File size: 25,292 Bytes
346a01f |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 |
2023-10-11 09:57:56,378 ----------------------------------------------------------------------------------------------------
2023-10-11 09:57:56,380 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 09:57:56,380 ----------------------------------------------------------------------------------------------------
2023-10-11 09:57:56,381 MultiCorpus: 1085 train + 148 dev + 364 test sentences
- NER_HIPE_2022 Corpus: 1085 train + 148 dev + 364 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/newseye/sv/with_doc_seperator
2023-10-11 09:57:56,381 ----------------------------------------------------------------------------------------------------
2023-10-11 09:57:56,381 Train: 1085 sentences
2023-10-11 09:57:56,381 (train_with_dev=False, train_with_test=False)
2023-10-11 09:57:56,381 ----------------------------------------------------------------------------------------------------
2023-10-11 09:57:56,381 Training Params:
2023-10-11 09:57:56,381 - learning_rate: "0.00016"
2023-10-11 09:57:56,381 - mini_batch_size: "4"
2023-10-11 09:57:56,381 - max_epochs: "10"
2023-10-11 09:57:56,381 - shuffle: "True"
2023-10-11 09:57:56,381 ----------------------------------------------------------------------------------------------------
2023-10-11 09:57:56,381 Plugins:
2023-10-11 09:57:56,381 - TensorboardLogger
2023-10-11 09:57:56,382 - LinearScheduler | warmup_fraction: '0.1'
2023-10-11 09:57:56,382 ----------------------------------------------------------------------------------------------------
2023-10-11 09:57:56,382 Final evaluation on model from best epoch (best-model.pt)
2023-10-11 09:57:56,382 - metric: "('micro avg', 'f1-score')"
2023-10-11 09:57:56,382 ----------------------------------------------------------------------------------------------------
2023-10-11 09:57:56,382 Computation:
2023-10-11 09:57:56,382 - compute on device: cuda:0
2023-10-11 09:57:56,382 - embedding storage: none
2023-10-11 09:57:56,382 ----------------------------------------------------------------------------------------------------
2023-10-11 09:57:56,382 Model training base path: "hmbench-newseye/sv-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs4-wsFalse-e10-lr0.00016-poolingfirst-layers-1-crfFalse-2"
2023-10-11 09:57:56,382 ----------------------------------------------------------------------------------------------------
2023-10-11 09:57:56,382 ----------------------------------------------------------------------------------------------------
2023-10-11 09:57:56,382 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-11 09:58:05,782 epoch 1 - iter 27/272 - loss 2.84961659 - time (sec): 9.40 - samples/sec: 576.00 - lr: 0.000015 - momentum: 0.000000
2023-10-11 09:58:14,469 epoch 1 - iter 54/272 - loss 2.83915292 - time (sec): 18.08 - samples/sec: 545.61 - lr: 0.000031 - momentum: 0.000000
2023-10-11 09:58:23,736 epoch 1 - iter 81/272 - loss 2.81803749 - time (sec): 27.35 - samples/sec: 553.45 - lr: 0.000047 - momentum: 0.000000
2023-10-11 09:58:33,589 epoch 1 - iter 108/272 - loss 2.74804522 - time (sec): 37.20 - samples/sec: 564.68 - lr: 0.000063 - momentum: 0.000000
2023-10-11 09:58:42,886 epoch 1 - iter 135/272 - loss 2.65294364 - time (sec): 46.50 - samples/sec: 566.48 - lr: 0.000079 - momentum: 0.000000
2023-10-11 09:58:51,421 epoch 1 - iter 162/272 - loss 2.56578574 - time (sec): 55.04 - samples/sec: 556.46 - lr: 0.000095 - momentum: 0.000000
2023-10-11 09:59:00,707 epoch 1 - iter 189/272 - loss 2.45465098 - time (sec): 64.32 - samples/sec: 554.14 - lr: 0.000111 - momentum: 0.000000
2023-10-11 09:59:09,883 epoch 1 - iter 216/272 - loss 2.33901060 - time (sec): 73.50 - samples/sec: 554.24 - lr: 0.000126 - momentum: 0.000000
2023-10-11 09:59:19,912 epoch 1 - iter 243/272 - loss 2.19029882 - time (sec): 83.53 - samples/sec: 558.64 - lr: 0.000142 - momentum: 0.000000
2023-10-11 09:59:29,206 epoch 1 - iter 270/272 - loss 2.07205638 - time (sec): 92.82 - samples/sec: 559.30 - lr: 0.000158 - momentum: 0.000000
2023-10-11 09:59:29,534 ----------------------------------------------------------------------------------------------------
2023-10-11 09:59:29,535 EPOCH 1 done: loss 2.0705 - lr: 0.000158
2023-10-11 09:59:34,650 DEV : loss 0.7345565557479858 - f1-score (micro avg) 0.0
2023-10-11 09:59:34,658 ----------------------------------------------------------------------------------------------------
2023-10-11 09:59:44,295 epoch 2 - iter 27/272 - loss 0.70414994 - time (sec): 9.63 - samples/sec: 601.26 - lr: 0.000158 - momentum: 0.000000
2023-10-11 09:59:53,165 epoch 2 - iter 54/272 - loss 0.63540563 - time (sec): 18.50 - samples/sec: 581.59 - lr: 0.000157 - momentum: 0.000000
2023-10-11 10:00:02,825 epoch 2 - iter 81/272 - loss 0.63286981 - time (sec): 28.16 - samples/sec: 590.61 - lr: 0.000155 - momentum: 0.000000
2023-10-11 10:00:12,024 epoch 2 - iter 108/272 - loss 0.60068837 - time (sec): 37.36 - samples/sec: 583.88 - lr: 0.000153 - momentum: 0.000000
2023-10-11 10:00:21,275 epoch 2 - iter 135/272 - loss 0.58169819 - time (sec): 46.61 - samples/sec: 578.50 - lr: 0.000151 - momentum: 0.000000
2023-10-11 10:00:29,801 epoch 2 - iter 162/272 - loss 0.55542542 - time (sec): 55.14 - samples/sec: 569.54 - lr: 0.000149 - momentum: 0.000000
2023-10-11 10:00:38,726 epoch 2 - iter 189/272 - loss 0.54212044 - time (sec): 64.07 - samples/sec: 563.12 - lr: 0.000148 - momentum: 0.000000
2023-10-11 10:00:48,281 epoch 2 - iter 216/272 - loss 0.51858578 - time (sec): 73.62 - samples/sec: 561.94 - lr: 0.000146 - momentum: 0.000000
2023-10-11 10:00:57,669 epoch 2 - iter 243/272 - loss 0.50379454 - time (sec): 83.01 - samples/sec: 557.51 - lr: 0.000144 - momentum: 0.000000
2023-10-11 10:01:08,182 epoch 2 - iter 270/272 - loss 0.49383828 - time (sec): 93.52 - samples/sec: 554.15 - lr: 0.000142 - momentum: 0.000000
2023-10-11 10:01:08,590 ----------------------------------------------------------------------------------------------------
2023-10-11 10:01:08,590 EPOCH 2 done: loss 0.4930 - lr: 0.000142
2023-10-11 10:01:14,578 DEV : loss 0.3049907982349396 - f1-score (micro avg) 0.2867
2023-10-11 10:01:14,589 saving best model
2023-10-11 10:01:15,564 ----------------------------------------------------------------------------------------------------
2023-10-11 10:01:26,175 epoch 3 - iter 27/272 - loss 0.36825189 - time (sec): 10.61 - samples/sec: 495.68 - lr: 0.000141 - momentum: 0.000000
2023-10-11 10:01:36,526 epoch 3 - iter 54/272 - loss 0.35828698 - time (sec): 20.96 - samples/sec: 489.94 - lr: 0.000139 - momentum: 0.000000
2023-10-11 10:01:46,586 epoch 3 - iter 81/272 - loss 0.33371894 - time (sec): 31.02 - samples/sec: 489.93 - lr: 0.000137 - momentum: 0.000000
2023-10-11 10:01:56,178 epoch 3 - iter 108/272 - loss 0.32842770 - time (sec): 40.61 - samples/sec: 500.40 - lr: 0.000135 - momentum: 0.000000
2023-10-11 10:02:06,182 epoch 3 - iter 135/272 - loss 0.32494522 - time (sec): 50.62 - samples/sec: 513.93 - lr: 0.000133 - momentum: 0.000000
2023-10-11 10:02:15,811 epoch 3 - iter 162/272 - loss 0.31495256 - time (sec): 60.24 - samples/sec: 514.72 - lr: 0.000132 - momentum: 0.000000
2023-10-11 10:02:26,550 epoch 3 - iter 189/272 - loss 0.31307865 - time (sec): 70.98 - samples/sec: 525.09 - lr: 0.000130 - momentum: 0.000000
2023-10-11 10:02:36,602 epoch 3 - iter 216/272 - loss 0.30261968 - time (sec): 81.04 - samples/sec: 527.34 - lr: 0.000128 - momentum: 0.000000
2023-10-11 10:02:45,602 epoch 3 - iter 243/272 - loss 0.30154536 - time (sec): 90.04 - samples/sec: 520.58 - lr: 0.000126 - momentum: 0.000000
2023-10-11 10:02:55,043 epoch 3 - iter 270/272 - loss 0.29990213 - time (sec): 99.48 - samples/sec: 519.97 - lr: 0.000125 - momentum: 0.000000
2023-10-11 10:02:55,531 ----------------------------------------------------------------------------------------------------
2023-10-11 10:02:55,531 EPOCH 3 done: loss 0.3001 - lr: 0.000125
2023-10-11 10:03:01,411 DEV : loss 0.23118416965007782 - f1-score (micro avg) 0.4514
2023-10-11 10:03:01,419 saving best model
2023-10-11 10:03:03,968 ----------------------------------------------------------------------------------------------------
2023-10-11 10:03:13,204 epoch 4 - iter 27/272 - loss 0.26389527 - time (sec): 9.23 - samples/sec: 543.60 - lr: 0.000123 - momentum: 0.000000
2023-10-11 10:03:22,197 epoch 4 - iter 54/272 - loss 0.23946997 - time (sec): 18.22 - samples/sec: 524.90 - lr: 0.000121 - momentum: 0.000000
2023-10-11 10:03:32,348 epoch 4 - iter 81/272 - loss 0.22494046 - time (sec): 28.37 - samples/sec: 549.85 - lr: 0.000119 - momentum: 0.000000
2023-10-11 10:03:42,207 epoch 4 - iter 108/272 - loss 0.22128238 - time (sec): 38.23 - samples/sec: 551.86 - lr: 0.000117 - momentum: 0.000000
2023-10-11 10:03:51,557 epoch 4 - iter 135/272 - loss 0.21790054 - time (sec): 47.58 - samples/sec: 546.55 - lr: 0.000116 - momentum: 0.000000
2023-10-11 10:04:01,550 epoch 4 - iter 162/272 - loss 0.21355483 - time (sec): 57.58 - samples/sec: 549.19 - lr: 0.000114 - momentum: 0.000000
2023-10-11 10:04:10,733 epoch 4 - iter 189/272 - loss 0.21461498 - time (sec): 66.76 - samples/sec: 543.62 - lr: 0.000112 - momentum: 0.000000
2023-10-11 10:04:20,210 epoch 4 - iter 216/272 - loss 0.21161711 - time (sec): 76.24 - samples/sec: 543.55 - lr: 0.000110 - momentum: 0.000000
2023-10-11 10:04:29,699 epoch 4 - iter 243/272 - loss 0.21439789 - time (sec): 85.72 - samples/sec: 544.83 - lr: 0.000109 - momentum: 0.000000
2023-10-11 10:04:38,989 epoch 4 - iter 270/272 - loss 0.21064388 - time (sec): 95.01 - samples/sec: 545.07 - lr: 0.000107 - momentum: 0.000000
2023-10-11 10:04:39,417 ----------------------------------------------------------------------------------------------------
2023-10-11 10:04:39,418 EPOCH 4 done: loss 0.2106 - lr: 0.000107
2023-10-11 10:04:45,299 DEV : loss 0.17590400576591492 - f1-score (micro avg) 0.5839
2023-10-11 10:04:45,309 saving best model
2023-10-11 10:04:47,847 ----------------------------------------------------------------------------------------------------
2023-10-11 10:04:56,859 epoch 5 - iter 27/272 - loss 0.16095320 - time (sec): 9.01 - samples/sec: 518.32 - lr: 0.000105 - momentum: 0.000000
2023-10-11 10:05:06,266 epoch 5 - iter 54/272 - loss 0.15303063 - time (sec): 18.41 - samples/sec: 545.17 - lr: 0.000103 - momentum: 0.000000
2023-10-11 10:05:15,581 epoch 5 - iter 81/272 - loss 0.14715890 - time (sec): 27.73 - samples/sec: 553.13 - lr: 0.000101 - momentum: 0.000000
2023-10-11 10:05:24,566 epoch 5 - iter 108/272 - loss 0.14859354 - time (sec): 36.71 - samples/sec: 551.36 - lr: 0.000100 - momentum: 0.000000
2023-10-11 10:05:33,383 epoch 5 - iter 135/272 - loss 0.13938521 - time (sec): 45.53 - samples/sec: 549.22 - lr: 0.000098 - momentum: 0.000000
2023-10-11 10:05:43,089 epoch 5 - iter 162/272 - loss 0.13872093 - time (sec): 55.24 - samples/sec: 559.36 - lr: 0.000096 - momentum: 0.000000
2023-10-11 10:05:52,177 epoch 5 - iter 189/272 - loss 0.14314448 - time (sec): 64.33 - samples/sec: 558.08 - lr: 0.000094 - momentum: 0.000000
2023-10-11 10:06:01,633 epoch 5 - iter 216/272 - loss 0.14519265 - time (sec): 73.78 - samples/sec: 559.16 - lr: 0.000093 - momentum: 0.000000
2023-10-11 10:06:11,241 epoch 5 - iter 243/272 - loss 0.14574570 - time (sec): 83.39 - samples/sec: 559.96 - lr: 0.000091 - momentum: 0.000000
2023-10-11 10:06:20,699 epoch 5 - iter 270/272 - loss 0.14307060 - time (sec): 92.85 - samples/sec: 556.93 - lr: 0.000089 - momentum: 0.000000
2023-10-11 10:06:21,219 ----------------------------------------------------------------------------------------------------
2023-10-11 10:06:21,219 EPOCH 5 done: loss 0.1430 - lr: 0.000089
2023-10-11 10:06:26,814 DEV : loss 0.15896683931350708 - f1-score (micro avg) 0.6123
2023-10-11 10:06:26,824 saving best model
2023-10-11 10:06:29,363 ----------------------------------------------------------------------------------------------------
2023-10-11 10:06:39,090 epoch 6 - iter 27/272 - loss 0.12861859 - time (sec): 9.72 - samples/sec: 563.75 - lr: 0.000087 - momentum: 0.000000
2023-10-11 10:06:48,027 epoch 6 - iter 54/272 - loss 0.12625714 - time (sec): 18.66 - samples/sec: 546.90 - lr: 0.000085 - momentum: 0.000000
2023-10-11 10:06:57,647 epoch 6 - iter 81/272 - loss 0.11812378 - time (sec): 28.28 - samples/sec: 551.60 - lr: 0.000084 - momentum: 0.000000
2023-10-11 10:07:07,496 epoch 6 - iter 108/272 - loss 0.11590135 - time (sec): 38.13 - samples/sec: 563.80 - lr: 0.000082 - momentum: 0.000000
2023-10-11 10:07:16,590 epoch 6 - iter 135/272 - loss 0.11257005 - time (sec): 47.22 - samples/sec: 548.24 - lr: 0.000080 - momentum: 0.000000
2023-10-11 10:07:26,339 epoch 6 - iter 162/272 - loss 0.10607560 - time (sec): 56.97 - samples/sec: 553.59 - lr: 0.000078 - momentum: 0.000000
2023-10-11 10:07:35,893 epoch 6 - iter 189/272 - loss 0.10194287 - time (sec): 66.53 - samples/sec: 550.51 - lr: 0.000077 - momentum: 0.000000
2023-10-11 10:07:45,318 epoch 6 - iter 216/272 - loss 0.10614524 - time (sec): 75.95 - samples/sec: 547.98 - lr: 0.000075 - momentum: 0.000000
2023-10-11 10:07:54,921 epoch 6 - iter 243/272 - loss 0.10334886 - time (sec): 85.55 - samples/sec: 547.68 - lr: 0.000073 - momentum: 0.000000
2023-10-11 10:08:04,092 epoch 6 - iter 270/272 - loss 0.10347814 - time (sec): 94.72 - samples/sec: 546.02 - lr: 0.000071 - momentum: 0.000000
2023-10-11 10:08:04,598 ----------------------------------------------------------------------------------------------------
2023-10-11 10:08:04,599 EPOCH 6 done: loss 0.1033 - lr: 0.000071
2023-10-11 10:08:10,290 DEV : loss 0.1453726589679718 - f1-score (micro avg) 0.6964
2023-10-11 10:08:10,299 saving best model
2023-10-11 10:08:12,837 ----------------------------------------------------------------------------------------------------
2023-10-11 10:08:22,005 epoch 7 - iter 27/272 - loss 0.08200013 - time (sec): 9.16 - samples/sec: 526.74 - lr: 0.000069 - momentum: 0.000000
2023-10-11 10:08:31,942 epoch 7 - iter 54/272 - loss 0.07169928 - time (sec): 19.10 - samples/sec: 549.62 - lr: 0.000068 - momentum: 0.000000
2023-10-11 10:08:41,790 epoch 7 - iter 81/272 - loss 0.06750559 - time (sec): 28.95 - samples/sec: 552.18 - lr: 0.000066 - momentum: 0.000000
2023-10-11 10:08:51,214 epoch 7 - iter 108/272 - loss 0.06876153 - time (sec): 38.37 - samples/sec: 549.77 - lr: 0.000064 - momentum: 0.000000
2023-10-11 10:09:01,123 epoch 7 - iter 135/272 - loss 0.07382666 - time (sec): 48.28 - samples/sec: 549.32 - lr: 0.000062 - momentum: 0.000000
2023-10-11 10:09:10,687 epoch 7 - iter 162/272 - loss 0.07376310 - time (sec): 57.85 - samples/sec: 542.87 - lr: 0.000061 - momentum: 0.000000
2023-10-11 10:09:19,909 epoch 7 - iter 189/272 - loss 0.07298171 - time (sec): 67.07 - samples/sec: 533.52 - lr: 0.000059 - momentum: 0.000000
2023-10-11 10:09:30,231 epoch 7 - iter 216/272 - loss 0.07154073 - time (sec): 77.39 - samples/sec: 535.81 - lr: 0.000057 - momentum: 0.000000
2023-10-11 10:09:39,463 epoch 7 - iter 243/272 - loss 0.07659368 - time (sec): 86.62 - samples/sec: 535.04 - lr: 0.000055 - momentum: 0.000000
2023-10-11 10:09:49,375 epoch 7 - iter 270/272 - loss 0.07547255 - time (sec): 96.53 - samples/sec: 536.22 - lr: 0.000054 - momentum: 0.000000
2023-10-11 10:09:49,841 ----------------------------------------------------------------------------------------------------
2023-10-11 10:09:49,841 EPOCH 7 done: loss 0.0753 - lr: 0.000054
2023-10-11 10:09:55,608 DEV : loss 0.1472298800945282 - f1-score (micro avg) 0.7478
2023-10-11 10:09:55,617 saving best model
2023-10-11 10:09:58,161 ----------------------------------------------------------------------------------------------------
2023-10-11 10:10:07,364 epoch 8 - iter 27/272 - loss 0.07327268 - time (sec): 9.20 - samples/sec: 499.40 - lr: 0.000052 - momentum: 0.000000
2023-10-11 10:10:16,567 epoch 8 - iter 54/272 - loss 0.05912708 - time (sec): 18.40 - samples/sec: 510.50 - lr: 0.000050 - momentum: 0.000000
2023-10-11 10:10:26,716 epoch 8 - iter 81/272 - loss 0.06315843 - time (sec): 28.55 - samples/sec: 538.72 - lr: 0.000048 - momentum: 0.000000
2023-10-11 10:10:36,403 epoch 8 - iter 108/272 - loss 0.06440136 - time (sec): 38.24 - samples/sec: 540.82 - lr: 0.000046 - momentum: 0.000000
2023-10-11 10:10:45,979 epoch 8 - iter 135/272 - loss 0.06543166 - time (sec): 47.81 - samples/sec: 541.26 - lr: 0.000045 - momentum: 0.000000
2023-10-11 10:10:54,660 epoch 8 - iter 162/272 - loss 0.06830101 - time (sec): 56.49 - samples/sec: 531.16 - lr: 0.000043 - momentum: 0.000000
2023-10-11 10:11:04,624 epoch 8 - iter 189/272 - loss 0.06687904 - time (sec): 66.46 - samples/sec: 535.98 - lr: 0.000041 - momentum: 0.000000
2023-10-11 10:11:15,233 epoch 8 - iter 216/272 - loss 0.06417230 - time (sec): 77.07 - samples/sec: 545.82 - lr: 0.000039 - momentum: 0.000000
2023-10-11 10:11:24,573 epoch 8 - iter 243/272 - loss 0.06203206 - time (sec): 86.41 - samples/sec: 541.39 - lr: 0.000038 - momentum: 0.000000
2023-10-11 10:11:34,240 epoch 8 - iter 270/272 - loss 0.05987987 - time (sec): 96.07 - samples/sec: 539.08 - lr: 0.000036 - momentum: 0.000000
2023-10-11 10:11:34,685 ----------------------------------------------------------------------------------------------------
2023-10-11 10:11:34,686 EPOCH 8 done: loss 0.0599 - lr: 0.000036
2023-10-11 10:11:40,788 DEV : loss 0.1424403041601181 - f1-score (micro avg) 0.7653
2023-10-11 10:11:40,796 saving best model
2023-10-11 10:11:43,341 ----------------------------------------------------------------------------------------------------
2023-10-11 10:11:51,856 epoch 9 - iter 27/272 - loss 0.07666643 - time (sec): 8.51 - samples/sec: 504.34 - lr: 0.000034 - momentum: 0.000000
2023-10-11 10:12:01,071 epoch 9 - iter 54/272 - loss 0.07280220 - time (sec): 17.73 - samples/sec: 533.59 - lr: 0.000032 - momentum: 0.000000
2023-10-11 10:12:10,786 epoch 9 - iter 81/272 - loss 0.05972387 - time (sec): 27.44 - samples/sec: 527.72 - lr: 0.000030 - momentum: 0.000000
2023-10-11 10:12:20,137 epoch 9 - iter 108/272 - loss 0.06294634 - time (sec): 36.79 - samples/sec: 531.16 - lr: 0.000029 - momentum: 0.000000
2023-10-11 10:12:29,121 epoch 9 - iter 135/272 - loss 0.06234610 - time (sec): 45.78 - samples/sec: 522.90 - lr: 0.000027 - momentum: 0.000000
2023-10-11 10:12:39,294 epoch 9 - iter 162/272 - loss 0.05914625 - time (sec): 55.95 - samples/sec: 534.53 - lr: 0.000025 - momentum: 0.000000
2023-10-11 10:12:48,743 epoch 9 - iter 189/272 - loss 0.05774082 - time (sec): 65.40 - samples/sec: 535.18 - lr: 0.000023 - momentum: 0.000000
2023-10-11 10:12:58,443 epoch 9 - iter 216/272 - loss 0.05539923 - time (sec): 75.10 - samples/sec: 537.09 - lr: 0.000022 - momentum: 0.000000
2023-10-11 10:13:08,510 epoch 9 - iter 243/272 - loss 0.05258815 - time (sec): 85.16 - samples/sec: 543.20 - lr: 0.000020 - momentum: 0.000000
2023-10-11 10:13:18,255 epoch 9 - iter 270/272 - loss 0.05055787 - time (sec): 94.91 - samples/sec: 545.35 - lr: 0.000018 - momentum: 0.000000
2023-10-11 10:13:18,703 ----------------------------------------------------------------------------------------------------
2023-10-11 10:13:18,703 EPOCH 9 done: loss 0.0506 - lr: 0.000018
2023-10-11 10:13:24,318 DEV : loss 0.14669708907604218 - f1-score (micro avg) 0.7576
2023-10-11 10:13:24,327 ----------------------------------------------------------------------------------------------------
2023-10-11 10:13:34,114 epoch 10 - iter 27/272 - loss 0.05704288 - time (sec): 9.79 - samples/sec: 547.25 - lr: 0.000016 - momentum: 0.000000
2023-10-11 10:13:43,165 epoch 10 - iter 54/272 - loss 0.06282741 - time (sec): 18.84 - samples/sec: 532.27 - lr: 0.000014 - momentum: 0.000000
2023-10-11 10:13:52,252 epoch 10 - iter 81/272 - loss 0.05470713 - time (sec): 27.92 - samples/sec: 535.00 - lr: 0.000013 - momentum: 0.000000
2023-10-11 10:14:01,156 epoch 10 - iter 108/272 - loss 0.05095954 - time (sec): 36.83 - samples/sec: 534.87 - lr: 0.000011 - momentum: 0.000000
2023-10-11 10:14:11,443 epoch 10 - iter 135/272 - loss 0.04981593 - time (sec): 47.11 - samples/sec: 550.36 - lr: 0.000009 - momentum: 0.000000
2023-10-11 10:14:20,845 epoch 10 - iter 162/272 - loss 0.04680562 - time (sec): 56.52 - samples/sec: 545.26 - lr: 0.000007 - momentum: 0.000000
2023-10-11 10:14:30,321 epoch 10 - iter 189/272 - loss 0.04579138 - time (sec): 65.99 - samples/sec: 545.81 - lr: 0.000005 - momentum: 0.000000
2023-10-11 10:14:39,821 epoch 10 - iter 216/272 - loss 0.04474162 - time (sec): 75.49 - samples/sec: 548.68 - lr: 0.000004 - momentum: 0.000000
2023-10-11 10:14:49,718 epoch 10 - iter 243/272 - loss 0.04341634 - time (sec): 85.39 - samples/sec: 549.72 - lr: 0.000002 - momentum: 0.000000
2023-10-11 10:14:58,779 epoch 10 - iter 270/272 - loss 0.04483462 - time (sec): 94.45 - samples/sec: 547.28 - lr: 0.000000 - momentum: 0.000000
2023-10-11 10:14:59,278 ----------------------------------------------------------------------------------------------------
2023-10-11 10:14:59,278 EPOCH 10 done: loss 0.0447 - lr: 0.000000
2023-10-11 10:15:04,833 DEV : loss 0.14579297602176666 - f1-score (micro avg) 0.7607
2023-10-11 10:15:05,711 ----------------------------------------------------------------------------------------------------
2023-10-11 10:15:05,713 Loading model from best epoch ...
2023-10-11 10:15:09,559 SequenceTagger predicts: Dictionary with 17 tags: O, S-LOC, B-LOC, E-LOC, I-LOC, S-PER, B-PER, E-PER, I-PER, S-HumanProd, B-HumanProd, E-HumanProd, I-HumanProd, S-ORG, B-ORG, E-ORG, I-ORG
2023-10-11 10:15:22,453
Results:
- F-score (micro) 0.7359
- F-score (macro) 0.6371
- Accuracy 0.6043
By class:
precision recall f1-score support
LOC 0.7330 0.8622 0.7923 312
PER 0.6784 0.8317 0.7473 208
ORG 0.3415 0.2545 0.2917 55
HumanProd 0.6129 0.8636 0.7170 22
micro avg 0.6844 0.7956 0.7359 597
macro avg 0.5914 0.7030 0.6371 597
weighted avg 0.6735 0.7956 0.7277 597
2023-10-11 10:15:22,454 ----------------------------------------------------------------------------------------------------
|