File size: 25,253 Bytes
b843ded |
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 |
2023-10-11 03:16:41,684 ----------------------------------------------------------------------------------------------------
2023-10-11 03:16:41,686 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): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True)
(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): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True)
(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): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True)
(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): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
)
(final_layer_norm): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
(locked_dropout): LockedDropout(p=0.5)
(linear): Linear(in_features=1472, out_features=17, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-11 03:16:41,686 ----------------------------------------------------------------------------------------------------
2023-10-11 03:16:41,686 MultiCorpus: 1166 train + 165 dev + 415 test sentences
- NER_HIPE_2022 Corpus: 1166 train + 165 dev + 415 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/newseye/fi/with_doc_seperator
2023-10-11 03:16:41,686 ----------------------------------------------------------------------------------------------------
2023-10-11 03:16:41,687 Train: 1166 sentences
2023-10-11 03:16:41,687 (train_with_dev=False, train_with_test=False)
2023-10-11 03:16:41,687 ----------------------------------------------------------------------------------------------------
2023-10-11 03:16:41,687 Training Params:
2023-10-11 03:16:41,687 - learning_rate: "0.00015"
2023-10-11 03:16:41,687 - mini_batch_size: "4"
2023-10-11 03:16:41,687 - max_epochs: "10"
2023-10-11 03:16:41,687 - shuffle: "True"
2023-10-11 03:16:41,687 ----------------------------------------------------------------------------------------------------
2023-10-11 03:16:41,687 Plugins:
2023-10-11 03:16:41,687 - TensorboardLogger
2023-10-11 03:16:41,687 - LinearScheduler | warmup_fraction: '0.1'
2023-10-11 03:16:41,687 ----------------------------------------------------------------------------------------------------
2023-10-11 03:16:41,687 Final evaluation on model from best epoch (best-model.pt)
2023-10-11 03:16:41,688 - metric: "('micro avg', 'f1-score')"
2023-10-11 03:16:41,688 ----------------------------------------------------------------------------------------------------
2023-10-11 03:16:41,688 Computation:
2023-10-11 03:16:41,688 - compute on device: cuda:0
2023-10-11 03:16:41,688 - embedding storage: none
2023-10-11 03:16:41,688 ----------------------------------------------------------------------------------------------------
2023-10-11 03:16:41,688 Model training base path: "hmbench-newseye/fi-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs4-wsFalse-e10-lr0.00015-poolingfirst-layers-1-crfFalse-5"
2023-10-11 03:16:41,688 ----------------------------------------------------------------------------------------------------
2023-10-11 03:16:41,688 ----------------------------------------------------------------------------------------------------
2023-10-11 03:16:41,688 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-11 03:16:51,822 epoch 1 - iter 29/292 - loss 2.81977350 - time (sec): 10.13 - samples/sec: 425.59 - lr: 0.000014 - momentum: 0.000000
2023-10-11 03:17:01,805 epoch 1 - iter 58/292 - loss 2.80688640 - time (sec): 20.12 - samples/sec: 410.33 - lr: 0.000029 - momentum: 0.000000
2023-10-11 03:17:12,096 epoch 1 - iter 87/292 - loss 2.78279036 - time (sec): 30.41 - samples/sec: 401.92 - lr: 0.000044 - momentum: 0.000000
2023-10-11 03:17:23,906 epoch 1 - iter 116/292 - loss 2.71003813 - time (sec): 42.22 - samples/sec: 413.94 - lr: 0.000059 - momentum: 0.000000
2023-10-11 03:17:34,740 epoch 1 - iter 145/292 - loss 2.62475305 - time (sec): 53.05 - samples/sec: 411.03 - lr: 0.000074 - momentum: 0.000000
2023-10-11 03:17:45,468 epoch 1 - iter 174/292 - loss 2.52043100 - time (sec): 63.78 - samples/sec: 419.27 - lr: 0.000089 - momentum: 0.000000
2023-10-11 03:17:56,403 epoch 1 - iter 203/292 - loss 2.39946272 - time (sec): 74.71 - samples/sec: 432.12 - lr: 0.000104 - momentum: 0.000000
2023-10-11 03:18:06,391 epoch 1 - iter 232/292 - loss 2.29647806 - time (sec): 84.70 - samples/sec: 430.10 - lr: 0.000119 - momentum: 0.000000
2023-10-11 03:18:16,949 epoch 1 - iter 261/292 - loss 2.17952313 - time (sec): 95.26 - samples/sec: 427.15 - lr: 0.000134 - momentum: 0.000000
2023-10-11 03:18:26,723 epoch 1 - iter 290/292 - loss 2.07842445 - time (sec): 105.03 - samples/sec: 422.13 - lr: 0.000148 - momentum: 0.000000
2023-10-11 03:18:27,176 ----------------------------------------------------------------------------------------------------
2023-10-11 03:18:27,176 EPOCH 1 done: loss 2.0752 - lr: 0.000148
2023-10-11 03:18:32,996 DEV : loss 0.7023366689682007 - f1-score (micro avg) 0.0
2023-10-11 03:18:33,006 ----------------------------------------------------------------------------------------------------
2023-10-11 03:18:42,653 epoch 2 - iter 29/292 - loss 0.73892904 - time (sec): 9.65 - samples/sec: 453.80 - lr: 0.000148 - momentum: 0.000000
2023-10-11 03:18:52,419 epoch 2 - iter 58/292 - loss 0.67633155 - time (sec): 19.41 - samples/sec: 442.23 - lr: 0.000147 - momentum: 0.000000
2023-10-11 03:19:02,214 epoch 2 - iter 87/292 - loss 0.66210392 - time (sec): 29.21 - samples/sec: 434.94 - lr: 0.000145 - momentum: 0.000000
2023-10-11 03:19:12,942 epoch 2 - iter 116/292 - loss 0.60646101 - time (sec): 39.93 - samples/sec: 435.65 - lr: 0.000143 - momentum: 0.000000
2023-10-11 03:19:23,013 epoch 2 - iter 145/292 - loss 0.65710441 - time (sec): 50.00 - samples/sec: 445.08 - lr: 0.000142 - momentum: 0.000000
2023-10-11 03:19:33,309 epoch 2 - iter 174/292 - loss 0.62278384 - time (sec): 60.30 - samples/sec: 443.37 - lr: 0.000140 - momentum: 0.000000
2023-10-11 03:19:43,005 epoch 2 - iter 203/292 - loss 0.59076254 - time (sec): 70.00 - samples/sec: 446.10 - lr: 0.000138 - momentum: 0.000000
2023-10-11 03:19:51,530 epoch 2 - iter 232/292 - loss 0.56424704 - time (sec): 78.52 - samples/sec: 444.22 - lr: 0.000137 - momentum: 0.000000
2023-10-11 03:20:01,917 epoch 2 - iter 261/292 - loss 0.55417964 - time (sec): 88.91 - samples/sec: 438.36 - lr: 0.000135 - momentum: 0.000000
2023-10-11 03:20:13,521 epoch 2 - iter 290/292 - loss 0.53594965 - time (sec): 100.51 - samples/sec: 439.75 - lr: 0.000134 - momentum: 0.000000
2023-10-11 03:20:14,065 ----------------------------------------------------------------------------------------------------
2023-10-11 03:20:14,066 EPOCH 2 done: loss 0.5351 - lr: 0.000134
2023-10-11 03:20:20,195 DEV : loss 0.297545462846756 - f1-score (micro avg) 0.0
2023-10-11 03:20:20,205 ----------------------------------------------------------------------------------------------------
2023-10-11 03:20:30,000 epoch 3 - iter 29/292 - loss 0.37459476 - time (sec): 9.79 - samples/sec: 407.46 - lr: 0.000132 - momentum: 0.000000
2023-10-11 03:20:38,908 epoch 3 - iter 58/292 - loss 0.36871425 - time (sec): 18.70 - samples/sec: 425.66 - lr: 0.000130 - momentum: 0.000000
2023-10-11 03:20:48,697 epoch 3 - iter 87/292 - loss 0.34885003 - time (sec): 28.49 - samples/sec: 439.40 - lr: 0.000128 - momentum: 0.000000
2023-10-11 03:20:58,273 epoch 3 - iter 116/292 - loss 0.32808207 - time (sec): 38.06 - samples/sec: 451.47 - lr: 0.000127 - momentum: 0.000000
2023-10-11 03:21:07,349 epoch 3 - iter 145/292 - loss 0.32374212 - time (sec): 47.14 - samples/sec: 452.13 - lr: 0.000125 - momentum: 0.000000
2023-10-11 03:21:17,462 epoch 3 - iter 174/292 - loss 0.30495891 - time (sec): 57.25 - samples/sec: 455.18 - lr: 0.000123 - momentum: 0.000000
2023-10-11 03:21:27,598 epoch 3 - iter 203/292 - loss 0.32430122 - time (sec): 67.39 - samples/sec: 455.47 - lr: 0.000122 - momentum: 0.000000
2023-10-11 03:21:36,787 epoch 3 - iter 232/292 - loss 0.32165796 - time (sec): 76.58 - samples/sec: 452.27 - lr: 0.000120 - momentum: 0.000000
2023-10-11 03:21:47,217 epoch 3 - iter 261/292 - loss 0.31557249 - time (sec): 87.01 - samples/sec: 456.71 - lr: 0.000119 - momentum: 0.000000
2023-10-11 03:21:56,871 epoch 3 - iter 290/292 - loss 0.31177726 - time (sec): 96.66 - samples/sec: 456.37 - lr: 0.000117 - momentum: 0.000000
2023-10-11 03:21:57,461 ----------------------------------------------------------------------------------------------------
2023-10-11 03:21:57,461 EPOCH 3 done: loss 0.3101 - lr: 0.000117
2023-10-11 03:22:03,017 DEV : loss 0.2035694271326065 - f1-score (micro avg) 0.4298
2023-10-11 03:22:03,027 saving best model
2023-10-11 03:22:04,096 ----------------------------------------------------------------------------------------------------
2023-10-11 03:22:14,038 epoch 4 - iter 29/292 - loss 0.20110119 - time (sec): 9.94 - samples/sec: 475.37 - lr: 0.000115 - momentum: 0.000000
2023-10-11 03:22:24,401 epoch 4 - iter 58/292 - loss 0.17826557 - time (sec): 20.30 - samples/sec: 495.26 - lr: 0.000113 - momentum: 0.000000
2023-10-11 03:22:34,413 epoch 4 - iter 87/292 - loss 0.20996155 - time (sec): 30.31 - samples/sec: 484.16 - lr: 0.000112 - momentum: 0.000000
2023-10-11 03:22:44,346 epoch 4 - iter 116/292 - loss 0.21692770 - time (sec): 40.25 - samples/sec: 482.16 - lr: 0.000110 - momentum: 0.000000
2023-10-11 03:22:53,827 epoch 4 - iter 145/292 - loss 0.21814470 - time (sec): 49.73 - samples/sec: 479.47 - lr: 0.000108 - momentum: 0.000000
2023-10-11 03:23:03,233 epoch 4 - iter 174/292 - loss 0.21089514 - time (sec): 59.14 - samples/sec: 474.44 - lr: 0.000107 - momentum: 0.000000
2023-10-11 03:23:12,483 epoch 4 - iter 203/292 - loss 0.21084487 - time (sec): 68.38 - samples/sec: 467.12 - lr: 0.000105 - momentum: 0.000000
2023-10-11 03:23:22,118 epoch 4 - iter 232/292 - loss 0.20482475 - time (sec): 78.02 - samples/sec: 463.58 - lr: 0.000104 - momentum: 0.000000
2023-10-11 03:23:30,759 epoch 4 - iter 261/292 - loss 0.20151559 - time (sec): 86.66 - samples/sec: 456.03 - lr: 0.000102 - momentum: 0.000000
2023-10-11 03:23:40,714 epoch 4 - iter 290/292 - loss 0.20188943 - time (sec): 96.62 - samples/sec: 458.75 - lr: 0.000100 - momentum: 0.000000
2023-10-11 03:23:41,130 ----------------------------------------------------------------------------------------------------
2023-10-11 03:23:41,131 EPOCH 4 done: loss 0.2023 - lr: 0.000100
2023-10-11 03:23:46,947 DEV : loss 0.15248891711235046 - f1-score (micro avg) 0.6021
2023-10-11 03:23:46,957 saving best model
2023-10-11 03:23:49,630 ----------------------------------------------------------------------------------------------------
2023-10-11 03:23:59,366 epoch 5 - iter 29/292 - loss 0.15498313 - time (sec): 9.73 - samples/sec: 459.85 - lr: 0.000098 - momentum: 0.000000
2023-10-11 03:24:09,829 epoch 5 - iter 58/292 - loss 0.13333345 - time (sec): 20.19 - samples/sec: 473.04 - lr: 0.000097 - momentum: 0.000000
2023-10-11 03:24:20,042 epoch 5 - iter 87/292 - loss 0.14821866 - time (sec): 30.41 - samples/sec: 470.84 - lr: 0.000095 - momentum: 0.000000
2023-10-11 03:24:30,243 epoch 5 - iter 116/292 - loss 0.14730098 - time (sec): 40.61 - samples/sec: 470.32 - lr: 0.000093 - momentum: 0.000000
2023-10-11 03:24:40,156 epoch 5 - iter 145/292 - loss 0.15044218 - time (sec): 50.52 - samples/sec: 466.81 - lr: 0.000092 - momentum: 0.000000
2023-10-11 03:24:50,125 epoch 5 - iter 174/292 - loss 0.14812398 - time (sec): 60.49 - samples/sec: 460.27 - lr: 0.000090 - momentum: 0.000000
2023-10-11 03:24:59,576 epoch 5 - iter 203/292 - loss 0.14525037 - time (sec): 69.94 - samples/sec: 454.03 - lr: 0.000089 - momentum: 0.000000
2023-10-11 03:25:09,036 epoch 5 - iter 232/292 - loss 0.14314177 - time (sec): 79.40 - samples/sec: 446.58 - lr: 0.000087 - momentum: 0.000000
2023-10-11 03:25:19,457 epoch 5 - iter 261/292 - loss 0.13766737 - time (sec): 89.82 - samples/sec: 448.05 - lr: 0.000085 - momentum: 0.000000
2023-10-11 03:25:29,102 epoch 5 - iter 290/292 - loss 0.13520449 - time (sec): 99.47 - samples/sec: 445.86 - lr: 0.000084 - momentum: 0.000000
2023-10-11 03:25:29,548 ----------------------------------------------------------------------------------------------------
2023-10-11 03:25:29,549 EPOCH 5 done: loss 0.1354 - lr: 0.000084
2023-10-11 03:25:35,482 DEV : loss 0.12166187167167664 - f1-score (micro avg) 0.7455
2023-10-11 03:25:35,492 saving best model
2023-10-11 03:25:38,127 ----------------------------------------------------------------------------------------------------
2023-10-11 03:25:47,622 epoch 6 - iter 29/292 - loss 0.08960265 - time (sec): 9.49 - samples/sec: 393.43 - lr: 0.000082 - momentum: 0.000000
2023-10-11 03:25:57,785 epoch 6 - iter 58/292 - loss 0.12767978 - time (sec): 19.65 - samples/sec: 396.46 - lr: 0.000080 - momentum: 0.000000
2023-10-11 03:26:08,293 epoch 6 - iter 87/292 - loss 0.11619278 - time (sec): 30.16 - samples/sec: 412.04 - lr: 0.000078 - momentum: 0.000000
2023-10-11 03:26:18,684 epoch 6 - iter 116/292 - loss 0.11102256 - time (sec): 40.55 - samples/sec: 410.65 - lr: 0.000077 - momentum: 0.000000
2023-10-11 03:26:29,492 epoch 6 - iter 145/292 - loss 0.10620087 - time (sec): 51.36 - samples/sec: 420.53 - lr: 0.000075 - momentum: 0.000000
2023-10-11 03:26:40,798 epoch 6 - iter 174/292 - loss 0.09868087 - time (sec): 62.67 - samples/sec: 424.26 - lr: 0.000074 - momentum: 0.000000
2023-10-11 03:26:50,496 epoch 6 - iter 203/292 - loss 0.10008914 - time (sec): 72.36 - samples/sec: 420.45 - lr: 0.000072 - momentum: 0.000000
2023-10-11 03:27:00,772 epoch 6 - iter 232/292 - loss 0.09776359 - time (sec): 82.64 - samples/sec: 412.02 - lr: 0.000070 - momentum: 0.000000
2023-10-11 03:27:12,362 epoch 6 - iter 261/292 - loss 0.09830523 - time (sec): 94.23 - samples/sec: 418.52 - lr: 0.000069 - momentum: 0.000000
2023-10-11 03:27:22,801 epoch 6 - iter 290/292 - loss 0.09807018 - time (sec): 104.67 - samples/sec: 421.15 - lr: 0.000067 - momentum: 0.000000
2023-10-11 03:27:23,552 ----------------------------------------------------------------------------------------------------
2023-10-11 03:27:23,553 EPOCH 6 done: loss 0.0974 - lr: 0.000067
2023-10-11 03:27:29,960 DEV : loss 0.11362636834383011 - f1-score (micro avg) 0.7778
2023-10-11 03:27:29,970 saving best model
2023-10-11 03:27:32,668 ----------------------------------------------------------------------------------------------------
2023-10-11 03:27:42,352 epoch 7 - iter 29/292 - loss 0.07419722 - time (sec): 9.67 - samples/sec: 394.50 - lr: 0.000065 - momentum: 0.000000
2023-10-11 03:27:52,002 epoch 7 - iter 58/292 - loss 0.07402345 - time (sec): 19.32 - samples/sec: 410.05 - lr: 0.000063 - momentum: 0.000000
2023-10-11 03:28:01,047 epoch 7 - iter 87/292 - loss 0.07690943 - time (sec): 28.37 - samples/sec: 401.83 - lr: 0.000062 - momentum: 0.000000
2023-10-11 03:28:12,534 epoch 7 - iter 116/292 - loss 0.07959332 - time (sec): 39.85 - samples/sec: 427.33 - lr: 0.000060 - momentum: 0.000000
2023-10-11 03:28:22,410 epoch 7 - iter 145/292 - loss 0.07518449 - time (sec): 49.73 - samples/sec: 424.22 - lr: 0.000059 - momentum: 0.000000
2023-10-11 03:28:32,859 epoch 7 - iter 174/292 - loss 0.07895677 - time (sec): 60.18 - samples/sec: 419.98 - lr: 0.000057 - momentum: 0.000000
2023-10-11 03:28:42,965 epoch 7 - iter 203/292 - loss 0.07832199 - time (sec): 70.29 - samples/sec: 422.31 - lr: 0.000055 - momentum: 0.000000
2023-10-11 03:28:53,183 epoch 7 - iter 232/292 - loss 0.07419731 - time (sec): 80.50 - samples/sec: 428.64 - lr: 0.000054 - momentum: 0.000000
2023-10-11 03:29:03,235 epoch 7 - iter 261/292 - loss 0.07485255 - time (sec): 90.56 - samples/sec: 433.91 - lr: 0.000052 - momentum: 0.000000
2023-10-11 03:29:13,519 epoch 7 - iter 290/292 - loss 0.07487242 - time (sec): 100.84 - samples/sec: 438.77 - lr: 0.000050 - momentum: 0.000000
2023-10-11 03:29:14,009 ----------------------------------------------------------------------------------------------------
2023-10-11 03:29:14,010 EPOCH 7 done: loss 0.0749 - lr: 0.000050
2023-10-11 03:29:19,721 DEV : loss 0.1100909486413002 - f1-score (micro avg) 0.7894
2023-10-11 03:29:19,730 saving best model
2023-10-11 03:29:22,380 ----------------------------------------------------------------------------------------------------
2023-10-11 03:29:32,069 epoch 8 - iter 29/292 - loss 0.06858685 - time (sec): 9.68 - samples/sec: 423.14 - lr: 0.000048 - momentum: 0.000000
2023-10-11 03:29:41,880 epoch 8 - iter 58/292 - loss 0.06385058 - time (sec): 19.50 - samples/sec: 433.12 - lr: 0.000047 - momentum: 0.000000
2023-10-11 03:29:51,908 epoch 8 - iter 87/292 - loss 0.06960465 - time (sec): 29.52 - samples/sec: 433.10 - lr: 0.000045 - momentum: 0.000000
2023-10-11 03:30:03,134 epoch 8 - iter 116/292 - loss 0.07164164 - time (sec): 40.75 - samples/sec: 437.11 - lr: 0.000044 - momentum: 0.000000
2023-10-11 03:30:13,585 epoch 8 - iter 145/292 - loss 0.07036557 - time (sec): 51.20 - samples/sec: 432.87 - lr: 0.000042 - momentum: 0.000000
2023-10-11 03:30:22,910 epoch 8 - iter 174/292 - loss 0.06603532 - time (sec): 60.53 - samples/sec: 427.80 - lr: 0.000040 - momentum: 0.000000
2023-10-11 03:30:33,264 epoch 8 - iter 203/292 - loss 0.06513753 - time (sec): 70.88 - samples/sec: 433.78 - lr: 0.000039 - momentum: 0.000000
2023-10-11 03:30:43,326 epoch 8 - iter 232/292 - loss 0.06712616 - time (sec): 80.94 - samples/sec: 436.36 - lr: 0.000037 - momentum: 0.000000
2023-10-11 03:30:52,519 epoch 8 - iter 261/292 - loss 0.06210932 - time (sec): 90.13 - samples/sec: 434.89 - lr: 0.000035 - momentum: 0.000000
2023-10-11 03:31:02,720 epoch 8 - iter 290/292 - loss 0.05877080 - time (sec): 100.34 - samples/sec: 440.52 - lr: 0.000034 - momentum: 0.000000
2023-10-11 03:31:03,298 ----------------------------------------------------------------------------------------------------
2023-10-11 03:31:03,298 EPOCH 8 done: loss 0.0588 - lr: 0.000034
2023-10-11 03:31:08,969 DEV : loss 0.11329486221075058 - f1-score (micro avg) 0.7913
2023-10-11 03:31:08,978 saving best model
2023-10-11 03:31:11,608 ----------------------------------------------------------------------------------------------------
2023-10-11 03:31:21,665 epoch 9 - iter 29/292 - loss 0.04382080 - time (sec): 10.05 - samples/sec: 457.28 - lr: 0.000032 - momentum: 0.000000
2023-10-11 03:31:30,877 epoch 9 - iter 58/292 - loss 0.05356168 - time (sec): 19.26 - samples/sec: 427.78 - lr: 0.000030 - momentum: 0.000000
2023-10-11 03:31:40,507 epoch 9 - iter 87/292 - loss 0.05086497 - time (sec): 28.89 - samples/sec: 429.42 - lr: 0.000029 - momentum: 0.000000
2023-10-11 03:31:50,143 epoch 9 - iter 116/292 - loss 0.05195064 - time (sec): 38.53 - samples/sec: 426.77 - lr: 0.000027 - momentum: 0.000000
2023-10-11 03:32:00,316 epoch 9 - iter 145/292 - loss 0.05449964 - time (sec): 48.70 - samples/sec: 441.65 - lr: 0.000025 - momentum: 0.000000
2023-10-11 03:32:09,886 epoch 9 - iter 174/292 - loss 0.05152749 - time (sec): 58.27 - samples/sec: 448.59 - lr: 0.000024 - momentum: 0.000000
2023-10-11 03:32:19,473 epoch 9 - iter 203/292 - loss 0.04877327 - time (sec): 67.86 - samples/sec: 452.81 - lr: 0.000022 - momentum: 0.000000
2023-10-11 03:32:29,586 epoch 9 - iter 232/292 - loss 0.04997456 - time (sec): 77.97 - samples/sec: 459.62 - lr: 0.000020 - momentum: 0.000000
2023-10-11 03:32:38,528 epoch 9 - iter 261/292 - loss 0.04881578 - time (sec): 86.92 - samples/sec: 459.93 - lr: 0.000019 - momentum: 0.000000
2023-10-11 03:32:47,820 epoch 9 - iter 290/292 - loss 0.04899412 - time (sec): 96.21 - samples/sec: 460.02 - lr: 0.000017 - momentum: 0.000000
2023-10-11 03:32:48,264 ----------------------------------------------------------------------------------------------------
2023-10-11 03:32:48,264 EPOCH 9 done: loss 0.0489 - lr: 0.000017
2023-10-11 03:32:53,792 DEV : loss 0.11269763857126236 - f1-score (micro avg) 0.7904
2023-10-11 03:32:53,801 ----------------------------------------------------------------------------------------------------
2023-10-11 03:33:02,891 epoch 10 - iter 29/292 - loss 0.04691422 - time (sec): 9.09 - samples/sec: 463.71 - lr: 0.000015 - momentum: 0.000000
2023-10-11 03:33:12,045 epoch 10 - iter 58/292 - loss 0.05143091 - time (sec): 18.24 - samples/sec: 457.69 - lr: 0.000014 - momentum: 0.000000
2023-10-11 03:33:22,171 epoch 10 - iter 87/292 - loss 0.04523086 - time (sec): 28.37 - samples/sec: 465.70 - lr: 0.000012 - momentum: 0.000000
2023-10-11 03:33:31,686 epoch 10 - iter 116/292 - loss 0.04088964 - time (sec): 37.88 - samples/sec: 465.65 - lr: 0.000010 - momentum: 0.000000
2023-10-11 03:33:41,168 epoch 10 - iter 145/292 - loss 0.04323243 - time (sec): 47.36 - samples/sec: 465.18 - lr: 0.000009 - momentum: 0.000000
2023-10-11 03:33:50,228 epoch 10 - iter 174/292 - loss 0.04318784 - time (sec): 56.43 - samples/sec: 457.83 - lr: 0.000007 - momentum: 0.000000
2023-10-11 03:34:00,927 epoch 10 - iter 203/292 - loss 0.04500761 - time (sec): 67.12 - samples/sec: 461.65 - lr: 0.000005 - momentum: 0.000000
2023-10-11 03:34:10,593 epoch 10 - iter 232/292 - loss 0.04463928 - time (sec): 76.79 - samples/sec: 458.99 - lr: 0.000004 - momentum: 0.000000
2023-10-11 03:34:20,237 epoch 10 - iter 261/292 - loss 0.04516349 - time (sec): 86.43 - samples/sec: 460.74 - lr: 0.000002 - momentum: 0.000000
2023-10-11 03:34:29,953 epoch 10 - iter 290/292 - loss 0.04578337 - time (sec): 96.15 - samples/sec: 457.82 - lr: 0.000000 - momentum: 0.000000
2023-10-11 03:34:30,663 ----------------------------------------------------------------------------------------------------
2023-10-11 03:34:30,664 EPOCH 10 done: loss 0.0454 - lr: 0.000000
2023-10-11 03:34:36,393 DEV : loss 0.11296340078115463 - f1-score (micro avg) 0.7913
2023-10-11 03:34:37,317 ----------------------------------------------------------------------------------------------------
2023-10-11 03:34:37,318 Loading model from best epoch ...
2023-10-11 03:34:41,397 SequenceTagger predicts: Dictionary with 17 tags: O, S-LOC, B-LOC, E-LOC, I-LOC, S-PER, B-PER, E-PER, I-PER, S-ORG, B-ORG, E-ORG, I-ORG, S-HumanProd, B-HumanProd, E-HumanProd, I-HumanProd
2023-10-11 03:34:54,193
Results:
- F-score (micro) 0.7433
- F-score (macro) 0.6945
- Accuracy 0.6094
By class:
precision recall f1-score support
PER 0.8065 0.8621 0.8333 348
LOC 0.6047 0.7854 0.6833 261
ORG 0.4255 0.3846 0.4040 52
HumanProd 0.9000 0.8182 0.8571 22
micro avg 0.6979 0.7950 0.7433 683
macro avg 0.6842 0.7126 0.6945 683
weighted avg 0.7034 0.7950 0.7441 683
2023-10-11 03:34:54,193 ----------------------------------------------------------------------------------------------------
|