File size: 25,293 Bytes
65c5835 |
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 10:50:03,874 ----------------------------------------------------------------------------------------------------
2023-10-11 10:50:03,877 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 10:50:03,877 ----------------------------------------------------------------------------------------------------
2023-10-11 10:50:03,877 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 10:50:03,877 ----------------------------------------------------------------------------------------------------
2023-10-11 10:50:03,877 Train: 1085 sentences
2023-10-11 10:50:03,877 (train_with_dev=False, train_with_test=False)
2023-10-11 10:50:03,877 ----------------------------------------------------------------------------------------------------
2023-10-11 10:50:03,877 Training Params:
2023-10-11 10:50:03,877 - learning_rate: "0.00015"
2023-10-11 10:50:03,878 - mini_batch_size: "4"
2023-10-11 10:50:03,878 - max_epochs: "10"
2023-10-11 10:50:03,878 - shuffle: "True"
2023-10-11 10:50:03,878 ----------------------------------------------------------------------------------------------------
2023-10-11 10:50:03,878 Plugins:
2023-10-11 10:50:03,878 - TensorboardLogger
2023-10-11 10:50:03,878 - LinearScheduler | warmup_fraction: '0.1'
2023-10-11 10:50:03,878 ----------------------------------------------------------------------------------------------------
2023-10-11 10:50:03,878 Final evaluation on model from best epoch (best-model.pt)
2023-10-11 10:50:03,878 - metric: "('micro avg', 'f1-score')"
2023-10-11 10:50:03,878 ----------------------------------------------------------------------------------------------------
2023-10-11 10:50:03,878 Computation:
2023-10-11 10:50:03,878 - compute on device: cuda:0
2023-10-11 10:50:03,878 - embedding storage: none
2023-10-11 10:50:03,878 ----------------------------------------------------------------------------------------------------
2023-10-11 10:50:03,878 Model training base path: "hmbench-newseye/sv-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs4-wsFalse-e10-lr0.00015-poolingfirst-layers-1-crfFalse-3"
2023-10-11 10:50:03,879 ----------------------------------------------------------------------------------------------------
2023-10-11 10:50:03,879 ----------------------------------------------------------------------------------------------------
2023-10-11 10:50:03,879 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-11 10:50:13,487 epoch 1 - iter 27/272 - loss 2.82659249 - time (sec): 9.61 - samples/sec: 598.94 - lr: 0.000014 - momentum: 0.000000
2023-10-11 10:50:22,796 epoch 1 - iter 54/272 - loss 2.81733222 - time (sec): 18.92 - samples/sec: 592.05 - lr: 0.000029 - momentum: 0.000000
2023-10-11 10:50:32,122 epoch 1 - iter 81/272 - loss 2.79616710 - time (sec): 28.24 - samples/sec: 580.96 - lr: 0.000044 - momentum: 0.000000
2023-10-11 10:50:41,422 epoch 1 - iter 108/272 - loss 2.75078140 - time (sec): 37.54 - samples/sec: 572.55 - lr: 0.000059 - momentum: 0.000000
2023-10-11 10:50:50,767 epoch 1 - iter 135/272 - loss 2.66645771 - time (sec): 46.89 - samples/sec: 571.66 - lr: 0.000074 - momentum: 0.000000
2023-10-11 10:50:59,522 epoch 1 - iter 162/272 - loss 2.58487111 - time (sec): 55.64 - samples/sec: 564.72 - lr: 0.000089 - momentum: 0.000000
2023-10-11 10:51:08,232 epoch 1 - iter 189/272 - loss 2.48987222 - time (sec): 64.35 - samples/sec: 557.49 - lr: 0.000104 - momentum: 0.000000
2023-10-11 10:51:17,966 epoch 1 - iter 216/272 - loss 2.36263029 - time (sec): 74.09 - samples/sec: 561.83 - lr: 0.000119 - momentum: 0.000000
2023-10-11 10:51:26,940 epoch 1 - iter 243/272 - loss 2.24603417 - time (sec): 83.06 - samples/sec: 560.39 - lr: 0.000133 - momentum: 0.000000
2023-10-11 10:51:36,410 epoch 1 - iter 270/272 - loss 2.12867006 - time (sec): 92.53 - samples/sec: 558.34 - lr: 0.000148 - momentum: 0.000000
2023-10-11 10:51:36,944 ----------------------------------------------------------------------------------------------------
2023-10-11 10:51:36,945 EPOCH 1 done: loss 2.1205 - lr: 0.000148
2023-10-11 10:51:41,788 DEV : loss 0.8181562423706055 - f1-score (micro avg) 0.0
2023-10-11 10:51:41,796 ----------------------------------------------------------------------------------------------------
2023-10-11 10:51:51,485 epoch 2 - iter 27/272 - loss 0.77588845 - time (sec): 9.69 - samples/sec: 602.16 - lr: 0.000148 - momentum: 0.000000
2023-10-11 10:52:00,658 epoch 2 - iter 54/272 - loss 0.73677202 - time (sec): 18.86 - samples/sec: 597.29 - lr: 0.000147 - momentum: 0.000000
2023-10-11 10:52:09,718 epoch 2 - iter 81/272 - loss 0.70960269 - time (sec): 27.92 - samples/sec: 587.29 - lr: 0.000145 - momentum: 0.000000
2023-10-11 10:52:18,476 epoch 2 - iter 108/272 - loss 0.67231305 - time (sec): 36.68 - samples/sec: 574.11 - lr: 0.000143 - momentum: 0.000000
2023-10-11 10:52:27,575 epoch 2 - iter 135/272 - loss 0.63398133 - time (sec): 45.78 - samples/sec: 571.45 - lr: 0.000142 - momentum: 0.000000
2023-10-11 10:52:36,162 epoch 2 - iter 162/272 - loss 0.60685794 - time (sec): 54.36 - samples/sec: 561.27 - lr: 0.000140 - momentum: 0.000000
2023-10-11 10:52:45,245 epoch 2 - iter 189/272 - loss 0.57321778 - time (sec): 63.45 - samples/sec: 556.34 - lr: 0.000138 - momentum: 0.000000
2023-10-11 10:52:54,524 epoch 2 - iter 216/272 - loss 0.55390885 - time (sec): 72.73 - samples/sec: 554.04 - lr: 0.000137 - momentum: 0.000000
2023-10-11 10:53:04,155 epoch 2 - iter 243/272 - loss 0.53466124 - time (sec): 82.36 - samples/sec: 553.70 - lr: 0.000135 - momentum: 0.000000
2023-10-11 10:53:14,344 epoch 2 - iter 270/272 - loss 0.51513046 - time (sec): 92.55 - samples/sec: 558.92 - lr: 0.000134 - momentum: 0.000000
2023-10-11 10:53:14,819 ----------------------------------------------------------------------------------------------------
2023-10-11 10:53:14,819 EPOCH 2 done: loss 0.5129 - lr: 0.000134
2023-10-11 10:53:20,319 DEV : loss 0.30094385147094727 - f1-score (micro avg) 0.318
2023-10-11 10:53:20,328 saving best model
2023-10-11 10:53:21,176 ----------------------------------------------------------------------------------------------------
2023-10-11 10:53:30,341 epoch 3 - iter 27/272 - loss 0.36013238 - time (sec): 9.16 - samples/sec: 557.38 - lr: 0.000132 - momentum: 0.000000
2023-10-11 10:53:39,870 epoch 3 - iter 54/272 - loss 0.33093762 - time (sec): 18.69 - samples/sec: 566.19 - lr: 0.000130 - momentum: 0.000000
2023-10-11 10:53:49,723 epoch 3 - iter 81/272 - loss 0.32749068 - time (sec): 28.54 - samples/sec: 583.92 - lr: 0.000128 - momentum: 0.000000
2023-10-11 10:53:58,383 epoch 3 - iter 108/272 - loss 0.33782747 - time (sec): 37.20 - samples/sec: 565.12 - lr: 0.000127 - momentum: 0.000000
2023-10-11 10:54:07,944 epoch 3 - iter 135/272 - loss 0.33686699 - time (sec): 46.77 - samples/sec: 551.43 - lr: 0.000125 - momentum: 0.000000
2023-10-11 10:54:18,169 epoch 3 - iter 162/272 - loss 0.32243636 - time (sec): 56.99 - samples/sec: 557.39 - lr: 0.000123 - momentum: 0.000000
2023-10-11 10:54:27,286 epoch 3 - iter 189/272 - loss 0.31031933 - time (sec): 66.11 - samples/sec: 558.27 - lr: 0.000122 - momentum: 0.000000
2023-10-11 10:54:36,651 epoch 3 - iter 216/272 - loss 0.29566962 - time (sec): 75.47 - samples/sec: 555.53 - lr: 0.000120 - momentum: 0.000000
2023-10-11 10:54:45,706 epoch 3 - iter 243/272 - loss 0.28469825 - time (sec): 84.53 - samples/sec: 552.48 - lr: 0.000119 - momentum: 0.000000
2023-10-11 10:54:54,928 epoch 3 - iter 270/272 - loss 0.28376485 - time (sec): 93.75 - samples/sec: 552.23 - lr: 0.000117 - momentum: 0.000000
2023-10-11 10:54:55,379 ----------------------------------------------------------------------------------------------------
2023-10-11 10:54:55,380 EPOCH 3 done: loss 0.2831 - lr: 0.000117
2023-10-11 10:55:00,783 DEV : loss 0.20688828825950623 - f1-score (micro avg) 0.5714
2023-10-11 10:55:00,792 saving best model
2023-10-11 10:55:03,337 ----------------------------------------------------------------------------------------------------
2023-10-11 10:55:12,297 epoch 4 - iter 27/272 - loss 0.18883592 - time (sec): 8.96 - samples/sec: 538.43 - lr: 0.000115 - momentum: 0.000000
2023-10-11 10:55:20,299 epoch 4 - iter 54/272 - loss 0.21454971 - time (sec): 16.96 - samples/sec: 510.04 - lr: 0.000113 - momentum: 0.000000
2023-10-11 10:55:29,545 epoch 4 - iter 81/272 - loss 0.22342544 - time (sec): 26.20 - samples/sec: 537.13 - lr: 0.000112 - momentum: 0.000000
2023-10-11 10:55:38,397 epoch 4 - iter 108/272 - loss 0.20820804 - time (sec): 35.06 - samples/sec: 541.85 - lr: 0.000110 - momentum: 0.000000
2023-10-11 10:55:47,376 epoch 4 - iter 135/272 - loss 0.20439682 - time (sec): 44.03 - samples/sec: 543.39 - lr: 0.000108 - momentum: 0.000000
2023-10-11 10:55:55,921 epoch 4 - iter 162/272 - loss 0.19740033 - time (sec): 52.58 - samples/sec: 537.66 - lr: 0.000107 - momentum: 0.000000
2023-10-11 10:56:06,557 epoch 4 - iter 189/272 - loss 0.18830402 - time (sec): 63.22 - samples/sec: 555.38 - lr: 0.000105 - momentum: 0.000000
2023-10-11 10:56:15,674 epoch 4 - iter 216/272 - loss 0.18412200 - time (sec): 72.33 - samples/sec: 557.92 - lr: 0.000103 - momentum: 0.000000
2023-10-11 10:56:25,318 epoch 4 - iter 243/272 - loss 0.17948316 - time (sec): 81.98 - samples/sec: 559.34 - lr: 0.000102 - momentum: 0.000000
2023-10-11 10:56:35,122 epoch 4 - iter 270/272 - loss 0.17495580 - time (sec): 91.78 - samples/sec: 562.76 - lr: 0.000100 - momentum: 0.000000
2023-10-11 10:56:35,663 ----------------------------------------------------------------------------------------------------
2023-10-11 10:56:35,663 EPOCH 4 done: loss 0.1745 - lr: 0.000100
2023-10-11 10:56:41,543 DEV : loss 0.16108796000480652 - f1-score (micro avg) 0.636
2023-10-11 10:56:41,552 saving best model
2023-10-11 10:56:44,080 ----------------------------------------------------------------------------------------------------
2023-10-11 10:56:54,081 epoch 5 - iter 27/272 - loss 0.11279072 - time (sec): 10.00 - samples/sec: 575.65 - lr: 0.000098 - momentum: 0.000000
2023-10-11 10:57:04,141 epoch 5 - iter 54/272 - loss 0.12313934 - time (sec): 20.06 - samples/sec: 554.27 - lr: 0.000097 - momentum: 0.000000
2023-10-11 10:57:13,844 epoch 5 - iter 81/272 - loss 0.11655880 - time (sec): 29.76 - samples/sec: 552.92 - lr: 0.000095 - momentum: 0.000000
2023-10-11 10:57:22,628 epoch 5 - iter 108/272 - loss 0.11981029 - time (sec): 38.54 - samples/sec: 540.70 - lr: 0.000093 - momentum: 0.000000
2023-10-11 10:57:31,316 epoch 5 - iter 135/272 - loss 0.12033932 - time (sec): 47.23 - samples/sec: 534.06 - lr: 0.000092 - momentum: 0.000000
2023-10-11 10:57:40,567 epoch 5 - iter 162/272 - loss 0.12185608 - time (sec): 56.48 - samples/sec: 536.12 - lr: 0.000090 - momentum: 0.000000
2023-10-11 10:57:49,789 epoch 5 - iter 189/272 - loss 0.11473005 - time (sec): 65.71 - samples/sec: 536.88 - lr: 0.000088 - momentum: 0.000000
2023-10-11 10:57:59,333 epoch 5 - iter 216/272 - loss 0.11661259 - time (sec): 75.25 - samples/sec: 538.49 - lr: 0.000087 - momentum: 0.000000
2023-10-11 10:58:09,636 epoch 5 - iter 243/272 - loss 0.11767849 - time (sec): 85.55 - samples/sec: 547.14 - lr: 0.000085 - momentum: 0.000000
2023-10-11 10:58:18,905 epoch 5 - iter 270/272 - loss 0.11723516 - time (sec): 94.82 - samples/sec: 546.86 - lr: 0.000084 - momentum: 0.000000
2023-10-11 10:58:19,303 ----------------------------------------------------------------------------------------------------
2023-10-11 10:58:19,303 EPOCH 5 done: loss 0.1171 - lr: 0.000084
2023-10-11 10:58:24,880 DEV : loss 0.13878858089447021 - f1-score (micro avg) 0.7559
2023-10-11 10:58:24,889 saving best model
2023-10-11 10:58:27,426 ----------------------------------------------------------------------------------------------------
2023-10-11 10:58:36,265 epoch 6 - iter 27/272 - loss 0.07309716 - time (sec): 8.83 - samples/sec: 529.07 - lr: 0.000082 - momentum: 0.000000
2023-10-11 10:58:45,128 epoch 6 - iter 54/272 - loss 0.08693004 - time (sec): 17.70 - samples/sec: 521.43 - lr: 0.000080 - momentum: 0.000000
2023-10-11 10:58:54,405 epoch 6 - iter 81/272 - loss 0.09456387 - time (sec): 26.97 - samples/sec: 523.42 - lr: 0.000078 - momentum: 0.000000
2023-10-11 10:59:04,184 epoch 6 - iter 108/272 - loss 0.08984032 - time (sec): 36.75 - samples/sec: 537.09 - lr: 0.000077 - momentum: 0.000000
2023-10-11 10:59:14,018 epoch 6 - iter 135/272 - loss 0.08508482 - time (sec): 46.59 - samples/sec: 552.64 - lr: 0.000075 - momentum: 0.000000
2023-10-11 10:59:23,202 epoch 6 - iter 162/272 - loss 0.08590144 - time (sec): 55.77 - samples/sec: 541.76 - lr: 0.000073 - momentum: 0.000000
2023-10-11 10:59:32,673 epoch 6 - iter 189/272 - loss 0.08188597 - time (sec): 65.24 - samples/sec: 546.70 - lr: 0.000072 - momentum: 0.000000
2023-10-11 10:59:41,961 epoch 6 - iter 216/272 - loss 0.08590837 - time (sec): 74.53 - samples/sec: 544.08 - lr: 0.000070 - momentum: 0.000000
2023-10-11 10:59:51,615 epoch 6 - iter 243/272 - loss 0.08500069 - time (sec): 84.18 - samples/sec: 546.89 - lr: 0.000069 - momentum: 0.000000
2023-10-11 11:00:01,257 epoch 6 - iter 270/272 - loss 0.08427618 - time (sec): 93.83 - samples/sec: 548.96 - lr: 0.000067 - momentum: 0.000000
2023-10-11 11:00:01,944 ----------------------------------------------------------------------------------------------------
2023-10-11 11:00:01,944 EPOCH 6 done: loss 0.0842 - lr: 0.000067
2023-10-11 11:00:07,586 DEV : loss 0.13508526980876923 - f1-score (micro avg) 0.7784
2023-10-11 11:00:07,594 saving best model
2023-10-11 11:00:10,077 ----------------------------------------------------------------------------------------------------
2023-10-11 11:00:19,490 epoch 7 - iter 27/272 - loss 0.06349629 - time (sec): 9.41 - samples/sec: 572.82 - lr: 0.000065 - momentum: 0.000000
2023-10-11 11:00:28,354 epoch 7 - iter 54/272 - loss 0.07168990 - time (sec): 18.27 - samples/sec: 558.79 - lr: 0.000063 - momentum: 0.000000
2023-10-11 11:00:38,273 epoch 7 - iter 81/272 - loss 0.06744160 - time (sec): 28.19 - samples/sec: 569.92 - lr: 0.000062 - momentum: 0.000000
2023-10-11 11:00:47,594 epoch 7 - iter 108/272 - loss 0.06533570 - time (sec): 37.51 - samples/sec: 569.37 - lr: 0.000060 - momentum: 0.000000
2023-10-11 11:00:56,990 epoch 7 - iter 135/272 - loss 0.07004227 - time (sec): 46.91 - samples/sec: 569.69 - lr: 0.000058 - momentum: 0.000000
2023-10-11 11:01:06,041 epoch 7 - iter 162/272 - loss 0.06680624 - time (sec): 55.96 - samples/sec: 564.74 - lr: 0.000057 - momentum: 0.000000
2023-10-11 11:01:15,636 epoch 7 - iter 189/272 - loss 0.07011620 - time (sec): 65.56 - samples/sec: 563.17 - lr: 0.000055 - momentum: 0.000000
2023-10-11 11:01:24,444 epoch 7 - iter 216/272 - loss 0.06882717 - time (sec): 74.36 - samples/sec: 557.55 - lr: 0.000053 - momentum: 0.000000
2023-10-11 11:01:34,079 epoch 7 - iter 243/272 - loss 0.06683075 - time (sec): 84.00 - samples/sec: 557.83 - lr: 0.000052 - momentum: 0.000000
2023-10-11 11:01:43,455 epoch 7 - iter 270/272 - loss 0.06398691 - time (sec): 93.37 - samples/sec: 554.75 - lr: 0.000050 - momentum: 0.000000
2023-10-11 11:01:43,855 ----------------------------------------------------------------------------------------------------
2023-10-11 11:01:43,855 EPOCH 7 done: loss 0.0638 - lr: 0.000050
2023-10-11 11:01:49,480 DEV : loss 0.13904628157615662 - f1-score (micro avg) 0.7544
2023-10-11 11:01:49,488 ----------------------------------------------------------------------------------------------------
2023-10-11 11:01:58,810 epoch 8 - iter 27/272 - loss 0.05151019 - time (sec): 9.32 - samples/sec: 555.05 - lr: 0.000048 - momentum: 0.000000
2023-10-11 11:02:07,667 epoch 8 - iter 54/272 - loss 0.04875397 - time (sec): 18.18 - samples/sec: 538.60 - lr: 0.000047 - momentum: 0.000000
2023-10-11 11:02:17,611 epoch 8 - iter 81/272 - loss 0.05284993 - time (sec): 28.12 - samples/sec: 547.95 - lr: 0.000045 - momentum: 0.000000
2023-10-11 11:02:26,809 epoch 8 - iter 108/272 - loss 0.05991550 - time (sec): 37.32 - samples/sec: 544.64 - lr: 0.000043 - momentum: 0.000000
2023-10-11 11:02:36,284 epoch 8 - iter 135/272 - loss 0.05652978 - time (sec): 46.79 - samples/sec: 546.34 - lr: 0.000042 - momentum: 0.000000
2023-10-11 11:02:45,701 epoch 8 - iter 162/272 - loss 0.05626381 - time (sec): 56.21 - samples/sec: 547.48 - lr: 0.000040 - momentum: 0.000000
2023-10-11 11:02:55,335 epoch 8 - iter 189/272 - loss 0.05419822 - time (sec): 65.85 - samples/sec: 549.05 - lr: 0.000038 - momentum: 0.000000
2023-10-11 11:03:05,159 epoch 8 - iter 216/272 - loss 0.05356223 - time (sec): 75.67 - samples/sec: 549.84 - lr: 0.000037 - momentum: 0.000000
2023-10-11 11:03:14,930 epoch 8 - iter 243/272 - loss 0.05043893 - time (sec): 85.44 - samples/sec: 551.54 - lr: 0.000035 - momentum: 0.000000
2023-10-11 11:03:23,724 epoch 8 - iter 270/272 - loss 0.05041046 - time (sec): 94.23 - samples/sec: 548.22 - lr: 0.000034 - momentum: 0.000000
2023-10-11 11:03:24,272 ----------------------------------------------------------------------------------------------------
2023-10-11 11:03:24,272 EPOCH 8 done: loss 0.0502 - lr: 0.000034
2023-10-11 11:03:30,009 DEV : loss 0.13503918051719666 - f1-score (micro avg) 0.7964
2023-10-11 11:03:30,018 saving best model
2023-10-11 11:03:32,544 ----------------------------------------------------------------------------------------------------
2023-10-11 11:03:42,316 epoch 9 - iter 27/272 - loss 0.04416837 - time (sec): 9.77 - samples/sec: 597.14 - lr: 0.000032 - momentum: 0.000000
2023-10-11 11:03:51,802 epoch 9 - iter 54/272 - loss 0.04781217 - time (sec): 19.25 - samples/sec: 577.94 - lr: 0.000030 - momentum: 0.000000
2023-10-11 11:04:00,376 epoch 9 - iter 81/272 - loss 0.04608928 - time (sec): 27.83 - samples/sec: 558.18 - lr: 0.000028 - momentum: 0.000000
2023-10-11 11:04:09,553 epoch 9 - iter 108/272 - loss 0.04616787 - time (sec): 37.01 - samples/sec: 556.51 - lr: 0.000027 - momentum: 0.000000
2023-10-11 11:04:19,183 epoch 9 - iter 135/272 - loss 0.04546524 - time (sec): 46.63 - samples/sec: 558.30 - lr: 0.000025 - momentum: 0.000000
2023-10-11 11:04:28,689 epoch 9 - iter 162/272 - loss 0.04381471 - time (sec): 56.14 - samples/sec: 559.12 - lr: 0.000023 - momentum: 0.000000
2023-10-11 11:04:38,019 epoch 9 - iter 189/272 - loss 0.04496020 - time (sec): 65.47 - samples/sec: 554.38 - lr: 0.000022 - momentum: 0.000000
2023-10-11 11:04:47,008 epoch 9 - iter 216/272 - loss 0.04477837 - time (sec): 74.46 - samples/sec: 551.81 - lr: 0.000020 - momentum: 0.000000
2023-10-11 11:04:57,105 epoch 9 - iter 243/272 - loss 0.04362140 - time (sec): 84.56 - samples/sec: 556.23 - lr: 0.000019 - momentum: 0.000000
2023-10-11 11:05:06,298 epoch 9 - iter 270/272 - loss 0.04226270 - time (sec): 93.75 - samples/sec: 553.27 - lr: 0.000017 - momentum: 0.000000
2023-10-11 11:05:06,661 ----------------------------------------------------------------------------------------------------
2023-10-11 11:05:06,662 EPOCH 9 done: loss 0.0423 - lr: 0.000017
2023-10-11 11:05:12,189 DEV : loss 0.13971544802188873 - f1-score (micro avg) 0.7899
2023-10-11 11:05:12,198 ----------------------------------------------------------------------------------------------------
2023-10-11 11:05:22,115 epoch 10 - iter 27/272 - loss 0.04362523 - time (sec): 9.92 - samples/sec: 549.26 - lr: 0.000015 - momentum: 0.000000
2023-10-11 11:05:30,817 epoch 10 - iter 54/272 - loss 0.04805838 - time (sec): 18.62 - samples/sec: 523.99 - lr: 0.000013 - momentum: 0.000000
2023-10-11 11:05:41,534 epoch 10 - iter 81/272 - loss 0.04944586 - time (sec): 29.33 - samples/sec: 557.68 - lr: 0.000012 - momentum: 0.000000
2023-10-11 11:05:51,898 epoch 10 - iter 108/272 - loss 0.04975532 - time (sec): 39.70 - samples/sec: 565.25 - lr: 0.000010 - momentum: 0.000000
2023-10-11 11:06:01,700 epoch 10 - iter 135/272 - loss 0.04678516 - time (sec): 49.50 - samples/sec: 561.01 - lr: 0.000008 - momentum: 0.000000
2023-10-11 11:06:10,214 epoch 10 - iter 162/272 - loss 0.04446697 - time (sec): 58.01 - samples/sec: 549.99 - lr: 0.000007 - momentum: 0.000000
2023-10-11 11:06:20,152 epoch 10 - iter 189/272 - loss 0.04288849 - time (sec): 67.95 - samples/sec: 550.37 - lr: 0.000005 - momentum: 0.000000
2023-10-11 11:06:29,363 epoch 10 - iter 216/272 - loss 0.04070236 - time (sec): 77.16 - samples/sec: 543.35 - lr: 0.000003 - momentum: 0.000000
2023-10-11 11:06:38,730 epoch 10 - iter 243/272 - loss 0.03991694 - time (sec): 86.53 - samples/sec: 543.33 - lr: 0.000002 - momentum: 0.000000
2023-10-11 11:06:47,930 epoch 10 - iter 270/272 - loss 0.03815397 - time (sec): 95.73 - samples/sec: 540.55 - lr: 0.000000 - momentum: 0.000000
2023-10-11 11:06:48,386 ----------------------------------------------------------------------------------------------------
2023-10-11 11:06:48,387 EPOCH 10 done: loss 0.0382 - lr: 0.000000
2023-10-11 11:06:54,102 DEV : loss 0.1388687789440155 - f1-score (micro avg) 0.7985
2023-10-11 11:06:54,111 saving best model
2023-10-11 11:06:57,461 ----------------------------------------------------------------------------------------------------
2023-10-11 11:06:57,463 Loading model from best epoch ...
2023-10-11 11:07:01,068 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 11:07:13,976
Results:
- F-score (micro) 0.7448
- F-score (macro) 0.6813
- Accuracy 0.6137
By class:
precision recall f1-score support
LOC 0.7652 0.8462 0.8037 312
PER 0.6357 0.8558 0.7295 208
ORG 0.4706 0.4364 0.4528 55
HumanProd 0.7083 0.7727 0.7391 22
micro avg 0.6900 0.8090 0.7448 597
macro avg 0.6450 0.7278 0.6813 597
weighted avg 0.6909 0.8090 0.7431 597
2023-10-11 11:07:13,976 ----------------------------------------------------------------------------------------------------
|