File size: 24,991 Bytes
d71a669 |
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 |
2023-10-08 18:16:32,446 ----------------------------------------------------------------------------------------------------
2023-10-08 18:16:32,447 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()
)"
2023-10-08 18:16:32,447 ----------------------------------------------------------------------------------------------------
2023-10-08 18:16:32,447 MultiCorpus: 966 train + 219 dev + 204 test sentences
- NER_HIPE_2022 Corpus: 966 train + 219 dev + 204 test sentences - /app/.flair/datasets/ner_hipe_2022/v2.1/ajmc/fr/with_doc_seperator
2023-10-08 18:16:32,448 ----------------------------------------------------------------------------------------------------
2023-10-08 18:16:32,448 Train: 966 sentences
2023-10-08 18:16:32,448 (train_with_dev=False, train_with_test=False)
2023-10-08 18:16:32,448 ----------------------------------------------------------------------------------------------------
2023-10-08 18:16:32,448 Training Params:
2023-10-08 18:16:32,448 - learning_rate: "0.00015"
2023-10-08 18:16:32,448 - mini_batch_size: "4"
2023-10-08 18:16:32,448 - max_epochs: "10"
2023-10-08 18:16:32,448 - shuffle: "True"
2023-10-08 18:16:32,448 ----------------------------------------------------------------------------------------------------
2023-10-08 18:16:32,448 Plugins:
2023-10-08 18:16:32,448 - TensorboardLogger
2023-10-08 18:16:32,448 - LinearScheduler | warmup_fraction: '0.1'
2023-10-08 18:16:32,448 ----------------------------------------------------------------------------------------------------
2023-10-08 18:16:32,448 Final evaluation on model from best epoch (best-model.pt)
2023-10-08 18:16:32,448 - metric: "('micro avg', 'f1-score')"
2023-10-08 18:16:32,448 ----------------------------------------------------------------------------------------------------
2023-10-08 18:16:32,448 Computation:
2023-10-08 18:16:32,449 - compute on device: cuda:0
2023-10-08 18:16:32,449 - embedding storage: none
2023-10-08 18:16:32,449 ----------------------------------------------------------------------------------------------------
2023-10-08 18:16:32,449 Model training base path: "hmbench-ajmc/fr-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs4-wsFalse-e10-lr0.00015-poolingfirst-layers-1-crfFalse-1"
2023-10-08 18:16:32,449 ----------------------------------------------------------------------------------------------------
2023-10-08 18:16:32,449 ----------------------------------------------------------------------------------------------------
2023-10-08 18:16:32,449 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-08 18:16:42,184 epoch 1 - iter 24/242 - loss 3.23269142 - time (sec): 9.73 - samples/sec: 224.89 - lr: 0.000014 - momentum: 0.000000
2023-10-08 18:16:51,473 epoch 1 - iter 48/242 - loss 3.22249485 - time (sec): 19.02 - samples/sec: 244.18 - lr: 0.000029 - momentum: 0.000000
2023-10-08 18:17:00,441 epoch 1 - iter 72/242 - loss 3.20551448 - time (sec): 27.99 - samples/sec: 247.62 - lr: 0.000044 - momentum: 0.000000
2023-10-08 18:17:09,710 epoch 1 - iter 96/242 - loss 3.15823056 - time (sec): 37.26 - samples/sec: 253.14 - lr: 0.000059 - momentum: 0.000000
2023-10-08 18:17:18,886 epoch 1 - iter 120/242 - loss 3.07644769 - time (sec): 46.44 - samples/sec: 253.57 - lr: 0.000074 - momentum: 0.000000
2023-10-08 18:17:28,777 epoch 1 - iter 144/242 - loss 2.97338064 - time (sec): 56.33 - samples/sec: 255.90 - lr: 0.000089 - momentum: 0.000000
2023-10-08 18:17:37,776 epoch 1 - iter 168/242 - loss 2.87431993 - time (sec): 65.33 - samples/sec: 256.24 - lr: 0.000104 - momentum: 0.000000
2023-10-08 18:17:46,696 epoch 1 - iter 192/242 - loss 2.76814205 - time (sec): 74.25 - samples/sec: 256.36 - lr: 0.000118 - momentum: 0.000000
2023-10-08 18:17:56,559 epoch 1 - iter 216/242 - loss 2.63563561 - time (sec): 84.11 - samples/sec: 258.43 - lr: 0.000133 - momentum: 0.000000
2023-10-08 18:18:06,722 epoch 1 - iter 240/242 - loss 2.49337003 - time (sec): 94.27 - samples/sec: 260.02 - lr: 0.000148 - momentum: 0.000000
2023-10-08 18:18:07,447 ----------------------------------------------------------------------------------------------------
2023-10-08 18:18:07,447 EPOCH 1 done: loss 2.4815 - lr: 0.000148
2023-10-08 18:18:13,344 DEV : loss 1.187881588935852 - f1-score (micro avg) 0.0
2023-10-08 18:18:13,350 ----------------------------------------------------------------------------------------------------
2023-10-08 18:18:22,466 epoch 2 - iter 24/242 - loss 1.16961126 - time (sec): 9.11 - samples/sec: 249.16 - lr: 0.000148 - momentum: 0.000000
2023-10-08 18:18:31,815 epoch 2 - iter 48/242 - loss 1.03643335 - time (sec): 18.46 - samples/sec: 252.66 - lr: 0.000147 - momentum: 0.000000
2023-10-08 18:18:41,103 epoch 2 - iter 72/242 - loss 0.94733732 - time (sec): 27.75 - samples/sec: 256.46 - lr: 0.000145 - momentum: 0.000000
2023-10-08 18:18:50,518 epoch 2 - iter 96/242 - loss 0.87122298 - time (sec): 37.17 - samples/sec: 259.14 - lr: 0.000143 - momentum: 0.000000
2023-10-08 18:18:59,786 epoch 2 - iter 120/242 - loss 0.81457811 - time (sec): 46.43 - samples/sec: 258.47 - lr: 0.000142 - momentum: 0.000000
2023-10-08 18:19:09,536 epoch 2 - iter 144/242 - loss 0.77131873 - time (sec): 56.18 - samples/sec: 261.82 - lr: 0.000140 - momentum: 0.000000
2023-10-08 18:19:19,642 epoch 2 - iter 168/242 - loss 0.72892369 - time (sec): 66.29 - samples/sec: 261.41 - lr: 0.000139 - momentum: 0.000000
2023-10-08 18:19:28,785 epoch 2 - iter 192/242 - loss 0.68941919 - time (sec): 75.43 - samples/sec: 261.16 - lr: 0.000137 - momentum: 0.000000
2023-10-08 18:19:37,777 epoch 2 - iter 216/242 - loss 0.65715679 - time (sec): 84.43 - samples/sec: 260.48 - lr: 0.000135 - momentum: 0.000000
2023-10-08 18:19:47,470 epoch 2 - iter 240/242 - loss 0.62460856 - time (sec): 94.12 - samples/sec: 261.61 - lr: 0.000134 - momentum: 0.000000
2023-10-08 18:19:48,012 ----------------------------------------------------------------------------------------------------
2023-10-08 18:19:48,012 EPOCH 2 done: loss 0.6235 - lr: 0.000134
2023-10-08 18:19:53,806 DEV : loss 0.3897227346897125 - f1-score (micro avg) 0.3731
2023-10-08 18:19:53,813 saving best model
2023-10-08 18:19:54,725 ----------------------------------------------------------------------------------------------------
2023-10-08 18:20:04,366 epoch 3 - iter 24/242 - loss 0.35697873 - time (sec): 9.64 - samples/sec: 268.39 - lr: 0.000132 - momentum: 0.000000
2023-10-08 18:20:13,604 epoch 3 - iter 48/242 - loss 0.37113707 - time (sec): 18.88 - samples/sec: 267.62 - lr: 0.000130 - momentum: 0.000000
2023-10-08 18:20:23,337 epoch 3 - iter 72/242 - loss 0.34367840 - time (sec): 28.61 - samples/sec: 270.64 - lr: 0.000128 - momentum: 0.000000
2023-10-08 18:20:32,526 epoch 3 - iter 96/242 - loss 0.32984894 - time (sec): 37.80 - samples/sec: 266.51 - lr: 0.000127 - momentum: 0.000000
2023-10-08 18:20:41,455 epoch 3 - iter 120/242 - loss 0.31221782 - time (sec): 46.73 - samples/sec: 264.59 - lr: 0.000125 - momentum: 0.000000
2023-10-08 18:20:50,329 epoch 3 - iter 144/242 - loss 0.30435470 - time (sec): 55.60 - samples/sec: 263.57 - lr: 0.000124 - momentum: 0.000000
2023-10-08 18:21:00,276 epoch 3 - iter 168/242 - loss 0.29962563 - time (sec): 65.55 - samples/sec: 263.91 - lr: 0.000122 - momentum: 0.000000
2023-10-08 18:21:09,953 epoch 3 - iter 192/242 - loss 0.29351352 - time (sec): 75.23 - samples/sec: 264.92 - lr: 0.000120 - momentum: 0.000000
2023-10-08 18:21:18,884 epoch 3 - iter 216/242 - loss 0.28484928 - time (sec): 84.16 - samples/sec: 263.64 - lr: 0.000119 - momentum: 0.000000
2023-10-08 18:21:28,210 epoch 3 - iter 240/242 - loss 0.27921957 - time (sec): 93.48 - samples/sec: 263.07 - lr: 0.000117 - momentum: 0.000000
2023-10-08 18:21:28,826 ----------------------------------------------------------------------------------------------------
2023-10-08 18:21:28,826 EPOCH 3 done: loss 0.2786 - lr: 0.000117
2023-10-08 18:21:34,635 DEV : loss 0.22860029339790344 - f1-score (micro avg) 0.5446
2023-10-08 18:21:34,641 saving best model
2023-10-08 18:21:39,854 ----------------------------------------------------------------------------------------------------
2023-10-08 18:21:48,411 epoch 4 - iter 24/242 - loss 0.23513209 - time (sec): 8.56 - samples/sec: 256.67 - lr: 0.000115 - momentum: 0.000000
2023-10-08 18:21:58,114 epoch 4 - iter 48/242 - loss 0.22118636 - time (sec): 18.26 - samples/sec: 265.13 - lr: 0.000113 - momentum: 0.000000
2023-10-08 18:22:07,595 epoch 4 - iter 72/242 - loss 0.20467054 - time (sec): 27.74 - samples/sec: 262.11 - lr: 0.000112 - momentum: 0.000000
2023-10-08 18:22:16,768 epoch 4 - iter 96/242 - loss 0.19305601 - time (sec): 36.91 - samples/sec: 262.29 - lr: 0.000110 - momentum: 0.000000
2023-10-08 18:22:25,785 epoch 4 - iter 120/242 - loss 0.19060223 - time (sec): 45.93 - samples/sec: 261.90 - lr: 0.000109 - momentum: 0.000000
2023-10-08 18:22:36,021 epoch 4 - iter 144/242 - loss 0.18708839 - time (sec): 56.17 - samples/sec: 263.75 - lr: 0.000107 - momentum: 0.000000
2023-10-08 18:22:45,696 epoch 4 - iter 168/242 - loss 0.17861271 - time (sec): 65.84 - samples/sec: 264.17 - lr: 0.000105 - momentum: 0.000000
2023-10-08 18:22:54,966 epoch 4 - iter 192/242 - loss 0.17748223 - time (sec): 75.11 - samples/sec: 262.29 - lr: 0.000104 - momentum: 0.000000
2023-10-08 18:23:04,486 epoch 4 - iter 216/242 - loss 0.17548225 - time (sec): 84.63 - samples/sec: 261.77 - lr: 0.000102 - momentum: 0.000000
2023-10-08 18:23:13,862 epoch 4 - iter 240/242 - loss 0.17101350 - time (sec): 94.01 - samples/sec: 261.05 - lr: 0.000100 - momentum: 0.000000
2023-10-08 18:23:14,560 ----------------------------------------------------------------------------------------------------
2023-10-08 18:23:14,561 EPOCH 4 done: loss 0.1710 - lr: 0.000100
2023-10-08 18:23:20,444 DEV : loss 0.15941345691680908 - f1-score (micro avg) 0.8212
2023-10-08 18:23:20,451 saving best model
2023-10-08 18:23:24,842 ----------------------------------------------------------------------------------------------------
2023-10-08 18:23:34,448 epoch 5 - iter 24/242 - loss 0.16486530 - time (sec): 9.61 - samples/sec: 263.61 - lr: 0.000098 - momentum: 0.000000
2023-10-08 18:23:43,910 epoch 5 - iter 48/242 - loss 0.13645891 - time (sec): 19.07 - samples/sec: 258.40 - lr: 0.000097 - momentum: 0.000000
2023-10-08 18:23:53,993 epoch 5 - iter 72/242 - loss 0.13013903 - time (sec): 29.15 - samples/sec: 264.42 - lr: 0.000095 - momentum: 0.000000
2023-10-08 18:24:04,037 epoch 5 - iter 96/242 - loss 0.12377710 - time (sec): 39.19 - samples/sec: 265.32 - lr: 0.000094 - momentum: 0.000000
2023-10-08 18:24:13,765 epoch 5 - iter 120/242 - loss 0.12205066 - time (sec): 48.92 - samples/sec: 262.72 - lr: 0.000092 - momentum: 0.000000
2023-10-08 18:24:22,979 epoch 5 - iter 144/242 - loss 0.12395128 - time (sec): 58.14 - samples/sec: 260.01 - lr: 0.000090 - momentum: 0.000000
2023-10-08 18:24:32,801 epoch 5 - iter 168/242 - loss 0.11820058 - time (sec): 67.96 - samples/sec: 259.84 - lr: 0.000089 - momentum: 0.000000
2023-10-08 18:24:41,768 epoch 5 - iter 192/242 - loss 0.11799339 - time (sec): 76.92 - samples/sec: 257.75 - lr: 0.000087 - momentum: 0.000000
2023-10-08 18:24:51,316 epoch 5 - iter 216/242 - loss 0.11722304 - time (sec): 86.47 - samples/sec: 256.40 - lr: 0.000085 - momentum: 0.000000
2023-10-08 18:25:00,897 epoch 5 - iter 240/242 - loss 0.11333769 - time (sec): 96.05 - samples/sec: 255.45 - lr: 0.000084 - momentum: 0.000000
2023-10-08 18:25:01,673 ----------------------------------------------------------------------------------------------------
2023-10-08 18:25:01,673 EPOCH 5 done: loss 0.1129 - lr: 0.000084
2023-10-08 18:25:07,915 DEV : loss 0.1479427069425583 - f1-score (micro avg) 0.8097
2023-10-08 18:25:07,921 ----------------------------------------------------------------------------------------------------
2023-10-08 18:25:17,380 epoch 6 - iter 24/242 - loss 0.09476938 - time (sec): 9.46 - samples/sec: 237.69 - lr: 0.000082 - momentum: 0.000000
2023-10-08 18:25:27,454 epoch 6 - iter 48/242 - loss 0.08965422 - time (sec): 19.53 - samples/sec: 245.46 - lr: 0.000080 - momentum: 0.000000
2023-10-08 18:25:37,034 epoch 6 - iter 72/242 - loss 0.08409169 - time (sec): 29.11 - samples/sec: 243.79 - lr: 0.000079 - momentum: 0.000000
2023-10-08 18:25:47,244 epoch 6 - iter 96/242 - loss 0.08653234 - time (sec): 39.32 - samples/sec: 244.98 - lr: 0.000077 - momentum: 0.000000
2023-10-08 18:25:57,462 epoch 6 - iter 120/242 - loss 0.08433397 - time (sec): 49.54 - samples/sec: 245.74 - lr: 0.000075 - momentum: 0.000000
2023-10-08 18:26:08,047 epoch 6 - iter 144/242 - loss 0.08364166 - time (sec): 60.12 - samples/sec: 247.54 - lr: 0.000074 - momentum: 0.000000
2023-10-08 18:26:18,163 epoch 6 - iter 168/242 - loss 0.08837454 - time (sec): 70.24 - samples/sec: 248.70 - lr: 0.000072 - momentum: 0.000000
2023-10-08 18:26:27,960 epoch 6 - iter 192/242 - loss 0.08797324 - time (sec): 80.04 - samples/sec: 246.32 - lr: 0.000070 - momentum: 0.000000
2023-10-08 18:26:37,514 epoch 6 - iter 216/242 - loss 0.08549075 - time (sec): 89.59 - samples/sec: 245.10 - lr: 0.000069 - momentum: 0.000000
2023-10-08 18:26:48,162 epoch 6 - iter 240/242 - loss 0.08456891 - time (sec): 100.24 - samples/sec: 244.47 - lr: 0.000067 - momentum: 0.000000
2023-10-08 18:26:49,071 ----------------------------------------------------------------------------------------------------
2023-10-08 18:26:49,072 EPOCH 6 done: loss 0.0846 - lr: 0.000067
2023-10-08 18:26:55,670 DEV : loss 0.13671354949474335 - f1-score (micro avg) 0.824
2023-10-08 18:26:55,676 saving best model
2023-10-08 18:27:00,080 ----------------------------------------------------------------------------------------------------
2023-10-08 18:27:08,919 epoch 7 - iter 24/242 - loss 0.04626953 - time (sec): 8.84 - samples/sec: 221.55 - lr: 0.000065 - momentum: 0.000000
2023-10-08 18:27:19,128 epoch 7 - iter 48/242 - loss 0.06148305 - time (sec): 19.05 - samples/sec: 239.21 - lr: 0.000064 - momentum: 0.000000
2023-10-08 18:27:29,148 epoch 7 - iter 72/242 - loss 0.06595940 - time (sec): 29.07 - samples/sec: 240.18 - lr: 0.000062 - momentum: 0.000000
2023-10-08 18:27:39,577 epoch 7 - iter 96/242 - loss 0.06887665 - time (sec): 39.50 - samples/sec: 242.89 - lr: 0.000060 - momentum: 0.000000
2023-10-08 18:27:49,708 epoch 7 - iter 120/242 - loss 0.06674868 - time (sec): 49.63 - samples/sec: 243.18 - lr: 0.000059 - momentum: 0.000000
2023-10-08 18:27:58,945 epoch 7 - iter 144/242 - loss 0.06165700 - time (sec): 58.86 - samples/sec: 240.61 - lr: 0.000057 - momentum: 0.000000
2023-10-08 18:28:09,127 epoch 7 - iter 168/242 - loss 0.06476297 - time (sec): 69.05 - samples/sec: 240.87 - lr: 0.000055 - momentum: 0.000000
2023-10-08 18:28:19,497 epoch 7 - iter 192/242 - loss 0.06659121 - time (sec): 79.41 - samples/sec: 241.57 - lr: 0.000054 - momentum: 0.000000
2023-10-08 18:28:30,481 epoch 7 - iter 216/242 - loss 0.06620009 - time (sec): 90.40 - samples/sec: 242.38 - lr: 0.000052 - momentum: 0.000000
2023-10-08 18:28:41,271 epoch 7 - iter 240/242 - loss 0.06433525 - time (sec): 101.19 - samples/sec: 242.96 - lr: 0.000050 - momentum: 0.000000
2023-10-08 18:28:41,908 ----------------------------------------------------------------------------------------------------
2023-10-08 18:28:41,908 EPOCH 7 done: loss 0.0641 - lr: 0.000050
2023-10-08 18:28:48,483 DEV : loss 0.1378345787525177 - f1-score (micro avg) 0.8149
2023-10-08 18:28:48,489 ----------------------------------------------------------------------------------------------------
2023-10-08 18:28:57,955 epoch 8 - iter 24/242 - loss 0.05161041 - time (sec): 9.46 - samples/sec: 222.09 - lr: 0.000049 - momentum: 0.000000
2023-10-08 18:29:08,141 epoch 8 - iter 48/242 - loss 0.05952830 - time (sec): 19.65 - samples/sec: 236.58 - lr: 0.000047 - momentum: 0.000000
2023-10-08 18:29:18,619 epoch 8 - iter 72/242 - loss 0.05676119 - time (sec): 30.13 - samples/sec: 243.02 - lr: 0.000045 - momentum: 0.000000
2023-10-08 18:29:28,729 epoch 8 - iter 96/242 - loss 0.04846381 - time (sec): 40.24 - samples/sec: 243.03 - lr: 0.000044 - momentum: 0.000000
2023-10-08 18:29:39,172 epoch 8 - iter 120/242 - loss 0.04918867 - time (sec): 50.68 - samples/sec: 243.12 - lr: 0.000042 - momentum: 0.000000
2023-10-08 18:29:49,395 epoch 8 - iter 144/242 - loss 0.04964492 - time (sec): 60.91 - samples/sec: 244.08 - lr: 0.000040 - momentum: 0.000000
2023-10-08 18:29:59,859 epoch 8 - iter 168/242 - loss 0.05311315 - time (sec): 71.37 - samples/sec: 245.18 - lr: 0.000039 - momentum: 0.000000
2023-10-08 18:30:10,010 epoch 8 - iter 192/242 - loss 0.05009814 - time (sec): 81.52 - samples/sec: 244.73 - lr: 0.000037 - momentum: 0.000000
2023-10-08 18:30:19,787 epoch 8 - iter 216/242 - loss 0.04810545 - time (sec): 91.30 - samples/sec: 243.95 - lr: 0.000035 - momentum: 0.000000
2023-10-08 18:30:29,559 epoch 8 - iter 240/242 - loss 0.04878436 - time (sec): 101.07 - samples/sec: 243.65 - lr: 0.000034 - momentum: 0.000000
2023-10-08 18:30:30,089 ----------------------------------------------------------------------------------------------------
2023-10-08 18:30:30,089 EPOCH 8 done: loss 0.0487 - lr: 0.000034
2023-10-08 18:30:36,560 DEV : loss 0.14124426245689392 - f1-score (micro avg) 0.8231
2023-10-08 18:30:36,566 ----------------------------------------------------------------------------------------------------
2023-10-08 18:30:46,085 epoch 9 - iter 24/242 - loss 0.05078836 - time (sec): 9.52 - samples/sec: 234.52 - lr: 0.000032 - momentum: 0.000000
2023-10-08 18:30:55,435 epoch 9 - iter 48/242 - loss 0.04517817 - time (sec): 18.87 - samples/sec: 233.10 - lr: 0.000030 - momentum: 0.000000
2023-10-08 18:31:05,694 epoch 9 - iter 72/242 - loss 0.03891168 - time (sec): 29.13 - samples/sec: 237.11 - lr: 0.000029 - momentum: 0.000000
2023-10-08 18:31:15,788 epoch 9 - iter 96/242 - loss 0.04475568 - time (sec): 39.22 - samples/sec: 240.97 - lr: 0.000027 - momentum: 0.000000
2023-10-08 18:31:26,418 epoch 9 - iter 120/242 - loss 0.04568425 - time (sec): 49.85 - samples/sec: 243.13 - lr: 0.000025 - momentum: 0.000000
2023-10-08 18:31:37,134 epoch 9 - iter 144/242 - loss 0.04606648 - time (sec): 60.57 - samples/sec: 242.53 - lr: 0.000024 - momentum: 0.000000
2023-10-08 18:31:47,375 epoch 9 - iter 168/242 - loss 0.04397738 - time (sec): 70.81 - samples/sec: 242.77 - lr: 0.000022 - momentum: 0.000000
2023-10-08 18:31:57,299 epoch 9 - iter 192/242 - loss 0.04245585 - time (sec): 80.73 - samples/sec: 242.51 - lr: 0.000020 - momentum: 0.000000
2023-10-08 18:32:07,349 epoch 9 - iter 216/242 - loss 0.04348688 - time (sec): 90.78 - samples/sec: 242.62 - lr: 0.000019 - momentum: 0.000000
2023-10-08 18:32:17,558 epoch 9 - iter 240/242 - loss 0.04244140 - time (sec): 100.99 - samples/sec: 243.14 - lr: 0.000017 - momentum: 0.000000
2023-10-08 18:32:18,254 ----------------------------------------------------------------------------------------------------
2023-10-08 18:32:18,255 EPOCH 9 done: loss 0.0426 - lr: 0.000017
2023-10-08 18:32:24,676 DEV : loss 0.14517952501773834 - f1-score (micro avg) 0.8109
2023-10-08 18:32:24,682 ----------------------------------------------------------------------------------------------------
2023-10-08 18:32:34,783 epoch 10 - iter 24/242 - loss 0.03127148 - time (sec): 10.10 - samples/sec: 242.77 - lr: 0.000015 - momentum: 0.000000
2023-10-08 18:32:45,274 epoch 10 - iter 48/242 - loss 0.02653376 - time (sec): 20.59 - samples/sec: 246.08 - lr: 0.000014 - momentum: 0.000000
2023-10-08 18:32:55,388 epoch 10 - iter 72/242 - loss 0.02893367 - time (sec): 30.70 - samples/sec: 245.21 - lr: 0.000012 - momentum: 0.000000
2023-10-08 18:33:05,766 epoch 10 - iter 96/242 - loss 0.03364083 - time (sec): 41.08 - samples/sec: 245.84 - lr: 0.000010 - momentum: 0.000000
2023-10-08 18:33:15,066 epoch 10 - iter 120/242 - loss 0.03636518 - time (sec): 50.38 - samples/sec: 242.54 - lr: 0.000009 - momentum: 0.000000
2023-10-08 18:33:24,798 epoch 10 - iter 144/242 - loss 0.03644534 - time (sec): 60.11 - samples/sec: 241.39 - lr: 0.000007 - momentum: 0.000000
2023-10-08 18:33:34,831 epoch 10 - iter 168/242 - loss 0.03646037 - time (sec): 70.15 - samples/sec: 241.73 - lr: 0.000005 - momentum: 0.000000
2023-10-08 18:33:45,003 epoch 10 - iter 192/242 - loss 0.03701845 - time (sec): 80.32 - samples/sec: 242.52 - lr: 0.000004 - momentum: 0.000000
2023-10-08 18:33:55,174 epoch 10 - iter 216/242 - loss 0.03595125 - time (sec): 90.49 - samples/sec: 242.81 - lr: 0.000002 - momentum: 0.000000
2023-10-08 18:34:05,446 epoch 10 - iter 240/242 - loss 0.03766310 - time (sec): 100.76 - samples/sec: 243.91 - lr: 0.000000 - momentum: 0.000000
2023-10-08 18:34:06,228 ----------------------------------------------------------------------------------------------------
2023-10-08 18:34:06,228 EPOCH 10 done: loss 0.0379 - lr: 0.000000
2023-10-08 18:34:12,689 DEV : loss 0.1463211625814438 - f1-score (micro avg) 0.8225
2023-10-08 18:34:13,565 ----------------------------------------------------------------------------------------------------
2023-10-08 18:34:13,566 Loading model from best epoch ...
2023-10-08 18:34:19,025 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
2023-10-08 18:34:25,383
Results:
- F-score (micro) 0.779
- F-score (macro) 0.4674
- Accuracy 0.6737
By class:
precision recall f1-score support
pers 0.8188 0.8129 0.8159 139
scope 0.7931 0.8915 0.8394 129
work 0.6162 0.7625 0.6816 80
loc 0.0000 0.0000 0.0000 9
date 0.0000 0.0000 0.0000 3
micro avg 0.7565 0.8028 0.7790 360
macro avg 0.4456 0.4934 0.4674 360
weighted avg 0.7373 0.8028 0.7673 360
2023-10-08 18:34:25,383 ----------------------------------------------------------------------------------------------------
|