File size: 23,912 Bytes
0485b3b |
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 |
2023-10-17 08:53:34,961 ----------------------------------------------------------------------------------------------------
2023-10-17 08:53:34,962 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=25, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-17 08:53:34,962 ----------------------------------------------------------------------------------------------------
2023-10-17 08:53:34,963 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
2023-10-17 08:53:34,963 ----------------------------------------------------------------------------------------------------
2023-10-17 08:53:34,963 Train: 1100 sentences
2023-10-17 08:53:34,963 (train_with_dev=False, train_with_test=False)
2023-10-17 08:53:34,963 ----------------------------------------------------------------------------------------------------
2023-10-17 08:53:34,963 Training Params:
2023-10-17 08:53:34,963 - learning_rate: "5e-05"
2023-10-17 08:53:34,963 - mini_batch_size: "8"
2023-10-17 08:53:34,963 - max_epochs: "10"
2023-10-17 08:53:34,963 - shuffle: "True"
2023-10-17 08:53:34,963 ----------------------------------------------------------------------------------------------------
2023-10-17 08:53:34,963 Plugins:
2023-10-17 08:53:34,963 - TensorboardLogger
2023-10-17 08:53:34,963 - LinearScheduler | warmup_fraction: '0.1'
2023-10-17 08:53:34,963 ----------------------------------------------------------------------------------------------------
2023-10-17 08:53:34,963 Final evaluation on model from best epoch (best-model.pt)
2023-10-17 08:53:34,963 - metric: "('micro avg', 'f1-score')"
2023-10-17 08:53:34,963 ----------------------------------------------------------------------------------------------------
2023-10-17 08:53:34,963 Computation:
2023-10-17 08:53:34,963 - compute on device: cuda:0
2023-10-17 08:53:34,963 - embedding storage: none
2023-10-17 08:53:34,963 ----------------------------------------------------------------------------------------------------
2023-10-17 08:53:34,963 Model training base path: "hmbench-ajmc/de-hmteams/teams-base-historic-multilingual-discriminator-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-4"
2023-10-17 08:53:34,963 ----------------------------------------------------------------------------------------------------
2023-10-17 08:53:34,963 ----------------------------------------------------------------------------------------------------
2023-10-17 08:53:34,963 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-17 08:53:35,685 epoch 1 - iter 13/138 - loss 4.06947742 - time (sec): 0.72 - samples/sec: 3084.80 - lr: 0.000004 - momentum: 0.000000
2023-10-17 08:53:36,373 epoch 1 - iter 26/138 - loss 3.53460833 - time (sec): 1.41 - samples/sec: 2884.84 - lr: 0.000009 - momentum: 0.000000
2023-10-17 08:53:37,113 epoch 1 - iter 39/138 - loss 2.81906985 - time (sec): 2.15 - samples/sec: 2934.39 - lr: 0.000014 - momentum: 0.000000
2023-10-17 08:53:37,859 epoch 1 - iter 52/138 - loss 2.32739470 - time (sec): 2.89 - samples/sec: 2912.07 - lr: 0.000018 - momentum: 0.000000
2023-10-17 08:53:38,624 epoch 1 - iter 65/138 - loss 1.97469116 - time (sec): 3.66 - samples/sec: 2879.38 - lr: 0.000023 - momentum: 0.000000
2023-10-17 08:53:39,364 epoch 1 - iter 78/138 - loss 1.72761590 - time (sec): 4.40 - samples/sec: 2869.36 - lr: 0.000028 - momentum: 0.000000
2023-10-17 08:53:40,121 epoch 1 - iter 91/138 - loss 1.51466269 - time (sec): 5.16 - samples/sec: 2916.91 - lr: 0.000033 - momentum: 0.000000
2023-10-17 08:53:40,865 epoch 1 - iter 104/138 - loss 1.35593559 - time (sec): 5.90 - samples/sec: 2974.88 - lr: 0.000037 - momentum: 0.000000
2023-10-17 08:53:41,568 epoch 1 - iter 117/138 - loss 1.24744656 - time (sec): 6.60 - samples/sec: 2961.80 - lr: 0.000042 - momentum: 0.000000
2023-10-17 08:53:42,325 epoch 1 - iter 130/138 - loss 1.15825875 - time (sec): 7.36 - samples/sec: 2932.06 - lr: 0.000047 - momentum: 0.000000
2023-10-17 08:53:42,773 ----------------------------------------------------------------------------------------------------
2023-10-17 08:53:42,773 EPOCH 1 done: loss 1.1129 - lr: 0.000047
2023-10-17 08:53:43,555 DEV : loss 0.1926119476556778 - f1-score (micro avg) 0.7766
2023-10-17 08:53:43,560 saving best model
2023-10-17 08:53:43,920 ----------------------------------------------------------------------------------------------------
2023-10-17 08:53:44,655 epoch 2 - iter 13/138 - loss 0.17238555 - time (sec): 0.73 - samples/sec: 2968.02 - lr: 0.000050 - momentum: 0.000000
2023-10-17 08:53:45,359 epoch 2 - iter 26/138 - loss 0.17242125 - time (sec): 1.44 - samples/sec: 2980.23 - lr: 0.000049 - momentum: 0.000000
2023-10-17 08:53:46,110 epoch 2 - iter 39/138 - loss 0.17201537 - time (sec): 2.19 - samples/sec: 2920.07 - lr: 0.000048 - momentum: 0.000000
2023-10-17 08:53:46,864 epoch 2 - iter 52/138 - loss 0.17632595 - time (sec): 2.94 - samples/sec: 2963.47 - lr: 0.000048 - momentum: 0.000000
2023-10-17 08:53:47,592 epoch 2 - iter 65/138 - loss 0.16585204 - time (sec): 3.67 - samples/sec: 2930.80 - lr: 0.000047 - momentum: 0.000000
2023-10-17 08:53:48,358 epoch 2 - iter 78/138 - loss 0.17233931 - time (sec): 4.44 - samples/sec: 2950.18 - lr: 0.000047 - momentum: 0.000000
2023-10-17 08:53:49,132 epoch 2 - iter 91/138 - loss 0.17024609 - time (sec): 5.21 - samples/sec: 2926.88 - lr: 0.000046 - momentum: 0.000000
2023-10-17 08:53:49,959 epoch 2 - iter 104/138 - loss 0.17589124 - time (sec): 6.04 - samples/sec: 2925.32 - lr: 0.000046 - momentum: 0.000000
2023-10-17 08:53:50,690 epoch 2 - iter 117/138 - loss 0.16910689 - time (sec): 6.77 - samples/sec: 2913.93 - lr: 0.000045 - momentum: 0.000000
2023-10-17 08:53:51,426 epoch 2 - iter 130/138 - loss 0.16263917 - time (sec): 7.50 - samples/sec: 2887.16 - lr: 0.000045 - momentum: 0.000000
2023-10-17 08:53:51,877 ----------------------------------------------------------------------------------------------------
2023-10-17 08:53:51,877 EPOCH 2 done: loss 0.1644 - lr: 0.000045
2023-10-17 08:53:52,509 DEV : loss 0.1401386559009552 - f1-score (micro avg) 0.8
2023-10-17 08:53:52,514 saving best model
2023-10-17 08:53:52,950 ----------------------------------------------------------------------------------------------------
2023-10-17 08:53:53,657 epoch 3 - iter 13/138 - loss 0.10839392 - time (sec): 0.71 - samples/sec: 2862.79 - lr: 0.000044 - momentum: 0.000000
2023-10-17 08:53:54,345 epoch 3 - iter 26/138 - loss 0.09821987 - time (sec): 1.39 - samples/sec: 2726.84 - lr: 0.000043 - momentum: 0.000000
2023-10-17 08:53:55,042 epoch 3 - iter 39/138 - loss 0.09920961 - time (sec): 2.09 - samples/sec: 2832.75 - lr: 0.000043 - momentum: 0.000000
2023-10-17 08:53:55,868 epoch 3 - iter 52/138 - loss 0.10149162 - time (sec): 2.92 - samples/sec: 2879.44 - lr: 0.000042 - momentum: 0.000000
2023-10-17 08:53:56,566 epoch 3 - iter 65/138 - loss 0.09871974 - time (sec): 3.61 - samples/sec: 2855.39 - lr: 0.000042 - momentum: 0.000000
2023-10-17 08:53:57,338 epoch 3 - iter 78/138 - loss 0.09599052 - time (sec): 4.39 - samples/sec: 2870.37 - lr: 0.000041 - momentum: 0.000000
2023-10-17 08:53:58,058 epoch 3 - iter 91/138 - loss 0.09023016 - time (sec): 5.11 - samples/sec: 2878.12 - lr: 0.000041 - momentum: 0.000000
2023-10-17 08:53:58,800 epoch 3 - iter 104/138 - loss 0.08954002 - time (sec): 5.85 - samples/sec: 2892.06 - lr: 0.000040 - momentum: 0.000000
2023-10-17 08:53:59,591 epoch 3 - iter 117/138 - loss 0.08970388 - time (sec): 6.64 - samples/sec: 2886.08 - lr: 0.000040 - momentum: 0.000000
2023-10-17 08:54:00,344 epoch 3 - iter 130/138 - loss 0.09656639 - time (sec): 7.39 - samples/sec: 2903.71 - lr: 0.000039 - momentum: 0.000000
2023-10-17 08:54:00,809 ----------------------------------------------------------------------------------------------------
2023-10-17 08:54:00,809 EPOCH 3 done: loss 0.0961 - lr: 0.000039
2023-10-17 08:54:01,436 DEV : loss 0.12580935657024384 - f1-score (micro avg) 0.8492
2023-10-17 08:54:01,440 saving best model
2023-10-17 08:54:01,874 ----------------------------------------------------------------------------------------------------
2023-10-17 08:54:02,587 epoch 4 - iter 13/138 - loss 0.07047690 - time (sec): 0.71 - samples/sec: 2860.23 - lr: 0.000038 - momentum: 0.000000
2023-10-17 08:54:03,311 epoch 4 - iter 26/138 - loss 0.05110513 - time (sec): 1.44 - samples/sec: 2869.51 - lr: 0.000038 - momentum: 0.000000
2023-10-17 08:54:04,036 epoch 4 - iter 39/138 - loss 0.08593546 - time (sec): 2.16 - samples/sec: 2894.35 - lr: 0.000037 - momentum: 0.000000
2023-10-17 08:54:04,775 epoch 4 - iter 52/138 - loss 0.09311636 - time (sec): 2.90 - samples/sec: 2907.89 - lr: 0.000037 - momentum: 0.000000
2023-10-17 08:54:05,531 epoch 4 - iter 65/138 - loss 0.08520031 - time (sec): 3.66 - samples/sec: 2866.50 - lr: 0.000036 - momentum: 0.000000
2023-10-17 08:54:06,289 epoch 4 - iter 78/138 - loss 0.07696350 - time (sec): 4.41 - samples/sec: 2883.48 - lr: 0.000036 - momentum: 0.000000
2023-10-17 08:54:07,056 epoch 4 - iter 91/138 - loss 0.07443197 - time (sec): 5.18 - samples/sec: 2868.28 - lr: 0.000035 - momentum: 0.000000
2023-10-17 08:54:07,813 epoch 4 - iter 104/138 - loss 0.07494235 - time (sec): 5.94 - samples/sec: 2888.26 - lr: 0.000035 - momentum: 0.000000
2023-10-17 08:54:08,578 epoch 4 - iter 117/138 - loss 0.07524586 - time (sec): 6.70 - samples/sec: 2887.19 - lr: 0.000034 - momentum: 0.000000
2023-10-17 08:54:09,360 epoch 4 - iter 130/138 - loss 0.07303091 - time (sec): 7.49 - samples/sec: 2870.51 - lr: 0.000034 - momentum: 0.000000
2023-10-17 08:54:09,846 ----------------------------------------------------------------------------------------------------
2023-10-17 08:54:09,846 EPOCH 4 done: loss 0.0738 - lr: 0.000034
2023-10-17 08:54:10,479 DEV : loss 0.1469777375459671 - f1-score (micro avg) 0.8681
2023-10-17 08:54:10,484 saving best model
2023-10-17 08:54:10,928 ----------------------------------------------------------------------------------------------------
2023-10-17 08:54:11,633 epoch 5 - iter 13/138 - loss 0.09518052 - time (sec): 0.70 - samples/sec: 3280.21 - lr: 0.000033 - momentum: 0.000000
2023-10-17 08:54:12,326 epoch 5 - iter 26/138 - loss 0.06688716 - time (sec): 1.40 - samples/sec: 3186.06 - lr: 0.000032 - momentum: 0.000000
2023-10-17 08:54:13,021 epoch 5 - iter 39/138 - loss 0.05861502 - time (sec): 2.09 - samples/sec: 3059.78 - lr: 0.000032 - momentum: 0.000000
2023-10-17 08:54:13,701 epoch 5 - iter 52/138 - loss 0.06250234 - time (sec): 2.77 - samples/sec: 3000.26 - lr: 0.000031 - momentum: 0.000000
2023-10-17 08:54:14,423 epoch 5 - iter 65/138 - loss 0.06755541 - time (sec): 3.49 - samples/sec: 2995.35 - lr: 0.000031 - momentum: 0.000000
2023-10-17 08:54:15,175 epoch 5 - iter 78/138 - loss 0.06415957 - time (sec): 4.25 - samples/sec: 2977.27 - lr: 0.000030 - momentum: 0.000000
2023-10-17 08:54:15,945 epoch 5 - iter 91/138 - loss 0.06402975 - time (sec): 5.02 - samples/sec: 2950.01 - lr: 0.000030 - momentum: 0.000000
2023-10-17 08:54:16,713 epoch 5 - iter 104/138 - loss 0.05989598 - time (sec): 5.78 - samples/sec: 2965.03 - lr: 0.000029 - momentum: 0.000000
2023-10-17 08:54:17,459 epoch 5 - iter 117/138 - loss 0.05868487 - time (sec): 6.53 - samples/sec: 2961.32 - lr: 0.000029 - momentum: 0.000000
2023-10-17 08:54:18,229 epoch 5 - iter 130/138 - loss 0.05774891 - time (sec): 7.30 - samples/sec: 2950.84 - lr: 0.000028 - momentum: 0.000000
2023-10-17 08:54:18,677 ----------------------------------------------------------------------------------------------------
2023-10-17 08:54:18,678 EPOCH 5 done: loss 0.0555 - lr: 0.000028
2023-10-17 08:54:19,310 DEV : loss 0.16393792629241943 - f1-score (micro avg) 0.8619
2023-10-17 08:54:19,314 ----------------------------------------------------------------------------------------------------
2023-10-17 08:54:20,016 epoch 6 - iter 13/138 - loss 0.03744158 - time (sec): 0.70 - samples/sec: 2748.59 - lr: 0.000027 - momentum: 0.000000
2023-10-17 08:54:20,736 epoch 6 - iter 26/138 - loss 0.04542780 - time (sec): 1.42 - samples/sec: 2836.48 - lr: 0.000027 - momentum: 0.000000
2023-10-17 08:54:21,486 epoch 6 - iter 39/138 - loss 0.04843577 - time (sec): 2.17 - samples/sec: 2839.99 - lr: 0.000026 - momentum: 0.000000
2023-10-17 08:54:22,237 epoch 6 - iter 52/138 - loss 0.03793948 - time (sec): 2.92 - samples/sec: 2830.12 - lr: 0.000026 - momentum: 0.000000
2023-10-17 08:54:22,975 epoch 6 - iter 65/138 - loss 0.04494788 - time (sec): 3.66 - samples/sec: 2886.63 - lr: 0.000025 - momentum: 0.000000
2023-10-17 08:54:23,723 epoch 6 - iter 78/138 - loss 0.03905436 - time (sec): 4.41 - samples/sec: 2887.40 - lr: 0.000025 - momentum: 0.000000
2023-10-17 08:54:24,485 epoch 6 - iter 91/138 - loss 0.03803730 - time (sec): 5.17 - samples/sec: 2861.38 - lr: 0.000024 - momentum: 0.000000
2023-10-17 08:54:25,272 epoch 6 - iter 104/138 - loss 0.04047424 - time (sec): 5.96 - samples/sec: 2851.01 - lr: 0.000024 - momentum: 0.000000
2023-10-17 08:54:26,105 epoch 6 - iter 117/138 - loss 0.04769041 - time (sec): 6.79 - samples/sec: 2850.07 - lr: 0.000023 - momentum: 0.000000
2023-10-17 08:54:26,875 epoch 6 - iter 130/138 - loss 0.04603896 - time (sec): 7.56 - samples/sec: 2860.08 - lr: 0.000023 - momentum: 0.000000
2023-10-17 08:54:27,299 ----------------------------------------------------------------------------------------------------
2023-10-17 08:54:27,300 EPOCH 6 done: loss 0.0444 - lr: 0.000023
2023-10-17 08:54:27,937 DEV : loss 0.1778355985879898 - f1-score (micro avg) 0.867
2023-10-17 08:54:27,941 ----------------------------------------------------------------------------------------------------
2023-10-17 08:54:28,667 epoch 7 - iter 13/138 - loss 0.08203944 - time (sec): 0.72 - samples/sec: 2779.90 - lr: 0.000022 - momentum: 0.000000
2023-10-17 08:54:29,442 epoch 7 - iter 26/138 - loss 0.06638147 - time (sec): 1.50 - samples/sec: 2795.68 - lr: 0.000021 - momentum: 0.000000
2023-10-17 08:54:30,208 epoch 7 - iter 39/138 - loss 0.07442183 - time (sec): 2.27 - samples/sec: 2872.06 - lr: 0.000021 - momentum: 0.000000
2023-10-17 08:54:30,956 epoch 7 - iter 52/138 - loss 0.06170432 - time (sec): 3.01 - samples/sec: 2872.64 - lr: 0.000020 - momentum: 0.000000
2023-10-17 08:54:31,711 epoch 7 - iter 65/138 - loss 0.05092242 - time (sec): 3.77 - samples/sec: 2837.54 - lr: 0.000020 - momentum: 0.000000
2023-10-17 08:54:32,455 epoch 7 - iter 78/138 - loss 0.04529981 - time (sec): 4.51 - samples/sec: 2858.77 - lr: 0.000019 - momentum: 0.000000
2023-10-17 08:54:33,249 epoch 7 - iter 91/138 - loss 0.04041344 - time (sec): 5.31 - samples/sec: 2855.92 - lr: 0.000019 - momentum: 0.000000
2023-10-17 08:54:33,992 epoch 7 - iter 104/138 - loss 0.04050339 - time (sec): 6.05 - samples/sec: 2871.54 - lr: 0.000018 - momentum: 0.000000
2023-10-17 08:54:34,755 epoch 7 - iter 117/138 - loss 0.03924770 - time (sec): 6.81 - samples/sec: 2842.01 - lr: 0.000018 - momentum: 0.000000
2023-10-17 08:54:35,464 epoch 7 - iter 130/138 - loss 0.03944895 - time (sec): 7.52 - samples/sec: 2858.12 - lr: 0.000017 - momentum: 0.000000
2023-10-17 08:54:35,904 ----------------------------------------------------------------------------------------------------
2023-10-17 08:54:35,905 EPOCH 7 done: loss 0.0376 - lr: 0.000017
2023-10-17 08:54:36,571 DEV : loss 0.19011272490024567 - f1-score (micro avg) 0.8633
2023-10-17 08:54:36,575 ----------------------------------------------------------------------------------------------------
2023-10-17 08:54:37,347 epoch 8 - iter 13/138 - loss 0.01406125 - time (sec): 0.77 - samples/sec: 3038.31 - lr: 0.000016 - momentum: 0.000000
2023-10-17 08:54:38,068 epoch 8 - iter 26/138 - loss 0.02028170 - time (sec): 1.49 - samples/sec: 3041.21 - lr: 0.000016 - momentum: 0.000000
2023-10-17 08:54:38,782 epoch 8 - iter 39/138 - loss 0.02305527 - time (sec): 2.21 - samples/sec: 2941.13 - lr: 0.000015 - momentum: 0.000000
2023-10-17 08:54:39,559 epoch 8 - iter 52/138 - loss 0.01949837 - time (sec): 2.98 - samples/sec: 2905.93 - lr: 0.000015 - momentum: 0.000000
2023-10-17 08:54:40,353 epoch 8 - iter 65/138 - loss 0.01753884 - time (sec): 3.78 - samples/sec: 2938.16 - lr: 0.000014 - momentum: 0.000000
2023-10-17 08:54:41,072 epoch 8 - iter 78/138 - loss 0.01859406 - time (sec): 4.50 - samples/sec: 2930.11 - lr: 0.000014 - momentum: 0.000000
2023-10-17 08:54:41,860 epoch 8 - iter 91/138 - loss 0.02017696 - time (sec): 5.28 - samples/sec: 2914.13 - lr: 0.000013 - momentum: 0.000000
2023-10-17 08:54:42,631 epoch 8 - iter 104/138 - loss 0.02936815 - time (sec): 6.06 - samples/sec: 2908.50 - lr: 0.000013 - momentum: 0.000000
2023-10-17 08:54:43,370 epoch 8 - iter 117/138 - loss 0.03072266 - time (sec): 6.79 - samples/sec: 2888.90 - lr: 0.000012 - momentum: 0.000000
2023-10-17 08:54:44,137 epoch 8 - iter 130/138 - loss 0.03009208 - time (sec): 7.56 - samples/sec: 2869.34 - lr: 0.000012 - momentum: 0.000000
2023-10-17 08:54:44,550 ----------------------------------------------------------------------------------------------------
2023-10-17 08:54:44,550 EPOCH 8 done: loss 0.0298 - lr: 0.000012
2023-10-17 08:54:45,191 DEV : loss 0.17324329912662506 - f1-score (micro avg) 0.8726
2023-10-17 08:54:45,197 saving best model
2023-10-17 08:54:45,655 ----------------------------------------------------------------------------------------------------
2023-10-17 08:54:46,459 epoch 9 - iter 13/138 - loss 0.03966138 - time (sec): 0.80 - samples/sec: 2842.12 - lr: 0.000011 - momentum: 0.000000
2023-10-17 08:54:47,277 epoch 9 - iter 26/138 - loss 0.02599233 - time (sec): 1.62 - samples/sec: 2846.41 - lr: 0.000010 - momentum: 0.000000
2023-10-17 08:54:48,009 epoch 9 - iter 39/138 - loss 0.02395130 - time (sec): 2.35 - samples/sec: 2821.06 - lr: 0.000010 - momentum: 0.000000
2023-10-17 08:54:48,740 epoch 9 - iter 52/138 - loss 0.02886960 - time (sec): 3.08 - samples/sec: 2828.78 - lr: 0.000009 - momentum: 0.000000
2023-10-17 08:54:49,446 epoch 9 - iter 65/138 - loss 0.02820047 - time (sec): 3.79 - samples/sec: 2779.66 - lr: 0.000009 - momentum: 0.000000
2023-10-17 08:54:50,180 epoch 9 - iter 78/138 - loss 0.03240256 - time (sec): 4.52 - samples/sec: 2843.95 - lr: 0.000008 - momentum: 0.000000
2023-10-17 08:54:50,889 epoch 9 - iter 91/138 - loss 0.03208418 - time (sec): 5.23 - samples/sec: 2861.50 - lr: 0.000008 - momentum: 0.000000
2023-10-17 08:54:51,651 epoch 9 - iter 104/138 - loss 0.02859865 - time (sec): 5.99 - samples/sec: 2838.39 - lr: 0.000007 - momentum: 0.000000
2023-10-17 08:54:52,406 epoch 9 - iter 117/138 - loss 0.02625805 - time (sec): 6.75 - samples/sec: 2840.41 - lr: 0.000007 - momentum: 0.000000
2023-10-17 08:54:53,128 epoch 9 - iter 130/138 - loss 0.02679964 - time (sec): 7.47 - samples/sec: 2863.51 - lr: 0.000006 - momentum: 0.000000
2023-10-17 08:54:53,616 ----------------------------------------------------------------------------------------------------
2023-10-17 08:54:53,616 EPOCH 9 done: loss 0.0259 - lr: 0.000006
2023-10-17 08:54:54,251 DEV : loss 0.1794566512107849 - f1-score (micro avg) 0.8719
2023-10-17 08:54:54,256 ----------------------------------------------------------------------------------------------------
2023-10-17 08:54:55,006 epoch 10 - iter 13/138 - loss 0.01099146 - time (sec): 0.75 - samples/sec: 2743.28 - lr: 0.000005 - momentum: 0.000000
2023-10-17 08:54:55,776 epoch 10 - iter 26/138 - loss 0.01326709 - time (sec): 1.52 - samples/sec: 2925.20 - lr: 0.000005 - momentum: 0.000000
2023-10-17 08:54:56,488 epoch 10 - iter 39/138 - loss 0.01241961 - time (sec): 2.23 - samples/sec: 2940.35 - lr: 0.000004 - momentum: 0.000000
2023-10-17 08:54:57,181 epoch 10 - iter 52/138 - loss 0.01622066 - time (sec): 2.92 - samples/sec: 2851.97 - lr: 0.000004 - momentum: 0.000000
2023-10-17 08:54:57,917 epoch 10 - iter 65/138 - loss 0.01665366 - time (sec): 3.66 - samples/sec: 2875.42 - lr: 0.000003 - momentum: 0.000000
2023-10-17 08:54:58,666 epoch 10 - iter 78/138 - loss 0.01562260 - time (sec): 4.41 - samples/sec: 2849.59 - lr: 0.000003 - momentum: 0.000000
2023-10-17 08:54:59,384 epoch 10 - iter 91/138 - loss 0.01461411 - time (sec): 5.13 - samples/sec: 2877.77 - lr: 0.000002 - momentum: 0.000000
2023-10-17 08:55:00,182 epoch 10 - iter 104/138 - loss 0.01947795 - time (sec): 5.92 - samples/sec: 2890.62 - lr: 0.000002 - momentum: 0.000000
2023-10-17 08:55:00,978 epoch 10 - iter 117/138 - loss 0.01852363 - time (sec): 6.72 - samples/sec: 2879.18 - lr: 0.000001 - momentum: 0.000000
2023-10-17 08:55:01,755 epoch 10 - iter 130/138 - loss 0.01707996 - time (sec): 7.50 - samples/sec: 2872.11 - lr: 0.000000 - momentum: 0.000000
2023-10-17 08:55:02,225 ----------------------------------------------------------------------------------------------------
2023-10-17 08:55:02,226 EPOCH 10 done: loss 0.0168 - lr: 0.000000
2023-10-17 08:55:02,944 DEV : loss 0.18495064973831177 - f1-score (micro avg) 0.8705
2023-10-17 08:55:03,287 ----------------------------------------------------------------------------------------------------
2023-10-17 08:55:03,289 Loading model from best epoch ...
2023-10-17 08:55:04,669 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
2023-10-17 08:55:05,488
Results:
- F-score (micro) 0.9029
- F-score (macro) 0.7414
- Accuracy 0.8309
By class:
precision recall f1-score support
scope 0.9029 0.8977 0.9003 176
pers 0.9672 0.9219 0.9440 128
work 0.8354 0.8919 0.8627 74
object 0.0000 0.0000 0.0000 2
loc 1.0000 1.0000 1.0000 2
micro avg 0.9053 0.9005 0.9029 382
macro avg 0.7411 0.7423 0.7414 382
weighted avg 0.9071 0.9005 0.9035 382
2023-10-17 08:55:05,488 ----------------------------------------------------------------------------------------------------
|