File size: 24,952 Bytes
835b3f6 |
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 |
2023-10-08 18:34:53,287 ----------------------------------------------------------------------------------------------------
2023-10-08 18:34:53,288 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 18:34:53,288 ----------------------------------------------------------------------------------------------------
2023-10-08 18:34:53,288 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 18:34:53,288 ----------------------------------------------------------------------------------------------------
2023-10-08 18:34:53,288 Train: 966 sentences
2023-10-08 18:34:53,288 (train_with_dev=False, train_with_test=False)
2023-10-08 18:34:53,288 ----------------------------------------------------------------------------------------------------
2023-10-08 18:34:53,288 Training Params:
2023-10-08 18:34:53,288 - learning_rate: "0.00016"
2023-10-08 18:34:53,289 - mini_batch_size: "4"
2023-10-08 18:34:53,289 - max_epochs: "10"
2023-10-08 18:34:53,289 - shuffle: "True"
2023-10-08 18:34:53,289 ----------------------------------------------------------------------------------------------------
2023-10-08 18:34:53,289 Plugins:
2023-10-08 18:34:53,289 - TensorboardLogger
2023-10-08 18:34:53,289 - LinearScheduler | warmup_fraction: '0.1'
2023-10-08 18:34:53,289 ----------------------------------------------------------------------------------------------------
2023-10-08 18:34:53,289 Final evaluation on model from best epoch (best-model.pt)
2023-10-08 18:34:53,289 - metric: "('micro avg', 'f1-score')"
2023-10-08 18:34:53,289 ----------------------------------------------------------------------------------------------------
2023-10-08 18:34:53,289 Computation:
2023-10-08 18:34:53,289 - compute on device: cuda:0
2023-10-08 18:34:53,289 - embedding storage: none
2023-10-08 18:34:53,289 ----------------------------------------------------------------------------------------------------
2023-10-08 18:34:53,289 Model training base path: "hmbench-ajmc/fr-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs4-wsFalse-e10-lr0.00016-poolingfirst-layers-1-crfFalse-1"
2023-10-08 18:34:53,289 ----------------------------------------------------------------------------------------------------
2023-10-08 18:34:53,289 ----------------------------------------------------------------------------------------------------
2023-10-08 18:34:53,289 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-08 18:35:02,673 epoch 1 - iter 24/242 - loss 3.23249107 - time (sec): 9.38 - samples/sec: 233.30 - lr: 0.000015 - momentum: 0.000000
2023-10-08 18:35:12,678 epoch 1 - iter 48/242 - loss 3.22160902 - time (sec): 19.39 - samples/sec: 239.60 - lr: 0.000031 - momentum: 0.000000
2023-10-08 18:35:22,313 epoch 1 - iter 72/242 - loss 3.20323790 - time (sec): 29.02 - samples/sec: 238.81 - lr: 0.000047 - momentum: 0.000000
2023-10-08 18:35:32,243 epoch 1 - iter 96/242 - loss 3.14989613 - time (sec): 38.95 - samples/sec: 242.14 - lr: 0.000063 - momentum: 0.000000
2023-10-08 18:35:42,136 epoch 1 - iter 120/242 - loss 3.06221572 - time (sec): 48.85 - samples/sec: 241.07 - lr: 0.000079 - momentum: 0.000000
2023-10-08 18:35:52,824 epoch 1 - iter 144/242 - loss 2.95428246 - time (sec): 59.53 - samples/sec: 242.12 - lr: 0.000095 - momentum: 0.000000
2023-10-08 18:36:02,538 epoch 1 - iter 168/242 - loss 2.85101367 - time (sec): 69.25 - samples/sec: 241.73 - lr: 0.000110 - momentum: 0.000000
2023-10-08 18:36:12,394 epoch 1 - iter 192/242 - loss 2.74034409 - time (sec): 79.10 - samples/sec: 240.62 - lr: 0.000126 - momentum: 0.000000
2023-10-08 18:36:23,043 epoch 1 - iter 216/242 - loss 2.60212408 - time (sec): 89.75 - samples/sec: 242.18 - lr: 0.000142 - momentum: 0.000000
2023-10-08 18:36:33,989 epoch 1 - iter 240/242 - loss 2.45380672 - time (sec): 100.70 - samples/sec: 243.43 - lr: 0.000158 - momentum: 0.000000
2023-10-08 18:36:34,764 ----------------------------------------------------------------------------------------------------
2023-10-08 18:36:34,764 EPOCH 1 done: loss 2.4415 - lr: 0.000158
2023-10-08 18:36:41,310 DEV : loss 1.10421884059906 - f1-score (micro avg) 0.0
2023-10-08 18:36:41,316 ----------------------------------------------------------------------------------------------------
2023-10-08 18:36:51,047 epoch 2 - iter 24/242 - loss 1.08751151 - time (sec): 9.73 - samples/sec: 233.42 - lr: 0.000158 - momentum: 0.000000
2023-10-08 18:37:00,942 epoch 2 - iter 48/242 - loss 0.95694896 - time (sec): 19.62 - samples/sec: 237.71 - lr: 0.000157 - momentum: 0.000000
2023-10-08 18:37:10,777 epoch 2 - iter 72/242 - loss 0.87876685 - time (sec): 29.46 - samples/sec: 241.59 - lr: 0.000155 - momentum: 0.000000
2023-10-08 18:37:20,908 epoch 2 - iter 96/242 - loss 0.81302044 - time (sec): 39.59 - samples/sec: 243.27 - lr: 0.000153 - momentum: 0.000000
2023-10-08 18:37:30,907 epoch 2 - iter 120/242 - loss 0.76456788 - time (sec): 49.59 - samples/sec: 242.03 - lr: 0.000151 - momentum: 0.000000
2023-10-08 18:37:41,488 epoch 2 - iter 144/242 - loss 0.71887917 - time (sec): 60.17 - samples/sec: 244.47 - lr: 0.000150 - momentum: 0.000000
2023-10-08 18:37:52,468 epoch 2 - iter 168/242 - loss 0.67430178 - time (sec): 71.15 - samples/sec: 243.56 - lr: 0.000148 - momentum: 0.000000
2023-10-08 18:38:02,384 epoch 2 - iter 192/242 - loss 0.63672173 - time (sec): 81.07 - samples/sec: 243.01 - lr: 0.000146 - momentum: 0.000000
2023-10-08 18:38:12,147 epoch 2 - iter 216/242 - loss 0.60703481 - time (sec): 90.83 - samples/sec: 242.11 - lr: 0.000144 - momentum: 0.000000
2023-10-08 18:38:22,686 epoch 2 - iter 240/242 - loss 0.57778911 - time (sec): 101.37 - samples/sec: 242.90 - lr: 0.000142 - momentum: 0.000000
2023-10-08 18:38:23,263 ----------------------------------------------------------------------------------------------------
2023-10-08 18:38:23,264 EPOCH 2 done: loss 0.5768 - lr: 0.000142
2023-10-08 18:38:29,811 DEV : loss 0.36735957860946655 - f1-score (micro avg) 0.4525
2023-10-08 18:38:29,817 saving best model
2023-10-08 18:38:30,707 ----------------------------------------------------------------------------------------------------
2023-10-08 18:38:40,732 epoch 3 - iter 24/242 - loss 0.33886112 - time (sec): 10.02 - samples/sec: 258.08 - lr: 0.000141 - momentum: 0.000000
2023-10-08 18:38:50,182 epoch 3 - iter 48/242 - loss 0.35382038 - time (sec): 19.47 - samples/sec: 259.42 - lr: 0.000139 - momentum: 0.000000
2023-10-08 18:39:00,142 epoch 3 - iter 72/242 - loss 0.32381848 - time (sec): 29.43 - samples/sec: 263.07 - lr: 0.000137 - momentum: 0.000000
2023-10-08 18:39:09,488 epoch 3 - iter 96/242 - loss 0.31041884 - time (sec): 38.78 - samples/sec: 259.78 - lr: 0.000135 - momentum: 0.000000
2023-10-08 18:39:18,559 epoch 3 - iter 120/242 - loss 0.29354167 - time (sec): 47.85 - samples/sec: 258.39 - lr: 0.000134 - momentum: 0.000000
2023-10-08 18:39:27,561 epoch 3 - iter 144/242 - loss 0.28584160 - time (sec): 56.85 - samples/sec: 257.77 - lr: 0.000132 - momentum: 0.000000
2023-10-08 18:39:37,506 epoch 3 - iter 168/242 - loss 0.28154401 - time (sec): 66.80 - samples/sec: 258.98 - lr: 0.000130 - momentum: 0.000000
2023-10-08 18:39:47,261 epoch 3 - iter 192/242 - loss 0.27573225 - time (sec): 76.55 - samples/sec: 260.33 - lr: 0.000128 - momentum: 0.000000
2023-10-08 18:39:56,248 epoch 3 - iter 216/242 - loss 0.26727131 - time (sec): 85.54 - samples/sec: 259.38 - lr: 0.000126 - momentum: 0.000000
2023-10-08 18:40:05,584 epoch 3 - iter 240/242 - loss 0.26232951 - time (sec): 94.88 - samples/sec: 259.21 - lr: 0.000125 - momentum: 0.000000
2023-10-08 18:40:06,193 ----------------------------------------------------------------------------------------------------
2023-10-08 18:40:06,193 EPOCH 3 done: loss 0.2619 - lr: 0.000125
2023-10-08 18:40:12,021 DEV : loss 0.20984935760498047 - f1-score (micro avg) 0.5559
2023-10-08 18:40:12,027 saving best model
2023-10-08 18:40:16,434 ----------------------------------------------------------------------------------------------------
2023-10-08 18:40:25,028 epoch 4 - iter 24/242 - loss 0.21633186 - time (sec): 8.59 - samples/sec: 255.58 - lr: 0.000123 - momentum: 0.000000
2023-10-08 18:40:34,737 epoch 4 - iter 48/242 - loss 0.20499870 - time (sec): 18.30 - samples/sec: 264.52 - lr: 0.000121 - momentum: 0.000000
2023-10-08 18:40:44,184 epoch 4 - iter 72/242 - loss 0.19040422 - time (sec): 27.75 - samples/sec: 262.04 - lr: 0.000119 - momentum: 0.000000
2023-10-08 18:40:53,336 epoch 4 - iter 96/242 - loss 0.17803656 - time (sec): 36.90 - samples/sec: 262.39 - lr: 0.000118 - momentum: 0.000000
2023-10-08 18:41:02,300 epoch 4 - iter 120/242 - loss 0.17506133 - time (sec): 45.86 - samples/sec: 262.28 - lr: 0.000116 - momentum: 0.000000
2023-10-08 18:41:12,466 epoch 4 - iter 144/242 - loss 0.17147606 - time (sec): 56.03 - samples/sec: 264.39 - lr: 0.000114 - momentum: 0.000000
2023-10-08 18:41:22,097 epoch 4 - iter 168/242 - loss 0.16370940 - time (sec): 65.66 - samples/sec: 264.89 - lr: 0.000112 - momentum: 0.000000
2023-10-08 18:41:31,301 epoch 4 - iter 192/242 - loss 0.16311405 - time (sec): 74.87 - samples/sec: 263.15 - lr: 0.000110 - momentum: 0.000000
2023-10-08 18:41:40,791 epoch 4 - iter 216/242 - loss 0.16120081 - time (sec): 84.35 - samples/sec: 262.63 - lr: 0.000109 - momentum: 0.000000
2023-10-08 18:41:50,100 epoch 4 - iter 240/242 - loss 0.15725999 - time (sec): 93.66 - samples/sec: 262.00 - lr: 0.000107 - momentum: 0.000000
2023-10-08 18:41:50,796 ----------------------------------------------------------------------------------------------------
2023-10-08 18:41:50,796 EPOCH 4 done: loss 0.1573 - lr: 0.000107
2023-10-08 18:41:56,634 DEV : loss 0.14482806622982025 - f1-score (micro avg) 0.8417
2023-10-08 18:41:56,640 saving best model
2023-10-08 18:42:01,009 ----------------------------------------------------------------------------------------------------
2023-10-08 18:42:10,593 epoch 5 - iter 24/242 - loss 0.14974282 - time (sec): 9.58 - samples/sec: 264.24 - lr: 0.000105 - momentum: 0.000000
2023-10-08 18:42:19,980 epoch 5 - iter 48/242 - loss 0.12573972 - time (sec): 18.97 - samples/sec: 259.74 - lr: 0.000103 - momentum: 0.000000
2023-10-08 18:42:29,922 epoch 5 - iter 72/242 - loss 0.12078938 - time (sec): 28.91 - samples/sec: 266.61 - lr: 0.000102 - momentum: 0.000000
2023-10-08 18:42:39,756 epoch 5 - iter 96/242 - loss 0.11455455 - time (sec): 38.74 - samples/sec: 268.40 - lr: 0.000100 - momentum: 0.000000
2023-10-08 18:42:49,278 epoch 5 - iter 120/242 - loss 0.11444902 - time (sec): 48.27 - samples/sec: 266.29 - lr: 0.000098 - momentum: 0.000000
2023-10-08 18:42:58,251 epoch 5 - iter 144/242 - loss 0.11447821 - time (sec): 57.24 - samples/sec: 264.08 - lr: 0.000096 - momentum: 0.000000
2023-10-08 18:43:07,744 epoch 5 - iter 168/242 - loss 0.10989477 - time (sec): 66.73 - samples/sec: 264.61 - lr: 0.000094 - momentum: 0.000000
2023-10-08 18:43:16,374 epoch 5 - iter 192/242 - loss 0.11006937 - time (sec): 75.36 - samples/sec: 263.09 - lr: 0.000093 - momentum: 0.000000
2023-10-08 18:43:25,524 epoch 5 - iter 216/242 - loss 0.10934520 - time (sec): 84.51 - samples/sec: 262.35 - lr: 0.000091 - momentum: 0.000000
2023-10-08 18:43:34,623 epoch 5 - iter 240/242 - loss 0.10539988 - time (sec): 93.61 - samples/sec: 262.11 - lr: 0.000089 - momentum: 0.000000
2023-10-08 18:43:35,364 ----------------------------------------------------------------------------------------------------
2023-10-08 18:43:35,365 EPOCH 5 done: loss 0.1050 - lr: 0.000089
2023-10-08 18:43:41,168 DEV : loss 0.13649354875087738 - f1-score (micro avg) 0.8192
2023-10-08 18:43:41,174 ----------------------------------------------------------------------------------------------------
2023-10-08 18:43:50,124 epoch 6 - iter 24/242 - loss 0.08990859 - time (sec): 8.95 - samples/sec: 251.23 - lr: 0.000087 - momentum: 0.000000
2023-10-08 18:43:59,640 epoch 6 - iter 48/242 - loss 0.08114002 - time (sec): 18.46 - samples/sec: 259.64 - lr: 0.000086 - momentum: 0.000000
2023-10-08 18:44:08,671 epoch 6 - iter 72/242 - loss 0.07601968 - time (sec): 27.50 - samples/sec: 258.12 - lr: 0.000084 - momentum: 0.000000
2023-10-08 18:44:18,225 epoch 6 - iter 96/242 - loss 0.07610495 - time (sec): 37.05 - samples/sec: 260.00 - lr: 0.000082 - momentum: 0.000000
2023-10-08 18:44:27,796 epoch 6 - iter 120/242 - loss 0.07416057 - time (sec): 46.62 - samples/sec: 261.13 - lr: 0.000080 - momentum: 0.000000
2023-10-08 18:44:37,627 epoch 6 - iter 144/242 - loss 0.07373746 - time (sec): 56.45 - samples/sec: 263.64 - lr: 0.000078 - momentum: 0.000000
2023-10-08 18:44:46,994 epoch 6 - iter 168/242 - loss 0.07727814 - time (sec): 65.82 - samples/sec: 265.41 - lr: 0.000077 - momentum: 0.000000
2023-10-08 18:44:56,080 epoch 6 - iter 192/242 - loss 0.07713696 - time (sec): 74.90 - samples/sec: 263.20 - lr: 0.000075 - momentum: 0.000000
2023-10-08 18:45:04,971 epoch 6 - iter 216/242 - loss 0.07461044 - time (sec): 83.80 - samples/sec: 262.06 - lr: 0.000073 - momentum: 0.000000
2023-10-08 18:45:14,841 epoch 6 - iter 240/242 - loss 0.07542934 - time (sec): 93.67 - samples/sec: 261.63 - lr: 0.000071 - momentum: 0.000000
2023-10-08 18:45:15,674 ----------------------------------------------------------------------------------------------------
2023-10-08 18:45:15,675 EPOCH 6 done: loss 0.0755 - lr: 0.000071
2023-10-08 18:45:21,569 DEV : loss 0.12205741554498672 - f1-score (micro avg) 0.8346
2023-10-08 18:45:21,576 ----------------------------------------------------------------------------------------------------
2023-10-08 18:45:29,908 epoch 7 - iter 24/242 - loss 0.04051827 - time (sec): 8.33 - samples/sec: 235.05 - lr: 0.000070 - momentum: 0.000000
2023-10-08 18:45:39,504 epoch 7 - iter 48/242 - loss 0.05486241 - time (sec): 17.93 - samples/sec: 254.15 - lr: 0.000068 - momentum: 0.000000
2023-10-08 18:45:48,946 epoch 7 - iter 72/242 - loss 0.05684543 - time (sec): 27.37 - samples/sec: 255.08 - lr: 0.000066 - momentum: 0.000000
2023-10-08 18:45:58,785 epoch 7 - iter 96/242 - loss 0.06216082 - time (sec): 37.21 - samples/sec: 257.83 - lr: 0.000064 - momentum: 0.000000
2023-10-08 18:46:08,303 epoch 7 - iter 120/242 - loss 0.05942616 - time (sec): 46.73 - samples/sec: 258.28 - lr: 0.000062 - momentum: 0.000000
2023-10-08 18:46:17,017 epoch 7 - iter 144/242 - loss 0.05433522 - time (sec): 55.44 - samples/sec: 255.47 - lr: 0.000061 - momentum: 0.000000
2023-10-08 18:46:26,640 epoch 7 - iter 168/242 - loss 0.05796718 - time (sec): 65.06 - samples/sec: 255.62 - lr: 0.000059 - momentum: 0.000000
2023-10-08 18:46:36,464 epoch 7 - iter 192/242 - loss 0.05899844 - time (sec): 74.89 - samples/sec: 256.18 - lr: 0.000057 - momentum: 0.000000
2023-10-08 18:46:46,862 epoch 7 - iter 216/242 - loss 0.05910219 - time (sec): 85.28 - samples/sec: 256.92 - lr: 0.000055 - momentum: 0.000000
2023-10-08 18:46:57,126 epoch 7 - iter 240/242 - loss 0.05689392 - time (sec): 95.55 - samples/sec: 257.31 - lr: 0.000054 - momentum: 0.000000
2023-10-08 18:46:57,730 ----------------------------------------------------------------------------------------------------
2023-10-08 18:46:57,730 EPOCH 7 done: loss 0.0568 - lr: 0.000054
2023-10-08 18:47:03,836 DEV : loss 0.12814383208751678 - f1-score (micro avg) 0.8224
2023-10-08 18:47:03,842 ----------------------------------------------------------------------------------------------------
2023-10-08 18:47:13,018 epoch 8 - iter 24/242 - loss 0.05203825 - time (sec): 9.17 - samples/sec: 229.12 - lr: 0.000052 - momentum: 0.000000
2023-10-08 18:47:22,854 epoch 8 - iter 48/242 - loss 0.05392342 - time (sec): 19.01 - samples/sec: 244.54 - lr: 0.000050 - momentum: 0.000000
2023-10-08 18:47:33,063 epoch 8 - iter 72/242 - loss 0.05134362 - time (sec): 29.22 - samples/sec: 250.58 - lr: 0.000048 - momentum: 0.000000
2023-10-08 18:47:42,930 epoch 8 - iter 96/242 - loss 0.04471543 - time (sec): 39.09 - samples/sec: 250.19 - lr: 0.000046 - momentum: 0.000000
2023-10-08 18:47:53,157 epoch 8 - iter 120/242 - loss 0.04434398 - time (sec): 49.31 - samples/sec: 249.87 - lr: 0.000045 - momentum: 0.000000
2023-10-08 18:48:03,273 epoch 8 - iter 144/242 - loss 0.04464736 - time (sec): 59.43 - samples/sec: 250.14 - lr: 0.000043 - momentum: 0.000000
2023-10-08 18:48:13,650 epoch 8 - iter 168/242 - loss 0.04805001 - time (sec): 69.81 - samples/sec: 250.66 - lr: 0.000041 - momentum: 0.000000
2023-10-08 18:48:23,745 epoch 8 - iter 192/242 - loss 0.04571917 - time (sec): 79.90 - samples/sec: 249.68 - lr: 0.000039 - momentum: 0.000000
2023-10-08 18:48:33,544 epoch 8 - iter 216/242 - loss 0.04399139 - time (sec): 89.70 - samples/sec: 248.29 - lr: 0.000038 - momentum: 0.000000
2023-10-08 18:48:43,388 epoch 8 - iter 240/242 - loss 0.04459372 - time (sec): 99.54 - samples/sec: 247.39 - lr: 0.000036 - momentum: 0.000000
2023-10-08 18:48:43,916 ----------------------------------------------------------------------------------------------------
2023-10-08 18:48:43,917 EPOCH 8 done: loss 0.0445 - lr: 0.000036
2023-10-08 18:48:50,468 DEV : loss 0.13605128228664398 - f1-score (micro avg) 0.8141
2023-10-08 18:48:50,475 ----------------------------------------------------------------------------------------------------
2023-10-08 18:49:00,109 epoch 9 - iter 24/242 - loss 0.05381900 - time (sec): 9.63 - samples/sec: 231.71 - lr: 0.000034 - momentum: 0.000000
2023-10-08 18:49:09,640 epoch 9 - iter 48/242 - loss 0.04401043 - time (sec): 19.16 - samples/sec: 229.50 - lr: 0.000032 - momentum: 0.000000
2023-10-08 18:49:20,009 epoch 9 - iter 72/242 - loss 0.03846786 - time (sec): 29.53 - samples/sec: 233.84 - lr: 0.000030 - momentum: 0.000000
2023-10-08 18:49:30,133 epoch 9 - iter 96/242 - loss 0.04296229 - time (sec): 39.66 - samples/sec: 238.32 - lr: 0.000029 - momentum: 0.000000
2023-10-08 18:49:40,670 epoch 9 - iter 120/242 - loss 0.04404865 - time (sec): 50.19 - samples/sec: 241.47 - lr: 0.000027 - momentum: 0.000000
2023-10-08 18:49:51,280 epoch 9 - iter 144/242 - loss 0.04306141 - time (sec): 60.80 - samples/sec: 241.58 - lr: 0.000025 - momentum: 0.000000
2023-10-08 18:50:01,468 epoch 9 - iter 168/242 - loss 0.04150438 - time (sec): 70.99 - samples/sec: 242.14 - lr: 0.000023 - momentum: 0.000000
2023-10-08 18:50:11,340 epoch 9 - iter 192/242 - loss 0.04021125 - time (sec): 80.86 - samples/sec: 242.11 - lr: 0.000022 - momentum: 0.000000
2023-10-08 18:50:21,375 epoch 9 - iter 216/242 - loss 0.04084934 - time (sec): 90.90 - samples/sec: 242.30 - lr: 0.000020 - momentum: 0.000000
2023-10-08 18:50:31,565 epoch 9 - iter 240/242 - loss 0.03905588 - time (sec): 101.09 - samples/sec: 242.90 - lr: 0.000018 - momentum: 0.000000
2023-10-08 18:50:32,262 ----------------------------------------------------------------------------------------------------
2023-10-08 18:50:32,263 EPOCH 9 done: loss 0.0393 - lr: 0.000018
2023-10-08 18:50:38,666 DEV : loss 0.1348445564508438 - f1-score (micro avg) 0.8256
2023-10-08 18:50:38,672 ----------------------------------------------------------------------------------------------------
2023-10-08 18:50:48,750 epoch 10 - iter 24/242 - loss 0.03126737 - time (sec): 10.08 - samples/sec: 243.34 - lr: 0.000016 - momentum: 0.000000
2023-10-08 18:50:59,044 epoch 10 - iter 48/242 - loss 0.02421919 - time (sec): 20.37 - samples/sec: 248.74 - lr: 0.000014 - momentum: 0.000000
2023-10-08 18:51:08,723 epoch 10 - iter 72/242 - loss 0.02499859 - time (sec): 30.05 - samples/sec: 250.55 - lr: 0.000013 - momentum: 0.000000
2023-10-08 18:51:18,437 epoch 10 - iter 96/242 - loss 0.03022215 - time (sec): 39.76 - samples/sec: 254.00 - lr: 0.000011 - momentum: 0.000000
2023-10-08 18:51:27,131 epoch 10 - iter 120/242 - loss 0.03176207 - time (sec): 48.46 - samples/sec: 252.18 - lr: 0.000009 - momentum: 0.000000
2023-10-08 18:51:36,241 epoch 10 - iter 144/242 - loss 0.03286198 - time (sec): 57.57 - samples/sec: 252.07 - lr: 0.000007 - momentum: 0.000000
2023-10-08 18:51:45,628 epoch 10 - iter 168/242 - loss 0.03336583 - time (sec): 66.95 - samples/sec: 253.26 - lr: 0.000006 - momentum: 0.000000
2023-10-08 18:51:55,144 epoch 10 - iter 192/242 - loss 0.03282448 - time (sec): 76.47 - samples/sec: 254.72 - lr: 0.000004 - momentum: 0.000000
2023-10-08 18:52:04,668 epoch 10 - iter 216/242 - loss 0.03205261 - time (sec): 86.00 - samples/sec: 255.50 - lr: 0.000002 - momentum: 0.000000
2023-10-08 18:52:14,309 epoch 10 - iter 240/242 - loss 0.03366127 - time (sec): 95.64 - samples/sec: 256.99 - lr: 0.000000 - momentum: 0.000000
2023-10-08 18:52:15,017 ----------------------------------------------------------------------------------------------------
2023-10-08 18:52:15,017 EPOCH 10 done: loss 0.0337 - lr: 0.000000
2023-10-08 18:52:20,852 DEV : loss 0.13635611534118652 - f1-score (micro avg) 0.8275
2023-10-08 18:52:21,698 ----------------------------------------------------------------------------------------------------
2023-10-08 18:52:21,699 Loading model from best epoch ...
2023-10-08 18:52:24,820 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 18:52:30,519
Results:
- F-score (micro) 0.7717
- F-score (macro) 0.4607
- Accuracy 0.6598
By class:
precision recall f1-score support
pers 0.7532 0.8561 0.8013 139
scope 0.7733 0.8992 0.8315 129
work 0.6667 0.6750 0.6708 80
loc 0.0000 0.0000 0.0000 9
date 0.0000 0.0000 0.0000 3
micro avg 0.7429 0.8028 0.7717 360
macro avg 0.4386 0.4861 0.4607 360
weighted avg 0.7161 0.8028 0.7564 360
2023-10-08 18:52:30,519 ----------------------------------------------------------------------------------------------------
|