File size: 24,998 Bytes
c154c2a |
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 21:48:25,144 ----------------------------------------------------------------------------------------------------
2023-10-08 21:48:25,145 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 21:48:25,145 ----------------------------------------------------------------------------------------------------
2023-10-08 21:48:25,145 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 21:48:25,145 ----------------------------------------------------------------------------------------------------
2023-10-08 21:48:25,145 Train: 966 sentences
2023-10-08 21:48:25,145 (train_with_dev=False, train_with_test=False)
2023-10-08 21:48:25,145 ----------------------------------------------------------------------------------------------------
2023-10-08 21:48:25,146 Training Params:
2023-10-08 21:48:25,146 - learning_rate: "0.00015"
2023-10-08 21:48:25,146 - mini_batch_size: "4"
2023-10-08 21:48:25,146 - max_epochs: "10"
2023-10-08 21:48:25,146 - shuffle: "True"
2023-10-08 21:48:25,146 ----------------------------------------------------------------------------------------------------
2023-10-08 21:48:25,146 Plugins:
2023-10-08 21:48:25,146 - TensorboardLogger
2023-10-08 21:48:25,146 - LinearScheduler | warmup_fraction: '0.1'
2023-10-08 21:48:25,146 ----------------------------------------------------------------------------------------------------
2023-10-08 21:48:25,146 Final evaluation on model from best epoch (best-model.pt)
2023-10-08 21:48:25,146 - metric: "('micro avg', 'f1-score')"
2023-10-08 21:48:25,146 ----------------------------------------------------------------------------------------------------
2023-10-08 21:48:25,146 Computation:
2023-10-08 21:48:25,146 - compute on device: cuda:0
2023-10-08 21:48:25,146 - embedding storage: none
2023-10-08 21:48:25,146 ----------------------------------------------------------------------------------------------------
2023-10-08 21:48:25,146 Model training base path: "hmbench-ajmc/fr-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs4-wsFalse-e10-lr0.00015-poolingfirst-layers-1-crfFalse-4"
2023-10-08 21:48:25,146 ----------------------------------------------------------------------------------------------------
2023-10-08 21:48:25,146 ----------------------------------------------------------------------------------------------------
2023-10-08 21:48:25,147 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-08 21:48:35,261 epoch 1 - iter 24/242 - loss 3.24936976 - time (sec): 10.11 - samples/sec: 248.78 - lr: 0.000014 - momentum: 0.000000
2023-10-08 21:48:44,932 epoch 1 - iter 48/242 - loss 3.24066758 - time (sec): 19.78 - samples/sec: 243.27 - lr: 0.000029 - momentum: 0.000000
2023-10-08 21:48:55,354 epoch 1 - iter 72/242 - loss 3.22032822 - time (sec): 30.21 - samples/sec: 246.04 - lr: 0.000044 - momentum: 0.000000
2023-10-08 21:49:05,319 epoch 1 - iter 96/242 - loss 3.17744610 - time (sec): 40.17 - samples/sec: 245.10 - lr: 0.000059 - momentum: 0.000000
2023-10-08 21:49:15,721 epoch 1 - iter 120/242 - loss 3.09588783 - time (sec): 50.57 - samples/sec: 246.51 - lr: 0.000074 - momentum: 0.000000
2023-10-08 21:49:25,624 epoch 1 - iter 144/242 - loss 2.99712605 - time (sec): 60.48 - samples/sec: 246.36 - lr: 0.000089 - momentum: 0.000000
2023-10-08 21:49:35,626 epoch 1 - iter 168/242 - loss 2.88979612 - time (sec): 70.48 - samples/sec: 245.73 - lr: 0.000104 - momentum: 0.000000
2023-10-08 21:49:45,530 epoch 1 - iter 192/242 - loss 2.78290090 - time (sec): 80.38 - samples/sec: 244.31 - lr: 0.000118 - momentum: 0.000000
2023-10-08 21:49:55,928 epoch 1 - iter 216/242 - loss 2.65496565 - time (sec): 90.78 - samples/sec: 245.28 - lr: 0.000133 - momentum: 0.000000
2023-10-08 21:50:05,679 epoch 1 - iter 240/242 - loss 2.53106870 - time (sec): 100.53 - samples/sec: 245.14 - lr: 0.000148 - momentum: 0.000000
2023-10-08 21:50:06,229 ----------------------------------------------------------------------------------------------------
2023-10-08 21:50:06,230 EPOCH 1 done: loss 2.5259 - lr: 0.000148
2023-10-08 21:50:12,701 DEV : loss 1.1719764471054077 - f1-score (micro avg) 0.0
2023-10-08 21:50:12,707 ----------------------------------------------------------------------------------------------------
2023-10-08 21:50:22,760 epoch 2 - iter 24/242 - loss 1.11332494 - time (sec): 10.05 - samples/sec: 243.75 - lr: 0.000148 - momentum: 0.000000
2023-10-08 21:50:33,542 epoch 2 - iter 48/242 - loss 0.95655996 - time (sec): 20.83 - samples/sec: 247.10 - lr: 0.000147 - momentum: 0.000000
2023-10-08 21:50:43,013 epoch 2 - iter 72/242 - loss 0.89066832 - time (sec): 30.30 - samples/sec: 242.01 - lr: 0.000145 - momentum: 0.000000
2023-10-08 21:50:53,116 epoch 2 - iter 96/242 - loss 0.82372175 - time (sec): 40.41 - samples/sec: 245.55 - lr: 0.000143 - momentum: 0.000000
2023-10-08 21:51:03,053 epoch 2 - iter 120/242 - loss 0.79572040 - time (sec): 50.34 - samples/sec: 246.74 - lr: 0.000142 - momentum: 0.000000
2023-10-08 21:51:12,702 epoch 2 - iter 144/242 - loss 0.74397531 - time (sec): 59.99 - samples/sec: 245.94 - lr: 0.000140 - momentum: 0.000000
2023-10-08 21:51:23,071 epoch 2 - iter 168/242 - loss 0.71229270 - time (sec): 70.36 - samples/sec: 246.20 - lr: 0.000139 - momentum: 0.000000
2023-10-08 21:51:33,943 epoch 2 - iter 192/242 - loss 0.67570002 - time (sec): 81.23 - samples/sec: 246.48 - lr: 0.000137 - momentum: 0.000000
2023-10-08 21:51:43,578 epoch 2 - iter 216/242 - loss 0.65912798 - time (sec): 90.87 - samples/sec: 245.00 - lr: 0.000135 - momentum: 0.000000
2023-10-08 21:51:53,476 epoch 2 - iter 240/242 - loss 0.63104209 - time (sec): 100.77 - samples/sec: 244.18 - lr: 0.000134 - momentum: 0.000000
2023-10-08 21:51:54,066 ----------------------------------------------------------------------------------------------------
2023-10-08 21:51:54,066 EPOCH 2 done: loss 0.6322 - lr: 0.000134
2023-10-08 21:52:00,544 DEV : loss 0.3960220217704773 - f1-score (micro avg) 0.0135
2023-10-08 21:52:00,550 saving best model
2023-10-08 21:52:01,401 ----------------------------------------------------------------------------------------------------
2023-10-08 21:52:10,877 epoch 3 - iter 24/242 - loss 0.40188551 - time (sec): 9.47 - samples/sec: 235.37 - lr: 0.000132 - momentum: 0.000000
2023-10-08 21:52:20,706 epoch 3 - iter 48/242 - loss 0.37672379 - time (sec): 19.30 - samples/sec: 236.06 - lr: 0.000130 - momentum: 0.000000
2023-10-08 21:52:31,476 epoch 3 - iter 72/242 - loss 0.35427374 - time (sec): 30.07 - samples/sec: 239.48 - lr: 0.000128 - momentum: 0.000000
2023-10-08 21:52:40,852 epoch 3 - iter 96/242 - loss 0.34923620 - time (sec): 39.45 - samples/sec: 240.10 - lr: 0.000127 - momentum: 0.000000
2023-10-08 21:52:50,923 epoch 3 - iter 120/242 - loss 0.33922801 - time (sec): 49.52 - samples/sec: 241.13 - lr: 0.000125 - momentum: 0.000000
2023-10-08 21:53:01,236 epoch 3 - iter 144/242 - loss 0.33523826 - time (sec): 59.83 - samples/sec: 243.12 - lr: 0.000124 - momentum: 0.000000
2023-10-08 21:53:12,104 epoch 3 - iter 168/242 - loss 0.32710972 - time (sec): 70.70 - samples/sec: 245.21 - lr: 0.000122 - momentum: 0.000000
2023-10-08 21:53:23,097 epoch 3 - iter 192/242 - loss 0.31052856 - time (sec): 81.69 - samples/sec: 246.04 - lr: 0.000120 - momentum: 0.000000
2023-10-08 21:53:32,472 epoch 3 - iter 216/242 - loss 0.30284166 - time (sec): 91.07 - samples/sec: 243.57 - lr: 0.000119 - momentum: 0.000000
2023-10-08 21:53:42,481 epoch 3 - iter 240/242 - loss 0.29493426 - time (sec): 101.08 - samples/sec: 243.64 - lr: 0.000117 - momentum: 0.000000
2023-10-08 21:53:43,061 ----------------------------------------------------------------------------------------------------
2023-10-08 21:53:43,061 EPOCH 3 done: loss 0.2940 - lr: 0.000117
2023-10-08 21:53:49,517 DEV : loss 0.24857072532176971 - f1-score (micro avg) 0.5387
2023-10-08 21:53:49,522 saving best model
2023-10-08 21:53:53,567 ----------------------------------------------------------------------------------------------------
2023-10-08 21:54:03,189 epoch 4 - iter 24/242 - loss 0.28695901 - time (sec): 9.62 - samples/sec: 244.51 - lr: 0.000115 - momentum: 0.000000
2023-10-08 21:54:13,509 epoch 4 - iter 48/242 - loss 0.25050755 - time (sec): 19.94 - samples/sec: 250.21 - lr: 0.000113 - momentum: 0.000000
2023-10-08 21:54:24,332 epoch 4 - iter 72/242 - loss 0.22200255 - time (sec): 30.76 - samples/sec: 250.20 - lr: 0.000112 - momentum: 0.000000
2023-10-08 21:54:33,704 epoch 4 - iter 96/242 - loss 0.21233290 - time (sec): 40.13 - samples/sec: 250.38 - lr: 0.000110 - momentum: 0.000000
2023-10-08 21:54:43,601 epoch 4 - iter 120/242 - loss 0.20391357 - time (sec): 50.03 - samples/sec: 253.24 - lr: 0.000109 - momentum: 0.000000
2023-10-08 21:54:53,848 epoch 4 - iter 144/242 - loss 0.19149682 - time (sec): 60.28 - samples/sec: 256.35 - lr: 0.000107 - momentum: 0.000000
2023-10-08 21:55:03,302 epoch 4 - iter 168/242 - loss 0.18804341 - time (sec): 69.73 - samples/sec: 256.97 - lr: 0.000105 - momentum: 0.000000
2023-10-08 21:55:12,735 epoch 4 - iter 192/242 - loss 0.18596665 - time (sec): 79.17 - samples/sec: 256.59 - lr: 0.000104 - momentum: 0.000000
2023-10-08 21:55:21,601 epoch 4 - iter 216/242 - loss 0.18153213 - time (sec): 88.03 - samples/sec: 256.09 - lr: 0.000102 - momentum: 0.000000
2023-10-08 21:55:30,210 epoch 4 - iter 240/242 - loss 0.18106988 - time (sec): 96.64 - samples/sec: 255.32 - lr: 0.000100 - momentum: 0.000000
2023-10-08 21:55:30,667 ----------------------------------------------------------------------------------------------------
2023-10-08 21:55:30,667 EPOCH 4 done: loss 0.1812 - lr: 0.000100
2023-10-08 21:55:36,479 DEV : loss 0.16338133811950684 - f1-score (micro avg) 0.7583
2023-10-08 21:55:36,485 saving best model
2023-10-08 21:55:40,871 ----------------------------------------------------------------------------------------------------
2023-10-08 21:55:49,856 epoch 5 - iter 24/242 - loss 0.11498730 - time (sec): 8.98 - samples/sec: 256.37 - lr: 0.000098 - momentum: 0.000000
2023-10-08 21:55:59,088 epoch 5 - iter 48/242 - loss 0.11312218 - time (sec): 18.22 - samples/sec: 257.20 - lr: 0.000097 - momentum: 0.000000
2023-10-08 21:56:08,767 epoch 5 - iter 72/242 - loss 0.11664110 - time (sec): 27.89 - samples/sec: 263.71 - lr: 0.000095 - momentum: 0.000000
2023-10-08 21:56:18,162 epoch 5 - iter 96/242 - loss 0.12130217 - time (sec): 37.29 - samples/sec: 261.06 - lr: 0.000094 - momentum: 0.000000
2023-10-08 21:56:27,699 epoch 5 - iter 120/242 - loss 0.11678047 - time (sec): 46.83 - samples/sec: 263.16 - lr: 0.000092 - momentum: 0.000000
2023-10-08 21:56:37,072 epoch 5 - iter 144/242 - loss 0.11466138 - time (sec): 56.20 - samples/sec: 262.65 - lr: 0.000090 - momentum: 0.000000
2023-10-08 21:56:46,589 epoch 5 - iter 168/242 - loss 0.12088344 - time (sec): 65.72 - samples/sec: 263.34 - lr: 0.000089 - momentum: 0.000000
2023-10-08 21:56:56,269 epoch 5 - iter 192/242 - loss 0.12137348 - time (sec): 75.40 - samples/sec: 262.53 - lr: 0.000087 - momentum: 0.000000
2023-10-08 21:57:04,904 epoch 5 - iter 216/242 - loss 0.11826361 - time (sec): 84.03 - samples/sec: 260.83 - lr: 0.000085 - momentum: 0.000000
2023-10-08 21:57:14,393 epoch 5 - iter 240/242 - loss 0.11764155 - time (sec): 93.52 - samples/sec: 261.72 - lr: 0.000084 - momentum: 0.000000
2023-10-08 21:57:15,315 ----------------------------------------------------------------------------------------------------
2023-10-08 21:57:15,315 EPOCH 5 done: loss 0.1168 - lr: 0.000084
2023-10-08 21:57:21,153 DEV : loss 0.1525898575782776 - f1-score (micro avg) 0.8204
2023-10-08 21:57:21,159 saving best model
2023-10-08 21:57:25,522 ----------------------------------------------------------------------------------------------------
2023-10-08 21:57:34,579 epoch 6 - iter 24/242 - loss 0.09059692 - time (sec): 9.06 - samples/sec: 263.15 - lr: 0.000082 - momentum: 0.000000
2023-10-08 21:57:44,205 epoch 6 - iter 48/242 - loss 0.08537644 - time (sec): 18.68 - samples/sec: 264.17 - lr: 0.000080 - momentum: 0.000000
2023-10-08 21:57:53,875 epoch 6 - iter 72/242 - loss 0.09074220 - time (sec): 28.35 - samples/sec: 265.31 - lr: 0.000079 - momentum: 0.000000
2023-10-08 21:58:02,944 epoch 6 - iter 96/242 - loss 0.08440456 - time (sec): 37.42 - samples/sec: 260.39 - lr: 0.000077 - momentum: 0.000000
2023-10-08 21:58:12,425 epoch 6 - iter 120/242 - loss 0.08569440 - time (sec): 46.90 - samples/sec: 260.53 - lr: 0.000075 - momentum: 0.000000
2023-10-08 21:58:21,866 epoch 6 - iter 144/242 - loss 0.08689621 - time (sec): 56.34 - samples/sec: 259.68 - lr: 0.000074 - momentum: 0.000000
2023-10-08 21:58:31,346 epoch 6 - iter 168/242 - loss 0.08364927 - time (sec): 65.82 - samples/sec: 260.12 - lr: 0.000072 - momentum: 0.000000
2023-10-08 21:58:41,144 epoch 6 - iter 192/242 - loss 0.08216564 - time (sec): 75.62 - samples/sec: 260.71 - lr: 0.000070 - momentum: 0.000000
2023-10-08 21:58:50,301 epoch 6 - iter 216/242 - loss 0.08163705 - time (sec): 84.78 - samples/sec: 259.85 - lr: 0.000069 - momentum: 0.000000
2023-10-08 21:58:59,914 epoch 6 - iter 240/242 - loss 0.08225899 - time (sec): 94.39 - samples/sec: 259.36 - lr: 0.000067 - momentum: 0.000000
2023-10-08 21:59:00,801 ----------------------------------------------------------------------------------------------------
2023-10-08 21:59:00,801 EPOCH 6 done: loss 0.0823 - lr: 0.000067
2023-10-08 21:59:06,845 DEV : loss 0.13395951688289642 - f1-score (micro avg) 0.8059
2023-10-08 21:59:06,851 ----------------------------------------------------------------------------------------------------
2023-10-08 21:59:17,232 epoch 7 - iter 24/242 - loss 0.04308971 - time (sec): 10.38 - samples/sec: 256.84 - lr: 0.000065 - momentum: 0.000000
2023-10-08 21:59:26,196 epoch 7 - iter 48/242 - loss 0.06023306 - time (sec): 19.34 - samples/sec: 248.86 - lr: 0.000064 - momentum: 0.000000
2023-10-08 21:59:35,484 epoch 7 - iter 72/242 - loss 0.05954315 - time (sec): 28.63 - samples/sec: 248.39 - lr: 0.000062 - momentum: 0.000000
2023-10-08 21:59:45,423 epoch 7 - iter 96/242 - loss 0.05624249 - time (sec): 38.57 - samples/sec: 247.16 - lr: 0.000060 - momentum: 0.000000
2023-10-08 21:59:54,968 epoch 7 - iter 120/242 - loss 0.05760461 - time (sec): 48.12 - samples/sec: 248.34 - lr: 0.000059 - momentum: 0.000000
2023-10-08 22:00:04,706 epoch 7 - iter 144/242 - loss 0.05798252 - time (sec): 57.85 - samples/sec: 248.27 - lr: 0.000057 - momentum: 0.000000
2023-10-08 22:00:14,766 epoch 7 - iter 168/242 - loss 0.05678083 - time (sec): 67.91 - samples/sec: 247.73 - lr: 0.000055 - momentum: 0.000000
2023-10-08 22:00:24,261 epoch 7 - iter 192/242 - loss 0.05577012 - time (sec): 77.41 - samples/sec: 247.08 - lr: 0.000054 - momentum: 0.000000
2023-10-08 22:00:35,124 epoch 7 - iter 216/242 - loss 0.06009719 - time (sec): 88.27 - samples/sec: 248.94 - lr: 0.000052 - momentum: 0.000000
2023-10-08 22:00:45,308 epoch 7 - iter 240/242 - loss 0.05937563 - time (sec): 98.46 - samples/sec: 250.43 - lr: 0.000050 - momentum: 0.000000
2023-10-08 22:00:45,842 ----------------------------------------------------------------------------------------------------
2023-10-08 22:00:45,842 EPOCH 7 done: loss 0.0594 - lr: 0.000050
2023-10-08 22:00:51,855 DEV : loss 0.15543989837169647 - f1-score (micro avg) 0.8116
2023-10-08 22:00:51,861 ----------------------------------------------------------------------------------------------------
2023-10-08 22:01:01,652 epoch 8 - iter 24/242 - loss 0.03467015 - time (sec): 9.79 - samples/sec: 265.70 - lr: 0.000049 - momentum: 0.000000
2023-10-08 22:01:10,803 epoch 8 - iter 48/242 - loss 0.04028766 - time (sec): 18.94 - samples/sec: 251.68 - lr: 0.000047 - momentum: 0.000000
2023-10-08 22:01:20,474 epoch 8 - iter 72/242 - loss 0.04085905 - time (sec): 28.61 - samples/sec: 258.92 - lr: 0.000045 - momentum: 0.000000
2023-10-08 22:01:29,605 epoch 8 - iter 96/242 - loss 0.04660420 - time (sec): 37.74 - samples/sec: 258.75 - lr: 0.000044 - momentum: 0.000000
2023-10-08 22:01:38,566 epoch 8 - iter 120/242 - loss 0.04973784 - time (sec): 46.70 - samples/sec: 256.38 - lr: 0.000042 - momentum: 0.000000
2023-10-08 22:01:48,046 epoch 8 - iter 144/242 - loss 0.05222928 - time (sec): 56.18 - samples/sec: 258.17 - lr: 0.000040 - momentum: 0.000000
2023-10-08 22:01:57,321 epoch 8 - iter 168/242 - loss 0.05142681 - time (sec): 65.46 - samples/sec: 258.20 - lr: 0.000039 - momentum: 0.000000
2023-10-08 22:02:06,790 epoch 8 - iter 192/242 - loss 0.04961141 - time (sec): 74.93 - samples/sec: 259.95 - lr: 0.000037 - momentum: 0.000000
2023-10-08 22:02:16,400 epoch 8 - iter 216/242 - loss 0.05054334 - time (sec): 84.54 - samples/sec: 260.36 - lr: 0.000035 - momentum: 0.000000
2023-10-08 22:02:25,934 epoch 8 - iter 240/242 - loss 0.04947282 - time (sec): 94.07 - samples/sec: 260.69 - lr: 0.000034 - momentum: 0.000000
2023-10-08 22:02:26,663 ----------------------------------------------------------------------------------------------------
2023-10-08 22:02:26,663 EPOCH 8 done: loss 0.0493 - lr: 0.000034
2023-10-08 22:02:32,428 DEV : loss 0.14887449145317078 - f1-score (micro avg) 0.8124
2023-10-08 22:02:32,434 ----------------------------------------------------------------------------------------------------
2023-10-08 22:02:43,185 epoch 9 - iter 24/242 - loss 0.04471556 - time (sec): 10.75 - samples/sec: 276.10 - lr: 0.000032 - momentum: 0.000000
2023-10-08 22:02:52,172 epoch 9 - iter 48/242 - loss 0.04892032 - time (sec): 19.74 - samples/sec: 269.49 - lr: 0.000030 - momentum: 0.000000
2023-10-08 22:03:00,937 epoch 9 - iter 72/242 - loss 0.04441681 - time (sec): 28.50 - samples/sec: 267.32 - lr: 0.000029 - momentum: 0.000000
2023-10-08 22:03:10,117 epoch 9 - iter 96/242 - loss 0.04682017 - time (sec): 37.68 - samples/sec: 265.64 - lr: 0.000027 - momentum: 0.000000
2023-10-08 22:03:19,439 epoch 9 - iter 120/242 - loss 0.04473883 - time (sec): 47.00 - samples/sec: 265.00 - lr: 0.000025 - momentum: 0.000000
2023-10-08 22:03:28,177 epoch 9 - iter 144/242 - loss 0.04405006 - time (sec): 55.74 - samples/sec: 263.03 - lr: 0.000024 - momentum: 0.000000
2023-10-08 22:03:38,026 epoch 9 - iter 168/242 - loss 0.04346054 - time (sec): 65.59 - samples/sec: 263.00 - lr: 0.000022 - momentum: 0.000000
2023-10-08 22:03:47,375 epoch 9 - iter 192/242 - loss 0.04284052 - time (sec): 74.94 - samples/sec: 263.32 - lr: 0.000020 - momentum: 0.000000
2023-10-08 22:03:56,583 epoch 9 - iter 216/242 - loss 0.04282043 - time (sec): 84.15 - samples/sec: 263.05 - lr: 0.000019 - momentum: 0.000000
2023-10-08 22:04:06,057 epoch 9 - iter 240/242 - loss 0.04294806 - time (sec): 93.62 - samples/sec: 262.79 - lr: 0.000017 - momentum: 0.000000
2023-10-08 22:04:06,655 ----------------------------------------------------------------------------------------------------
2023-10-08 22:04:06,655 EPOCH 9 done: loss 0.0427 - lr: 0.000017
2023-10-08 22:04:12,490 DEV : loss 0.15138302743434906 - f1-score (micro avg) 0.8165
2023-10-08 22:04:12,496 ----------------------------------------------------------------------------------------------------
2023-10-08 22:04:22,629 epoch 10 - iter 24/242 - loss 0.04445960 - time (sec): 10.13 - samples/sec: 277.85 - lr: 0.000015 - momentum: 0.000000
2023-10-08 22:04:31,391 epoch 10 - iter 48/242 - loss 0.03988212 - time (sec): 18.89 - samples/sec: 259.34 - lr: 0.000014 - momentum: 0.000000
2023-10-08 22:04:40,645 epoch 10 - iter 72/242 - loss 0.03622270 - time (sec): 28.15 - samples/sec: 260.27 - lr: 0.000012 - momentum: 0.000000
2023-10-08 22:04:49,980 epoch 10 - iter 96/242 - loss 0.03801687 - time (sec): 37.48 - samples/sec: 259.62 - lr: 0.000010 - momentum: 0.000000
2023-10-08 22:04:59,614 epoch 10 - iter 120/242 - loss 0.03744072 - time (sec): 47.12 - samples/sec: 257.94 - lr: 0.000009 - momentum: 0.000000
2023-10-08 22:05:09,252 epoch 10 - iter 144/242 - loss 0.03611060 - time (sec): 56.75 - samples/sec: 259.94 - lr: 0.000007 - momentum: 0.000000
2023-10-08 22:05:18,478 epoch 10 - iter 168/242 - loss 0.03672125 - time (sec): 65.98 - samples/sec: 260.53 - lr: 0.000005 - momentum: 0.000000
2023-10-08 22:05:27,822 epoch 10 - iter 192/242 - loss 0.03701310 - time (sec): 75.32 - samples/sec: 259.52 - lr: 0.000004 - momentum: 0.000000
2023-10-08 22:05:37,777 epoch 10 - iter 216/242 - loss 0.03824803 - time (sec): 85.28 - samples/sec: 260.66 - lr: 0.000002 - momentum: 0.000000
2023-10-08 22:05:46,925 epoch 10 - iter 240/242 - loss 0.03663811 - time (sec): 94.43 - samples/sec: 259.78 - lr: 0.000000 - momentum: 0.000000
2023-10-08 22:05:47,692 ----------------------------------------------------------------------------------------------------
2023-10-08 22:05:47,692 EPOCH 10 done: loss 0.0373 - lr: 0.000000
2023-10-08 22:05:53,853 DEV : loss 0.15766310691833496 - f1-score (micro avg) 0.8115
2023-10-08 22:05:54,761 ----------------------------------------------------------------------------------------------------
2023-10-08 22:05:54,763 Loading model from best epoch ...
2023-10-08 22:05:58,830 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 22:06:04,706
Results:
- F-score (micro) 0.7862
- F-score (macro) 0.4709
- Accuracy 0.6835
By class:
precision recall f1-score support
pers 0.8346 0.7986 0.8162 139
scope 0.8456 0.8915 0.8679 129
work 0.6146 0.7375 0.6705 80
loc 0.0000 0.0000 0.0000 9
date 0.0000 0.0000 0.0000 3
micro avg 0.7808 0.7917 0.7862 360
macro avg 0.4590 0.4855 0.4709 360
weighted avg 0.7618 0.7917 0.7751 360
2023-10-08 22:06:04,706 ----------------------------------------------------------------------------------------------------
|