File size: 25,143 Bytes
f4f3810 |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 |
2023-10-06 23:17:42,929 ----------------------------------------------------------------------------------------------------
2023-10-06 23:17:42,930 Model: "SequenceTagger(
(embeddings): ByT5Embeddings(
(model): T5EncoderModel(
(shared): Embedding(384, 1472)
(encoder): T5Stack(
(embed_tokens): Embedding(384, 1472)
(block): ModuleList(
(0): T5Block(
(layer): ModuleList(
(0): T5LayerSelfAttention(
(SelfAttention): T5Attention(
(q): Linear(in_features=1472, out_features=384, bias=False)
(k): Linear(in_features=1472, out_features=384, bias=False)
(v): Linear(in_features=1472, out_features=384, bias=False)
(o): Linear(in_features=384, out_features=1472, bias=False)
(relative_attention_bias): Embedding(32, 6)
)
(layer_norm): T5LayerNorm()
(dropout): Dropout(p=0.1, inplace=False)
)
(1): T5LayerFF(
(DenseReluDense): T5DenseGatedActDense(
(wi_0): Linear(in_features=1472, out_features=3584, bias=False)
(wi_1): Linear(in_features=1472, out_features=3584, bias=False)
(wo): Linear(in_features=3584, out_features=1472, bias=False)
(dropout): Dropout(p=0.1, inplace=False)
(act): NewGELUActivation()
)
(layer_norm): T5LayerNorm()
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
(1-11): 11 x T5Block(
(layer): ModuleList(
(0): T5LayerSelfAttention(
(SelfAttention): T5Attention(
(q): Linear(in_features=1472, out_features=384, bias=False)
(k): Linear(in_features=1472, out_features=384, bias=False)
(v): Linear(in_features=1472, out_features=384, bias=False)
(o): Linear(in_features=384, out_features=1472, bias=False)
)
(layer_norm): T5LayerNorm()
(dropout): Dropout(p=0.1, inplace=False)
)
(1): T5LayerFF(
(DenseReluDense): T5DenseGatedActDense(
(wi_0): Linear(in_features=1472, out_features=3584, bias=False)
(wi_1): Linear(in_features=1472, out_features=3584, bias=False)
(wo): Linear(in_features=3584, out_features=1472, bias=False)
(dropout): Dropout(p=0.1, inplace=False)
(act): NewGELUActivation()
)
(layer_norm): T5LayerNorm()
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
)
(final_layer_norm): T5LayerNorm()
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
(locked_dropout): LockedDropout(p=0.5)
(linear): Linear(in_features=1472, out_features=25, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-06 23:17:42,930 ----------------------------------------------------------------------------------------------------
2023-10-06 23:17:42,930 MultiCorpus: 1100 train + 206 dev + 240 test sentences
- NER_HIPE_2022 Corpus: 1100 train + 206 dev + 240 test sentences - /app/.flair/datasets/ner_hipe_2022/v2.1/ajmc/de/with_doc_seperator
2023-10-06 23:17:42,930 ----------------------------------------------------------------------------------------------------
2023-10-06 23:17:42,931 Train: 1100 sentences
2023-10-06 23:17:42,931 (train_with_dev=False, train_with_test=False)
2023-10-06 23:17:42,931 ----------------------------------------------------------------------------------------------------
2023-10-06 23:17:42,931 Training Params:
2023-10-06 23:17:42,931 - learning_rate: "0.00015"
2023-10-06 23:17:42,931 - mini_batch_size: "4"
2023-10-06 23:17:42,931 - max_epochs: "10"
2023-10-06 23:17:42,931 - shuffle: "True"
2023-10-06 23:17:42,931 ----------------------------------------------------------------------------------------------------
2023-10-06 23:17:42,931 Plugins:
2023-10-06 23:17:42,931 - TensorboardLogger
2023-10-06 23:17:42,931 - LinearScheduler | warmup_fraction: '0.1'
2023-10-06 23:17:42,931 ----------------------------------------------------------------------------------------------------
2023-10-06 23:17:42,931 Final evaluation on model from best epoch (best-model.pt)
2023-10-06 23:17:42,931 - metric: "('micro avg', 'f1-score')"
2023-10-06 23:17:42,931 ----------------------------------------------------------------------------------------------------
2023-10-06 23:17:42,931 Computation:
2023-10-06 23:17:42,931 - compute on device: cuda:0
2023-10-06 23:17:42,931 - embedding storage: none
2023-10-06 23:17:42,932 ----------------------------------------------------------------------------------------------------
2023-10-06 23:17:42,932 Model training base path: "hmbench-ajmc/de-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs4-wsFalse-e10-lr0.00015-poolingfirst-layers-1-crfFalse-3"
2023-10-06 23:17:42,932 ----------------------------------------------------------------------------------------------------
2023-10-06 23:17:42,932 ----------------------------------------------------------------------------------------------------
2023-10-06 23:17:42,932 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-06 23:17:53,081 epoch 1 - iter 27/275 - loss 3.20779212 - time (sec): 10.15 - samples/sec: 215.71 - lr: 0.000014 - momentum: 0.000000
2023-10-06 23:18:04,015 epoch 1 - iter 54/275 - loss 3.19893422 - time (sec): 21.08 - samples/sec: 212.08 - lr: 0.000029 - momentum: 0.000000
2023-10-06 23:18:15,565 epoch 1 - iter 81/275 - loss 3.18017401 - time (sec): 32.63 - samples/sec: 213.13 - lr: 0.000044 - momentum: 0.000000
2023-10-06 23:18:26,626 epoch 1 - iter 108/275 - loss 3.14294964 - time (sec): 43.69 - samples/sec: 210.67 - lr: 0.000058 - momentum: 0.000000
2023-10-06 23:18:38,029 epoch 1 - iter 135/275 - loss 3.06861851 - time (sec): 55.10 - samples/sec: 210.27 - lr: 0.000073 - momentum: 0.000000
2023-10-06 23:18:48,668 epoch 1 - iter 162/275 - loss 2.98386536 - time (sec): 65.74 - samples/sec: 208.75 - lr: 0.000088 - momentum: 0.000000
2023-10-06 23:18:59,398 epoch 1 - iter 189/275 - loss 2.87891269 - time (sec): 76.46 - samples/sec: 207.70 - lr: 0.000103 - momentum: 0.000000
2023-10-06 23:19:09,952 epoch 1 - iter 216/275 - loss 2.77298535 - time (sec): 87.02 - samples/sec: 207.98 - lr: 0.000117 - momentum: 0.000000
2023-10-06 23:19:20,467 epoch 1 - iter 243/275 - loss 2.66563305 - time (sec): 97.53 - samples/sec: 207.28 - lr: 0.000132 - momentum: 0.000000
2023-10-06 23:19:30,935 epoch 1 - iter 270/275 - loss 2.54633852 - time (sec): 108.00 - samples/sec: 206.16 - lr: 0.000147 - momentum: 0.000000
2023-10-06 23:19:33,191 ----------------------------------------------------------------------------------------------------
2023-10-06 23:19:33,192 EPOCH 1 done: loss 2.5171 - lr: 0.000147
2023-10-06 23:19:39,803 DEV : loss 1.1775470972061157 - f1-score (micro avg) 0.0
2023-10-06 23:19:39,809 ----------------------------------------------------------------------------------------------------
2023-10-06 23:19:50,771 epoch 2 - iter 27/275 - loss 1.03372971 - time (sec): 10.96 - samples/sec: 215.51 - lr: 0.000148 - momentum: 0.000000
2023-10-06 23:20:02,347 epoch 2 - iter 54/275 - loss 1.00951904 - time (sec): 22.54 - samples/sec: 210.82 - lr: 0.000147 - momentum: 0.000000
2023-10-06 23:20:13,223 epoch 2 - iter 81/275 - loss 0.91537880 - time (sec): 33.41 - samples/sec: 206.90 - lr: 0.000145 - momentum: 0.000000
2023-10-06 23:20:24,587 epoch 2 - iter 108/275 - loss 0.89903855 - time (sec): 44.78 - samples/sec: 208.39 - lr: 0.000144 - momentum: 0.000000
2023-10-06 23:20:35,651 epoch 2 - iter 135/275 - loss 0.85333467 - time (sec): 55.84 - samples/sec: 207.59 - lr: 0.000142 - momentum: 0.000000
2023-10-06 23:20:46,118 epoch 2 - iter 162/275 - loss 0.82046048 - time (sec): 66.31 - samples/sec: 206.46 - lr: 0.000140 - momentum: 0.000000
2023-10-06 23:20:57,032 epoch 2 - iter 189/275 - loss 0.80119432 - time (sec): 77.22 - samples/sec: 205.82 - lr: 0.000139 - momentum: 0.000000
2023-10-06 23:21:07,772 epoch 2 - iter 216/275 - loss 0.77353092 - time (sec): 87.96 - samples/sec: 206.14 - lr: 0.000137 - momentum: 0.000000
2023-10-06 23:21:17,964 epoch 2 - iter 243/275 - loss 0.74058223 - time (sec): 98.15 - samples/sec: 204.84 - lr: 0.000135 - momentum: 0.000000
2023-10-06 23:21:29,022 epoch 2 - iter 270/275 - loss 0.70448315 - time (sec): 109.21 - samples/sec: 204.95 - lr: 0.000134 - momentum: 0.000000
2023-10-06 23:21:30,900 ----------------------------------------------------------------------------------------------------
2023-10-06 23:21:30,900 EPOCH 2 done: loss 0.7014 - lr: 0.000134
2023-10-06 23:21:37,665 DEV : loss 0.42873144149780273 - f1-score (micro avg) 0.5582
2023-10-06 23:21:37,671 saving best model
2023-10-06 23:21:38,639 ----------------------------------------------------------------------------------------------------
2023-10-06 23:21:49,785 epoch 3 - iter 27/275 - loss 0.43704065 - time (sec): 11.14 - samples/sec: 212.84 - lr: 0.000132 - momentum: 0.000000
2023-10-06 23:22:00,388 epoch 3 - iter 54/275 - loss 0.39009542 - time (sec): 21.75 - samples/sec: 208.48 - lr: 0.000130 - momentum: 0.000000
2023-10-06 23:22:11,535 epoch 3 - iter 81/275 - loss 0.37638191 - time (sec): 32.89 - samples/sec: 210.19 - lr: 0.000129 - momentum: 0.000000
2023-10-06 23:22:22,286 epoch 3 - iter 108/275 - loss 0.36351112 - time (sec): 43.65 - samples/sec: 208.11 - lr: 0.000127 - momentum: 0.000000
2023-10-06 23:22:33,756 epoch 3 - iter 135/275 - loss 0.34380491 - time (sec): 55.12 - samples/sec: 207.60 - lr: 0.000125 - momentum: 0.000000
2023-10-06 23:22:44,252 epoch 3 - iter 162/275 - loss 0.33317147 - time (sec): 65.61 - samples/sec: 206.76 - lr: 0.000124 - momentum: 0.000000
2023-10-06 23:22:55,521 epoch 3 - iter 189/275 - loss 0.32264027 - time (sec): 76.88 - samples/sec: 206.42 - lr: 0.000122 - momentum: 0.000000
2023-10-06 23:23:06,782 epoch 3 - iter 216/275 - loss 0.31955630 - time (sec): 88.14 - samples/sec: 207.28 - lr: 0.000120 - momentum: 0.000000
2023-10-06 23:23:16,907 epoch 3 - iter 243/275 - loss 0.31378673 - time (sec): 98.27 - samples/sec: 205.84 - lr: 0.000119 - momentum: 0.000000
2023-10-06 23:23:27,385 epoch 3 - iter 270/275 - loss 0.30770208 - time (sec): 108.75 - samples/sec: 206.31 - lr: 0.000117 - momentum: 0.000000
2023-10-06 23:23:29,177 ----------------------------------------------------------------------------------------------------
2023-10-06 23:23:29,177 EPOCH 3 done: loss 0.3081 - lr: 0.000117
2023-10-06 23:23:35,739 DEV : loss 0.21868450939655304 - f1-score (micro avg) 0.7855
2023-10-06 23:23:35,744 saving best model
2023-10-06 23:23:36,679 ----------------------------------------------------------------------------------------------------
2023-10-06 23:23:47,415 epoch 4 - iter 27/275 - loss 0.23843334 - time (sec): 10.73 - samples/sec: 210.93 - lr: 0.000115 - momentum: 0.000000
2023-10-06 23:23:58,293 epoch 4 - iter 54/275 - loss 0.23321435 - time (sec): 21.61 - samples/sec: 213.77 - lr: 0.000114 - momentum: 0.000000
2023-10-06 23:24:09,026 epoch 4 - iter 81/275 - loss 0.22124299 - time (sec): 32.34 - samples/sec: 211.47 - lr: 0.000112 - momentum: 0.000000
2023-10-06 23:24:19,058 epoch 4 - iter 108/275 - loss 0.20355063 - time (sec): 42.38 - samples/sec: 206.93 - lr: 0.000110 - momentum: 0.000000
2023-10-06 23:24:30,407 epoch 4 - iter 135/275 - loss 0.19004237 - time (sec): 53.73 - samples/sec: 207.67 - lr: 0.000109 - momentum: 0.000000
2023-10-06 23:24:41,880 epoch 4 - iter 162/275 - loss 0.18252030 - time (sec): 65.20 - samples/sec: 209.48 - lr: 0.000107 - momentum: 0.000000
2023-10-06 23:24:52,348 epoch 4 - iter 189/275 - loss 0.17924747 - time (sec): 75.67 - samples/sec: 207.77 - lr: 0.000105 - momentum: 0.000000
2023-10-06 23:25:03,353 epoch 4 - iter 216/275 - loss 0.17416266 - time (sec): 86.67 - samples/sec: 207.01 - lr: 0.000104 - momentum: 0.000000
2023-10-06 23:25:13,778 epoch 4 - iter 243/275 - loss 0.16862196 - time (sec): 97.10 - samples/sec: 205.85 - lr: 0.000102 - momentum: 0.000000
2023-10-06 23:25:24,256 epoch 4 - iter 270/275 - loss 0.16310851 - time (sec): 107.58 - samples/sec: 206.76 - lr: 0.000101 - momentum: 0.000000
2023-10-06 23:25:26,700 ----------------------------------------------------------------------------------------------------
2023-10-06 23:25:26,700 EPOCH 4 done: loss 0.1624 - lr: 0.000101
2023-10-06 23:25:33,324 DEV : loss 0.1493232250213623 - f1-score (micro avg) 0.8382
2023-10-06 23:25:33,330 saving best model
2023-10-06 23:25:34,248 ----------------------------------------------------------------------------------------------------
2023-10-06 23:25:45,066 epoch 5 - iter 27/275 - loss 0.08466083 - time (sec): 10.82 - samples/sec: 202.38 - lr: 0.000099 - momentum: 0.000000
2023-10-06 23:25:55,975 epoch 5 - iter 54/275 - loss 0.09065331 - time (sec): 21.72 - samples/sec: 207.69 - lr: 0.000097 - momentum: 0.000000
2023-10-06 23:26:06,526 epoch 5 - iter 81/275 - loss 0.09365691 - time (sec): 32.28 - samples/sec: 206.13 - lr: 0.000095 - momentum: 0.000000
2023-10-06 23:26:17,220 epoch 5 - iter 108/275 - loss 0.10027437 - time (sec): 42.97 - samples/sec: 204.70 - lr: 0.000094 - momentum: 0.000000
2023-10-06 23:26:27,592 epoch 5 - iter 135/275 - loss 0.10170985 - time (sec): 53.34 - samples/sec: 203.78 - lr: 0.000092 - momentum: 0.000000
2023-10-06 23:26:39,008 epoch 5 - iter 162/275 - loss 0.09929130 - time (sec): 64.76 - samples/sec: 203.88 - lr: 0.000090 - momentum: 0.000000
2023-10-06 23:26:49,894 epoch 5 - iter 189/275 - loss 0.09494369 - time (sec): 75.64 - samples/sec: 205.05 - lr: 0.000089 - momentum: 0.000000
2023-10-06 23:27:00,929 epoch 5 - iter 216/275 - loss 0.09307176 - time (sec): 86.68 - samples/sec: 204.98 - lr: 0.000087 - momentum: 0.000000
2023-10-06 23:27:11,558 epoch 5 - iter 243/275 - loss 0.10031013 - time (sec): 97.31 - samples/sec: 205.28 - lr: 0.000086 - momentum: 0.000000
2023-10-06 23:27:22,376 epoch 5 - iter 270/275 - loss 0.09762639 - time (sec): 108.13 - samples/sec: 205.60 - lr: 0.000084 - momentum: 0.000000
2023-10-06 23:27:24,773 ----------------------------------------------------------------------------------------------------
2023-10-06 23:27:24,774 EPOCH 5 done: loss 0.0987 - lr: 0.000084
2023-10-06 23:27:31,457 DEV : loss 0.12611618638038635 - f1-score (micro avg) 0.8616
2023-10-06 23:27:31,465 saving best model
2023-10-06 23:27:32,384 ----------------------------------------------------------------------------------------------------
2023-10-06 23:27:43,387 epoch 6 - iter 27/275 - loss 0.07895610 - time (sec): 11.00 - samples/sec: 207.51 - lr: 0.000082 - momentum: 0.000000
2023-10-06 23:27:53,456 epoch 6 - iter 54/275 - loss 0.09273836 - time (sec): 21.07 - samples/sec: 204.64 - lr: 0.000080 - momentum: 0.000000
2023-10-06 23:28:03,998 epoch 6 - iter 81/275 - loss 0.08414141 - time (sec): 31.61 - samples/sec: 203.97 - lr: 0.000079 - momentum: 0.000000
2023-10-06 23:28:14,503 epoch 6 - iter 108/275 - loss 0.08316482 - time (sec): 42.12 - samples/sec: 203.43 - lr: 0.000077 - momentum: 0.000000
2023-10-06 23:28:25,564 epoch 6 - iter 135/275 - loss 0.07871307 - time (sec): 53.18 - samples/sec: 202.39 - lr: 0.000075 - momentum: 0.000000
2023-10-06 23:28:36,883 epoch 6 - iter 162/275 - loss 0.08138135 - time (sec): 64.50 - samples/sec: 204.35 - lr: 0.000074 - momentum: 0.000000
2023-10-06 23:28:47,438 epoch 6 - iter 189/275 - loss 0.08231468 - time (sec): 75.05 - samples/sec: 204.10 - lr: 0.000072 - momentum: 0.000000
2023-10-06 23:28:58,758 epoch 6 - iter 216/275 - loss 0.08734570 - time (sec): 86.37 - samples/sec: 204.81 - lr: 0.000071 - momentum: 0.000000
2023-10-06 23:29:09,868 epoch 6 - iter 243/275 - loss 0.08052102 - time (sec): 97.48 - samples/sec: 205.70 - lr: 0.000069 - momentum: 0.000000
2023-10-06 23:29:20,681 epoch 6 - iter 270/275 - loss 0.07520156 - time (sec): 108.30 - samples/sec: 206.50 - lr: 0.000067 - momentum: 0.000000
2023-10-06 23:29:22,697 ----------------------------------------------------------------------------------------------------
2023-10-06 23:29:22,697 EPOCH 6 done: loss 0.0743 - lr: 0.000067
2023-10-06 23:29:29,351 DEV : loss 0.12128882855176926 - f1-score (micro avg) 0.8758
2023-10-06 23:29:29,356 saving best model
2023-10-06 23:29:30,274 ----------------------------------------------------------------------------------------------------
2023-10-06 23:29:40,216 epoch 7 - iter 27/275 - loss 0.05932625 - time (sec): 9.94 - samples/sec: 200.08 - lr: 0.000065 - momentum: 0.000000
2023-10-06 23:29:51,868 epoch 7 - iter 54/275 - loss 0.04672326 - time (sec): 21.59 - samples/sec: 205.76 - lr: 0.000064 - momentum: 0.000000
2023-10-06 23:30:02,697 epoch 7 - iter 81/275 - loss 0.04533032 - time (sec): 32.42 - samples/sec: 207.73 - lr: 0.000062 - momentum: 0.000000
2023-10-06 23:30:13,412 epoch 7 - iter 108/275 - loss 0.04353625 - time (sec): 43.14 - samples/sec: 208.50 - lr: 0.000060 - momentum: 0.000000
2023-10-06 23:30:24,374 epoch 7 - iter 135/275 - loss 0.05327309 - time (sec): 54.10 - samples/sec: 210.45 - lr: 0.000059 - momentum: 0.000000
2023-10-06 23:30:35,356 epoch 7 - iter 162/275 - loss 0.05492008 - time (sec): 65.08 - samples/sec: 210.63 - lr: 0.000057 - momentum: 0.000000
2023-10-06 23:30:45,948 epoch 7 - iter 189/275 - loss 0.05356602 - time (sec): 75.67 - samples/sec: 209.59 - lr: 0.000056 - momentum: 0.000000
2023-10-06 23:30:56,774 epoch 7 - iter 216/275 - loss 0.05638404 - time (sec): 86.50 - samples/sec: 209.12 - lr: 0.000054 - momentum: 0.000000
2023-10-06 23:31:07,501 epoch 7 - iter 243/275 - loss 0.05727416 - time (sec): 97.23 - samples/sec: 207.60 - lr: 0.000052 - momentum: 0.000000
2023-10-06 23:31:18,650 epoch 7 - iter 270/275 - loss 0.05734359 - time (sec): 108.37 - samples/sec: 207.38 - lr: 0.000051 - momentum: 0.000000
2023-10-06 23:31:20,274 ----------------------------------------------------------------------------------------------------
2023-10-06 23:31:20,274 EPOCH 7 done: loss 0.0575 - lr: 0.000051
2023-10-06 23:31:26,927 DEV : loss 0.12337013334035873 - f1-score (micro avg) 0.8783
2023-10-06 23:31:26,932 saving best model
2023-10-06 23:31:28,029 ----------------------------------------------------------------------------------------------------
2023-10-06 23:31:38,496 epoch 8 - iter 27/275 - loss 0.06679142 - time (sec): 10.47 - samples/sec: 205.91 - lr: 0.000049 - momentum: 0.000000
2023-10-06 23:31:49,201 epoch 8 - iter 54/275 - loss 0.05175168 - time (sec): 21.17 - samples/sec: 206.19 - lr: 0.000047 - momentum: 0.000000
2023-10-06 23:31:59,409 epoch 8 - iter 81/275 - loss 0.04590241 - time (sec): 31.38 - samples/sec: 206.03 - lr: 0.000045 - momentum: 0.000000
2023-10-06 23:32:09,829 epoch 8 - iter 108/275 - loss 0.05025701 - time (sec): 41.80 - samples/sec: 206.68 - lr: 0.000044 - momentum: 0.000000
2023-10-06 23:32:20,652 epoch 8 - iter 135/275 - loss 0.04689769 - time (sec): 52.62 - samples/sec: 205.85 - lr: 0.000042 - momentum: 0.000000
2023-10-06 23:32:32,043 epoch 8 - iter 162/275 - loss 0.04687291 - time (sec): 64.01 - samples/sec: 207.51 - lr: 0.000041 - momentum: 0.000000
2023-10-06 23:32:43,660 epoch 8 - iter 189/275 - loss 0.04632883 - time (sec): 75.63 - samples/sec: 207.27 - lr: 0.000039 - momentum: 0.000000
2023-10-06 23:32:55,088 epoch 8 - iter 216/275 - loss 0.05000443 - time (sec): 87.06 - samples/sec: 207.66 - lr: 0.000037 - momentum: 0.000000
2023-10-06 23:33:05,581 epoch 8 - iter 243/275 - loss 0.04980880 - time (sec): 97.55 - samples/sec: 207.44 - lr: 0.000036 - momentum: 0.000000
2023-10-06 23:33:15,960 epoch 8 - iter 270/275 - loss 0.04839911 - time (sec): 107.93 - samples/sec: 207.10 - lr: 0.000034 - momentum: 0.000000
2023-10-06 23:33:18,029 ----------------------------------------------------------------------------------------------------
2023-10-06 23:33:18,030 EPOCH 8 done: loss 0.0477 - lr: 0.000034
2023-10-06 23:33:24,660 DEV : loss 0.127992644906044 - f1-score (micro avg) 0.873
2023-10-06 23:33:24,666 ----------------------------------------------------------------------------------------------------
2023-10-06 23:33:34,726 epoch 9 - iter 27/275 - loss 0.04178413 - time (sec): 10.06 - samples/sec: 199.43 - lr: 0.000032 - momentum: 0.000000
2023-10-06 23:33:45,547 epoch 9 - iter 54/275 - loss 0.05131075 - time (sec): 20.88 - samples/sec: 204.79 - lr: 0.000030 - momentum: 0.000000
2023-10-06 23:33:56,356 epoch 9 - iter 81/275 - loss 0.05860657 - time (sec): 31.69 - samples/sec: 206.35 - lr: 0.000029 - momentum: 0.000000
2023-10-06 23:34:06,475 epoch 9 - iter 108/275 - loss 0.05561689 - time (sec): 41.81 - samples/sec: 203.29 - lr: 0.000027 - momentum: 0.000000
2023-10-06 23:34:17,273 epoch 9 - iter 135/275 - loss 0.05086730 - time (sec): 52.61 - samples/sec: 203.17 - lr: 0.000026 - momentum: 0.000000
2023-10-06 23:34:28,378 epoch 9 - iter 162/275 - loss 0.04692919 - time (sec): 63.71 - samples/sec: 204.55 - lr: 0.000024 - momentum: 0.000000
2023-10-06 23:34:39,748 epoch 9 - iter 189/275 - loss 0.04599155 - time (sec): 75.08 - samples/sec: 206.27 - lr: 0.000022 - momentum: 0.000000
2023-10-06 23:34:51,078 epoch 9 - iter 216/275 - loss 0.04321166 - time (sec): 86.41 - samples/sec: 206.78 - lr: 0.000021 - momentum: 0.000000
2023-10-06 23:35:01,862 epoch 9 - iter 243/275 - loss 0.04328397 - time (sec): 97.19 - samples/sec: 206.47 - lr: 0.000019 - momentum: 0.000000
2023-10-06 23:35:12,650 epoch 9 - iter 270/275 - loss 0.04290629 - time (sec): 107.98 - samples/sec: 206.79 - lr: 0.000017 - momentum: 0.000000
2023-10-06 23:35:14,703 ----------------------------------------------------------------------------------------------------
2023-10-06 23:35:14,704 EPOCH 9 done: loss 0.0429 - lr: 0.000017
2023-10-06 23:35:21,542 DEV : loss 0.1252521574497223 - f1-score (micro avg) 0.883
2023-10-06 23:35:21,548 saving best model
2023-10-06 23:35:22,470 ----------------------------------------------------------------------------------------------------
2023-10-06 23:35:32,902 epoch 10 - iter 27/275 - loss 0.03780507 - time (sec): 10.43 - samples/sec: 203.07 - lr: 0.000015 - momentum: 0.000000
2023-10-06 23:35:44,530 epoch 10 - iter 54/275 - loss 0.02927955 - time (sec): 22.06 - samples/sec: 203.91 - lr: 0.000014 - momentum: 0.000000
2023-10-06 23:35:54,510 epoch 10 - iter 81/275 - loss 0.04116419 - time (sec): 32.04 - samples/sec: 201.14 - lr: 0.000012 - momentum: 0.000000
2023-10-06 23:36:04,734 epoch 10 - iter 108/275 - loss 0.04094626 - time (sec): 42.26 - samples/sec: 200.70 - lr: 0.000011 - momentum: 0.000000
2023-10-06 23:36:16,023 epoch 10 - iter 135/275 - loss 0.04004258 - time (sec): 53.55 - samples/sec: 203.73 - lr: 0.000009 - momentum: 0.000000
2023-10-06 23:36:26,817 epoch 10 - iter 162/275 - loss 0.03973698 - time (sec): 64.35 - samples/sec: 204.54 - lr: 0.000007 - momentum: 0.000000
2023-10-06 23:36:37,043 epoch 10 - iter 189/275 - loss 0.04053412 - time (sec): 74.57 - samples/sec: 203.43 - lr: 0.000006 - momentum: 0.000000
2023-10-06 23:36:48,182 epoch 10 - iter 216/275 - loss 0.04054429 - time (sec): 85.71 - samples/sec: 204.55 - lr: 0.000004 - momentum: 0.000000
2023-10-06 23:36:59,116 epoch 10 - iter 243/275 - loss 0.04240430 - time (sec): 96.64 - samples/sec: 205.37 - lr: 0.000002 - momentum: 0.000000
2023-10-06 23:37:10,460 epoch 10 - iter 270/275 - loss 0.04062955 - time (sec): 107.99 - samples/sec: 206.36 - lr: 0.000001 - momentum: 0.000000
2023-10-06 23:37:12,554 ----------------------------------------------------------------------------------------------------
2023-10-06 23:37:12,555 EPOCH 10 done: loss 0.0405 - lr: 0.000001
2023-10-06 23:37:19,202 DEV : loss 0.1265440583229065 - f1-score (micro avg) 0.8788
2023-10-06 23:37:20,098 ----------------------------------------------------------------------------------------------------
2023-10-06 23:37:20,100 Loading model from best epoch ...
2023-10-06 23:37:24,178 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
2023-10-06 23:37:31,313
Results:
- F-score (micro) 0.9129
- F-score (macro) 0.5465
- Accuracy 0.8561
By class:
precision recall f1-score support
scope 0.9011 0.9318 0.9162 176
pers 0.9457 0.9531 0.9494 128
work 0.8553 0.8784 0.8667 74
object 0.0000 0.0000 0.0000 2
loc 0.0000 0.0000 0.0000 2
micro avg 0.9070 0.9188 0.9129 382
macro avg 0.5404 0.5527 0.5465 382
weighted avg 0.8977 0.9188 0.9081 382
2023-10-06 23:37:31,313 ----------------------------------------------------------------------------------------------------
|