000000
2023-10-17 17:52:53,786 epoch 6 - iter 126/146 - loss 0.02975220 - time (sec): 13.63 - samples/sec: 2816.61 - lr: 0.000023 - momentum: 0.000000
2023-10-17 17:52:55,150 epoch 6 - iter 140/146 - loss 0.03021626 - time (sec): 14.99 - samples/sec: 2858.97 - lr: 0.000023 - momentum: 0.000000
2023-10-17 17:52:55,649 ----------------------------------------------------------------------------------------------------
2023-10-17 17:52:55,649 EPOCH 6 done: loss 0.0306 - lr: 0.000023
2023-10-17 17:52:56,904 DEV : loss 0.1446683704853058 - f1-score (micro avg) 0.7591
2023-10-17 17:52:56,909 ----------------------------------------------------------------------------------------------------
2023-10-17 17:52:58,252 epoch 7 - iter 14/146 - loss 0.01067881 - time (sec): 1.34 - samples/sec: 2768.15 - lr: 0.000022 - momentum: 0.000000
2023-10-17 17:52:59,748 epoch 7 - iter 28/146 - loss 0.01553625 - time (sec): 2.84 - samples/sec: 2972.30 - lr: 0.000021 - momentum: 0.000000
2023-10-17 17:53:01,210 epoch 7 - iter 42/146 - loss 0.02338274 - time (sec): 4.30 - samples/sec: 3013.27 - lr: 0.000021 - momentum: 0.000000
2023-10-17 17:53:02,788 epoch 7 - iter 56/146 - loss 0.02276518 - time (sec): 5.88 - samples/sec: 2958.23 - lr: 0.000020 - momentum: 0.000000
2023-10-17 17:53:04,135 epoch 7 - iter 70/146 - loss 0.02040107 - time (sec): 7.22 - samples/sec: 2979.45 - lr: 0.000020 - momentum: 0.000000
2023-10-17 17:53:05,707 epoch 7 - iter 84/146 - loss 0.01887587 - time (sec): 8.80 - samples/sec: 2899.23 - lr: 0.000019 - momentum: 0.000000
2023-10-17 17:53:07,090 epoch 7 - iter 98/146 - loss 0.02062489 - time (sec): 10.18 - samples/sec: 2927.81 - lr: 0.000019 - momentum: 0.000000
2023-10-17 17:53:08,404 epoch 7 - iter 112/146 - loss 0.02270664 - time (sec): 11.49 - samples/sec: 2967.68 - lr: 0.000018 - momentum: 0.000000
2023-10-17 17:53:10,026 epoch 7 - iter 126/146 - loss 0.02108995 - time (sec): 13.12 - samples/sec: 2940.26 - lr: 0.000018 - momentum: 0.000000
2023-10-17 17:53:11,723 epoch 7 - iter 140/146 - loss 0.02146590 - time (sec): 14.81 - samples/sec: 2904.13 - lr: 0.000017 - momentum: 0.000000
2023-10-17 17:53:12,214 ----------------------------------------------------------------------------------------------------
2023-10-17 17:53:12,214 EPOCH 7 done: loss 0.0221 - lr: 0.000017
2023-10-17 17:53:13,493 DEV : loss 0.13926248252391815 - f1-score (micro avg) 0.7588
2023-10-17 17:53:13,498 ----------------------------------------------------------------------------------------------------
2023-10-17 17:53:14,805 epoch 8 - iter 14/146 - loss 0.04199538 - time (sec): 1.31 - samples/sec: 2869.25 - lr: 0.000016 - momentum: 0.000000
2023-10-17 17:53:16,360 epoch 8 - iter 28/146 - loss 0.03623353 - time (sec): 2.86 - samples/sec: 2858.33 - lr: 0.000016 - momentum: 0.000000
2023-10-17 17:53:17,765 epoch 8 - iter 42/146 - loss 0.02909497 - time (sec): 4.27 - samples/sec: 2789.70 - lr: 0.000015 - momentum: 0.000000
2023-10-17 17:53:19,130 epoch 8 - iter 56/146 - loss 0.02582501 - time (sec): 5.63 - samples/sec: 2822.08 - lr: 0.000015 - momentum: 0.000000
2023-10-17 17:53:20,472 epoch 8 - iter 70/146 - loss 0.02248507 - time (sec): 6.97 - samples/sec: 2905.70 - lr: 0.000014 - momentum: 0.000000
2023-10-17 17:53:21,992 epoch 8 - iter 84/146 - loss 0.02039668 - time (sec): 8.49 - samples/sec: 2948.75 - lr: 0.000014 - momentum: 0.000000
2023-10-17 17:53:23,243 epoch 8 - iter 98/146 - loss 0.02065091 - time (sec): 9.74 - samples/sec: 2953.56 - lr: 0.000013 - momentum: 0.000000
2023-10-17 17:53:24,793 epoch 8 - iter 112/146 - loss 0.01976567 - time (sec): 11.29 - samples/sec: 2967.93 - lr: 0.000013 - momentum: 0.000000
2023-10-17 17:53:26,152 epoch 8 - iter 126/146 - loss 0.01878112 - time (sec): 12.65 - samples/sec: 2964.08 - lr: 0.000012 - momentum: 0.000000
2023-10-17 17:53:27,832 epoch 8 - iter 140/146 - loss 0.01820202 - time (sec): 14.33 - samples/sec: 2977.10 - lr: 0.000012 - momentum: 0.000000
2023-10-17 17:53:28,544 ----------------------------------------------------------------------------------------------------
2023-10-17 17:53:28,544 EPOCH 8 done: loss 0.0180 - lr: 0.000012
2023-10-17 17:53:29,816 DEV : loss 0.1338219791650772 - f1-score (micro avg) 0.8125
2023-10-17 17:53:29,821 saving best model
2023-10-17 17:53:30,265 ----------------------------------------------------------------------------------------------------
2023-10-17 17:53:31,690 epoch 9 - iter 14/146 - loss 0.00900173 - time (sec): 1.42 - samples/sec: 3146.38 - lr: 0.000011 - momentum: 0.000000
2023-10-17 17:53:33,172 epoch 9 - iter 28/146 - loss 0.01353775 - time (sec): 2.91 - samples/sec: 2969.76 - lr: 0.000010 - momentum: 0.000000
2023-10-17 17:53:34,800 epoch 9 - iter 42/146 - loss 0.01411638 - time (sec): 4.53 - samples/sec: 2925.56 - lr: 0.000010 - momentum: 0.000000
2023-10-17 17:53:36,570 epoch 9 - iter 56/146 - loss 0.01717406 - time (sec): 6.30 - samples/sec: 2832.55 - lr: 0.000009 - momentum: 0.000000
2023-10-17 17:53:38,322 epoch 9 - iter 70/146 - loss 0.01651724 - time (sec): 8.06 - samples/sec: 2779.14 - lr: 0.000009 - momentum: 0.000000
2023-10-17 17:53:39,653 epoch 9 - iter 84/146 - loss 0.01447022 - time (sec): 9.39 - samples/sec: 2809.42 - lr: 0.000008 - momentum: 0.000000
2023-10-17 17:53:41,289 epoch 9 - iter 98/146 - loss 0.01305284 - time (sec): 11.02 - samples/sec: 2793.58 - lr: 0.000008 - momentum: 0.000000
2023-10-17 17:53:42,484 epoch 9 - iter 112/146 - loss 0.01276152 - time (sec): 12.22 - samples/sec: 2817.73 - lr: 0.000007 - momentum: 0.000000
2023-10-17 17:53:43,877 epoch 9 - iter 126/146 - loss 0.01240197 - time (sec): 13.61 - samples/sec: 2825.26 - lr: 0.000007 - momentum: 0.000000
2023-10-17 17:53:45,221 epoch 9 - iter 140/146 - loss 0.01247661 - time (sec): 14.95 - samples/sec: 2864.35 - lr: 0.000006 - momentum: 0.000000
2023-10-17 17:53:45,888 ----------------------------------------------------------------------------------------------------
2023-10-17 17:53:45,888 EPOCH 9 done: loss 0.0125 - lr: 0.000006
2023-10-17 17:53:47,153 DEV : loss 0.1447986215353012 - f1-score (micro avg) 0.7973
2023-10-17 17:53:47,157 ----------------------------------------------------------------------------------------------------
2023-10-17 17:53:48,470 epoch 10 - iter 14/146 - loss 0.01128755 - time (sec): 1.31 - samples/sec: 3088.10 - lr: 0.000005 - momentum: 0.000000
2023-10-17 17:53:49,879 epoch 10 - iter 28/146 - loss 0.01029360 - time (sec): 2.72 - samples/sec: 2992.37 - lr: 0.000005 - momentum: 0.000000
2023-10-17 17:53:51,519 epoch 10 - iter 42/146 - loss 0.01240356 - time (sec): 4.36 - samples/sec: 2861.25 - lr: 0.000004 - momentum: 0.000000
2023-10-17 17:53:52,894 epoch 10 - iter 56/146 - loss 0.01088444 - time (sec): 5.74 - samples/sec: 2944.20 - lr: 0.000004 - momentum: 0.000000
2023-10-17 17:53:54,438 epoch 10 - iter 70/146 - loss 0.01024052 - time (sec): 7.28 - samples/sec: 2969.04 - lr: 0.000003 - momentum: 0.000000
2023-10-17 17:53:56,006 epoch 10 - iter 84/146 - loss 0.00903528 - time (sec): 8.85 - samples/sec: 2930.05 - lr: 0.000003 - momentum: 0.000000
2023-10-17 17:53:57,751 epoch 10 - iter 98/146 - loss 0.01043615 - time (sec): 10.59 - samples/sec: 2857.32 - lr: 0.000002 - momentum: 0.000000
2023-10-17 17:53:59,290 epoch 10 - iter 112/146 - loss 0.01019047 - time (sec): 12.13 - samples/sec: 2865.50 - lr: 0.000002 - momentum: 0.000000
2023-10-17 17:54:00,771 epoch 10 - iter 126/146 - loss 0.00966469 - time (sec): 13.61 - samples/sec: 2864.12 - lr: 0.000001 - momentum: 0.000000
2023-10-17 17:54:02,058 epoch 10 - iter 140/146 - loss 0.01137448 - time (sec): 14.90 - samples/sec: 2865.91 - lr: 0.000000 - momentum: 0.000000
2023-10-17 17:54:02,617 ----------------------------------------------------------------------------------------------------
2023-10-17 17:54:02,617 EPOCH 10 done: loss 0.0111 - lr: 0.000000
2023-10-17 17:54:03,903 DEV : loss 0.14230762422084808 - f1-score (micro avg) 0.8044
2023-10-17 17:54:04,247 ----------------------------------------------------------------------------------------------------
2023-10-17 17:54:04,249 Loading model from best epoch ...
2023-10-17 17:54:05,633 SequenceTagger predicts: Dictionary with 17 tags: O, S-LOC, B-LOC, E-LOC, I-LOC, S-PER, B-PER, E-PER, I-PER, S-ORG, B-ORG, E-ORG, I-ORG, S-HumanProd, B-HumanProd, E-HumanProd, I-HumanProd
2023-10-17 17:54:08,070
Results:
- F-score (micro) 0.7704
- F-score (macro) 0.7203
- Accuracy 0.6484
By class:
precision recall f1-score support
PER 0.8179 0.8649 0.8408 348
LOC 0.6503 0.8123 0.7223 261
ORG 0.5745 0.5192 0.5455 52
HumanProd 0.7727 0.7727 0.7727 22
micro avg 0.7300 0.8155 0.7704 683
macro avg 0.7039 0.7423 0.7203 683
weighted avg 0.7339 0.8155 0.7708 683
2023-10-17 17:54:08,070 ----------------------------------------------------------------------------------------------------
|