File size: 25,289 Bytes
458f2cb |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 |
2023-10-11 09:04:56,657 ----------------------------------------------------------------------------------------------------
2023-10-11 09:04:56,659 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 09:04:56,659 ----------------------------------------------------------------------------------------------------
2023-10-11 09:04:56,659 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 09:04:56,659 ----------------------------------------------------------------------------------------------------
2023-10-11 09:04:56,659 Train: 1085 sentences
2023-10-11 09:04:56,659 (train_with_dev=False, train_with_test=False)
2023-10-11 09:04:56,659 ----------------------------------------------------------------------------------------------------
2023-10-11 09:04:56,660 Training Params:
2023-10-11 09:04:56,660 - learning_rate: "0.00015"
2023-10-11 09:04:56,660 - mini_batch_size: "8"
2023-10-11 09:04:56,660 - max_epochs: "10"
2023-10-11 09:04:56,660 - shuffle: "True"
2023-10-11 09:04:56,660 ----------------------------------------------------------------------------------------------------
2023-10-11 09:04:56,660 Plugins:
2023-10-11 09:04:56,660 - TensorboardLogger
2023-10-11 09:04:56,660 - LinearScheduler | warmup_fraction: '0.1'
2023-10-11 09:04:56,660 ----------------------------------------------------------------------------------------------------
2023-10-11 09:04:56,660 Final evaluation on model from best epoch (best-model.pt)
2023-10-11 09:04:56,660 - metric: "('micro avg', 'f1-score')"
2023-10-11 09:04:56,660 ----------------------------------------------------------------------------------------------------
2023-10-11 09:04:56,660 Computation:
2023-10-11 09:04:56,660 - compute on device: cuda:0
2023-10-11 09:04:56,661 - embedding storage: none
2023-10-11 09:04:56,661 ----------------------------------------------------------------------------------------------------
2023-10-11 09:04:56,661 Model training base path: "hmbench-newseye/sv-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs8-wsFalse-e10-lr0.00015-poolingfirst-layers-1-crfFalse-2"
2023-10-11 09:04:56,661 ----------------------------------------------------------------------------------------------------
2023-10-11 09:04:56,661 ----------------------------------------------------------------------------------------------------
2023-10-11 09:04:56,661 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-11 09:05:05,324 epoch 1 - iter 13/136 - loss 2.85454415 - time (sec): 8.66 - samples/sec: 591.27 - lr: 0.000013 - momentum: 0.000000
2023-10-11 09:05:13,976 epoch 1 - iter 26/136 - loss 2.84857093 - time (sec): 17.31 - samples/sec: 539.68 - lr: 0.000028 - momentum: 0.000000
2023-10-11 09:05:22,695 epoch 1 - iter 39/136 - loss 2.83797201 - time (sec): 26.03 - samples/sec: 562.37 - lr: 0.000042 - momentum: 0.000000
2023-10-11 09:05:31,290 epoch 1 - iter 52/136 - loss 2.81925649 - time (sec): 34.63 - samples/sec: 571.88 - lr: 0.000056 - momentum: 0.000000
2023-10-11 09:05:39,850 epoch 1 - iter 65/136 - loss 2.78851902 - time (sec): 43.19 - samples/sec: 579.84 - lr: 0.000071 - momentum: 0.000000
2023-10-11 09:05:48,157 epoch 1 - iter 78/136 - loss 2.73824936 - time (sec): 51.49 - samples/sec: 579.76 - lr: 0.000085 - momentum: 0.000000
2023-10-11 09:05:56,653 epoch 1 - iter 91/136 - loss 2.66944775 - time (sec): 59.99 - samples/sec: 580.76 - lr: 0.000099 - momentum: 0.000000
2023-10-11 09:06:05,007 epoch 1 - iter 104/136 - loss 2.59680995 - time (sec): 68.34 - samples/sec: 579.67 - lr: 0.000114 - momentum: 0.000000
2023-10-11 09:06:14,148 epoch 1 - iter 117/136 - loss 2.50961737 - time (sec): 77.49 - samples/sec: 580.56 - lr: 0.000128 - momentum: 0.000000
2023-10-11 09:06:22,904 epoch 1 - iter 130/136 - loss 2.42829796 - time (sec): 86.24 - samples/sec: 581.62 - lr: 0.000142 - momentum: 0.000000
2023-10-11 09:06:26,301 ----------------------------------------------------------------------------------------------------
2023-10-11 09:06:26,302 EPOCH 1 done: loss 2.3990 - lr: 0.000142
2023-10-11 09:06:31,357 DEV : loss 1.4164044857025146 - f1-score (micro avg) 0.0
2023-10-11 09:06:31,365 ----------------------------------------------------------------------------------------------------
2023-10-11 09:06:40,344 epoch 2 - iter 13/136 - loss 1.39753845 - time (sec): 8.98 - samples/sec: 620.23 - lr: 0.000149 - momentum: 0.000000
2023-10-11 09:06:48,792 epoch 2 - iter 26/136 - loss 1.30612822 - time (sec): 17.43 - samples/sec: 595.58 - lr: 0.000147 - momentum: 0.000000
2023-10-11 09:06:58,251 epoch 2 - iter 39/136 - loss 1.24698017 - time (sec): 26.88 - samples/sec: 601.44 - lr: 0.000145 - momentum: 0.000000
2023-10-11 09:07:06,826 epoch 2 - iter 52/136 - loss 1.16189236 - time (sec): 35.46 - samples/sec: 591.11 - lr: 0.000144 - momentum: 0.000000
2023-10-11 09:07:15,457 epoch 2 - iter 65/136 - loss 1.10052914 - time (sec): 44.09 - samples/sec: 585.86 - lr: 0.000142 - momentum: 0.000000
2023-10-11 09:07:23,553 epoch 2 - iter 78/136 - loss 1.04318846 - time (sec): 52.19 - samples/sec: 579.64 - lr: 0.000141 - momentum: 0.000000
2023-10-11 09:07:32,021 epoch 2 - iter 91/136 - loss 0.99457757 - time (sec): 60.65 - samples/sec: 575.15 - lr: 0.000139 - momentum: 0.000000
2023-10-11 09:07:40,643 epoch 2 - iter 104/136 - loss 0.93611893 - time (sec): 69.28 - samples/sec: 575.52 - lr: 0.000137 - momentum: 0.000000
2023-10-11 09:07:49,116 epoch 2 - iter 117/136 - loss 0.90403324 - time (sec): 77.75 - samples/sec: 573.59 - lr: 0.000136 - momentum: 0.000000
2023-10-11 09:07:57,517 epoch 2 - iter 130/136 - loss 0.88225517 - time (sec): 86.15 - samples/sec: 572.13 - lr: 0.000134 - momentum: 0.000000
2023-10-11 09:08:01,685 ----------------------------------------------------------------------------------------------------
2023-10-11 09:08:01,685 EPOCH 2 done: loss 0.8649 - lr: 0.000134
2023-10-11 09:08:07,581 DEV : loss 0.49027636647224426 - f1-score (micro avg) 0.0
2023-10-11 09:08:07,589 ----------------------------------------------------------------------------------------------------
2023-10-11 09:08:16,565 epoch 3 - iter 13/136 - loss 0.54946705 - time (sec): 8.97 - samples/sec: 579.70 - lr: 0.000132 - momentum: 0.000000
2023-10-11 09:08:25,341 epoch 3 - iter 26/136 - loss 0.57802032 - time (sec): 17.75 - samples/sec: 563.01 - lr: 0.000130 - momentum: 0.000000
2023-10-11 09:08:34,098 epoch 3 - iter 39/136 - loss 0.52629489 - time (sec): 26.51 - samples/sec: 557.63 - lr: 0.000129 - momentum: 0.000000
2023-10-11 09:08:42,539 epoch 3 - iter 52/136 - loss 0.50784907 - time (sec): 34.95 - samples/sec: 550.19 - lr: 0.000127 - momentum: 0.000000
2023-10-11 09:08:52,029 epoch 3 - iter 65/136 - loss 0.49141220 - time (sec): 44.44 - samples/sec: 563.15 - lr: 0.000126 - momentum: 0.000000
2023-10-11 09:09:00,500 epoch 3 - iter 78/136 - loss 0.46973894 - time (sec): 52.91 - samples/sec: 564.60 - lr: 0.000124 - momentum: 0.000000
2023-10-11 09:09:10,091 epoch 3 - iter 91/136 - loss 0.45652392 - time (sec): 62.50 - samples/sec: 574.68 - lr: 0.000122 - momentum: 0.000000
2023-10-11 09:09:18,756 epoch 3 - iter 104/136 - loss 0.44771354 - time (sec): 71.17 - samples/sec: 575.80 - lr: 0.000121 - momentum: 0.000000
2023-10-11 09:09:27,472 epoch 3 - iter 117/136 - loss 0.43194864 - time (sec): 79.88 - samples/sec: 575.18 - lr: 0.000119 - momentum: 0.000000
2023-10-11 09:09:35,262 epoch 3 - iter 130/136 - loss 0.42684263 - time (sec): 87.67 - samples/sec: 570.29 - lr: 0.000118 - momentum: 0.000000
2023-10-11 09:09:38,903 ----------------------------------------------------------------------------------------------------
2023-10-11 09:09:38,903 EPOCH 3 done: loss 0.4273 - lr: 0.000118
2023-10-11 09:09:45,058 DEV : loss 0.30092161893844604 - f1-score (micro avg) 0.2491
2023-10-11 09:09:45,067 saving best model
2023-10-11 09:09:45,936 ----------------------------------------------------------------------------------------------------
2023-10-11 09:09:54,331 epoch 4 - iter 13/136 - loss 0.32672183 - time (sec): 8.39 - samples/sec: 568.98 - lr: 0.000115 - momentum: 0.000000
2023-10-11 09:10:02,521 epoch 4 - iter 26/136 - loss 0.33454165 - time (sec): 16.58 - samples/sec: 551.43 - lr: 0.000114 - momentum: 0.000000
2023-10-11 09:10:11,774 epoch 4 - iter 39/136 - loss 0.31845525 - time (sec): 25.84 - samples/sec: 577.18 - lr: 0.000112 - momentum: 0.000000
2023-10-11 09:10:20,950 epoch 4 - iter 52/136 - loss 0.31203326 - time (sec): 35.01 - samples/sec: 586.29 - lr: 0.000111 - momentum: 0.000000
2023-10-11 09:10:29,119 epoch 4 - iter 65/136 - loss 0.30935257 - time (sec): 43.18 - samples/sec: 581.35 - lr: 0.000109 - momentum: 0.000000
2023-10-11 09:10:37,905 epoch 4 - iter 78/136 - loss 0.30389004 - time (sec): 51.97 - samples/sec: 584.76 - lr: 0.000107 - momentum: 0.000000
2023-10-11 09:10:46,403 epoch 4 - iter 91/136 - loss 0.31155565 - time (sec): 60.46 - samples/sec: 579.12 - lr: 0.000106 - momentum: 0.000000
2023-10-11 09:10:55,125 epoch 4 - iter 104/136 - loss 0.30416576 - time (sec): 69.19 - samples/sec: 578.07 - lr: 0.000104 - momentum: 0.000000
2023-10-11 09:11:03,837 epoch 4 - iter 117/136 - loss 0.31157005 - time (sec): 77.90 - samples/sec: 574.88 - lr: 0.000103 - momentum: 0.000000
2023-10-11 09:11:12,717 epoch 4 - iter 130/136 - loss 0.31360435 - time (sec): 86.78 - samples/sec: 572.74 - lr: 0.000101 - momentum: 0.000000
2023-10-11 09:11:17,287 ----------------------------------------------------------------------------------------------------
2023-10-11 09:11:17,287 EPOCH 4 done: loss 0.3110 - lr: 0.000101
2023-10-11 09:11:23,281 DEV : loss 0.2556583285331726 - f1-score (micro avg) 0.3513
2023-10-11 09:11:23,297 saving best model
2023-10-11 09:11:25,880 ----------------------------------------------------------------------------------------------------
2023-10-11 09:11:33,874 epoch 5 - iter 13/136 - loss 0.28788963 - time (sec): 7.99 - samples/sec: 567.52 - lr: 0.000099 - momentum: 0.000000
2023-10-11 09:11:42,322 epoch 5 - iter 26/136 - loss 0.27712001 - time (sec): 16.44 - samples/sec: 590.10 - lr: 0.000097 - momentum: 0.000000
2023-10-11 09:11:51,282 epoch 5 - iter 39/136 - loss 0.26346164 - time (sec): 25.40 - samples/sec: 591.58 - lr: 0.000096 - momentum: 0.000000
2023-10-11 09:11:59,175 epoch 5 - iter 52/136 - loss 0.26906131 - time (sec): 33.29 - samples/sec: 579.28 - lr: 0.000094 - momentum: 0.000000
2023-10-11 09:12:07,515 epoch 5 - iter 65/136 - loss 0.25519490 - time (sec): 41.63 - samples/sec: 580.16 - lr: 0.000092 - momentum: 0.000000
2023-10-11 09:12:16,284 epoch 5 - iter 78/136 - loss 0.24701124 - time (sec): 50.40 - samples/sec: 584.95 - lr: 0.000091 - momentum: 0.000000
2023-10-11 09:12:25,010 epoch 5 - iter 91/136 - loss 0.24986196 - time (sec): 59.13 - samples/sec: 583.53 - lr: 0.000089 - momentum: 0.000000
2023-10-11 09:12:33,752 epoch 5 - iter 104/136 - loss 0.25823714 - time (sec): 67.87 - samples/sec: 585.17 - lr: 0.000088 - momentum: 0.000000
2023-10-11 09:12:42,624 epoch 5 - iter 117/136 - loss 0.26070813 - time (sec): 76.74 - samples/sec: 586.28 - lr: 0.000086 - momentum: 0.000000
2023-10-11 09:12:50,866 epoch 5 - iter 130/136 - loss 0.26182593 - time (sec): 84.98 - samples/sec: 583.96 - lr: 0.000084 - momentum: 0.000000
2023-10-11 09:12:54,779 ----------------------------------------------------------------------------------------------------
2023-10-11 09:12:54,779 EPOCH 5 done: loss 0.2621 - lr: 0.000084
2023-10-11 09:13:00,530 DEV : loss 0.2347497045993805 - f1-score (micro avg) 0.3522
2023-10-11 09:13:00,539 saving best model
2023-10-11 09:13:03,115 ----------------------------------------------------------------------------------------------------
2023-10-11 09:13:11,908 epoch 6 - iter 13/136 - loss 0.26730900 - time (sec): 8.79 - samples/sec: 590.52 - lr: 0.000082 - momentum: 0.000000
2023-10-11 09:13:20,432 epoch 6 - iter 26/136 - loss 0.25691153 - time (sec): 17.31 - samples/sec: 572.24 - lr: 0.000081 - momentum: 0.000000
2023-10-11 09:13:29,176 epoch 6 - iter 39/136 - loss 0.24647921 - time (sec): 26.06 - samples/sec: 573.94 - lr: 0.000079 - momentum: 0.000000
2023-10-11 09:13:37,933 epoch 6 - iter 52/136 - loss 0.25068436 - time (sec): 34.81 - samples/sec: 589.21 - lr: 0.000077 - momentum: 0.000000
2023-10-11 09:13:45,698 epoch 6 - iter 65/136 - loss 0.24521590 - time (sec): 42.58 - samples/sec: 582.87 - lr: 0.000076 - momentum: 0.000000
2023-10-11 09:13:54,218 epoch 6 - iter 78/136 - loss 0.23230441 - time (sec): 51.10 - samples/sec: 589.09 - lr: 0.000074 - momentum: 0.000000
2023-10-11 09:14:02,808 epoch 6 - iter 91/136 - loss 0.22651680 - time (sec): 59.69 - samples/sec: 594.18 - lr: 0.000073 - momentum: 0.000000
2023-10-11 09:14:10,918 epoch 6 - iter 104/136 - loss 0.23136907 - time (sec): 67.80 - samples/sec: 591.72 - lr: 0.000071 - momentum: 0.000000
2023-10-11 09:14:19,378 epoch 6 - iter 117/136 - loss 0.22970596 - time (sec): 76.26 - samples/sec: 589.50 - lr: 0.000069 - momentum: 0.000000
2023-10-11 09:14:27,857 epoch 6 - iter 130/136 - loss 0.22534419 - time (sec): 84.74 - samples/sec: 589.45 - lr: 0.000068 - momentum: 0.000000
2023-10-11 09:14:31,402 ----------------------------------------------------------------------------------------------------
2023-10-11 09:14:31,403 EPOCH 6 done: loss 0.2242 - lr: 0.000068
2023-10-11 09:14:37,061 DEV : loss 0.2328910082578659 - f1-score (micro avg) 0.4469
2023-10-11 09:14:37,069 saving best model
2023-10-11 09:14:39,620 ----------------------------------------------------------------------------------------------------
2023-10-11 09:14:48,048 epoch 7 - iter 13/136 - loss 0.20055562 - time (sec): 8.42 - samples/sec: 546.90 - lr: 0.000066 - momentum: 0.000000
2023-10-11 09:14:56,720 epoch 7 - iter 26/136 - loss 0.19132013 - time (sec): 17.10 - samples/sec: 584.66 - lr: 0.000064 - momentum: 0.000000
2023-10-11 09:15:05,870 epoch 7 - iter 39/136 - loss 0.18504756 - time (sec): 26.25 - samples/sec: 597.42 - lr: 0.000062 - momentum: 0.000000
2023-10-11 09:15:13,741 epoch 7 - iter 52/136 - loss 0.18716742 - time (sec): 34.12 - samples/sec: 592.46 - lr: 0.000061 - momentum: 0.000000
2023-10-11 09:15:22,691 epoch 7 - iter 65/136 - loss 0.18737032 - time (sec): 43.07 - samples/sec: 594.71 - lr: 0.000059 - momentum: 0.000000
2023-10-11 09:15:31,175 epoch 7 - iter 78/136 - loss 0.18893187 - time (sec): 51.55 - samples/sec: 590.61 - lr: 0.000058 - momentum: 0.000000
2023-10-11 09:15:38,773 epoch 7 - iter 91/136 - loss 0.19416762 - time (sec): 59.15 - samples/sec: 580.16 - lr: 0.000056 - momentum: 0.000000
2023-10-11 09:15:47,664 epoch 7 - iter 104/136 - loss 0.19167814 - time (sec): 68.04 - samples/sec: 579.16 - lr: 0.000054 - momentum: 0.000000
2023-10-11 09:15:56,563 epoch 7 - iter 117/136 - loss 0.19355860 - time (sec): 76.94 - samples/sec: 580.89 - lr: 0.000053 - momentum: 0.000000
2023-10-11 09:16:04,831 epoch 7 - iter 130/136 - loss 0.19427477 - time (sec): 85.21 - samples/sec: 577.98 - lr: 0.000051 - momentum: 0.000000
2023-10-11 09:16:09,097 ----------------------------------------------------------------------------------------------------
2023-10-11 09:16:09,097 EPOCH 7 done: loss 0.1945 - lr: 0.000051
2023-10-11 09:16:14,969 DEV : loss 0.20801755785942078 - f1-score (micro avg) 0.5008
2023-10-11 09:16:14,978 saving best model
2023-10-11 09:16:17,554 ----------------------------------------------------------------------------------------------------
2023-10-11 09:16:25,728 epoch 8 - iter 13/136 - loss 0.19111255 - time (sec): 8.17 - samples/sec: 553.12 - lr: 0.000049 - momentum: 0.000000
2023-10-11 09:16:33,725 epoch 8 - iter 26/136 - loss 0.16263921 - time (sec): 16.17 - samples/sec: 555.57 - lr: 0.000047 - momentum: 0.000000
2023-10-11 09:16:42,962 epoch 8 - iter 39/136 - loss 0.17294011 - time (sec): 25.40 - samples/sec: 580.06 - lr: 0.000046 - momentum: 0.000000
2023-10-11 09:16:51,627 epoch 8 - iter 52/136 - loss 0.17653631 - time (sec): 34.07 - samples/sec: 583.04 - lr: 0.000044 - momentum: 0.000000
2023-10-11 09:17:00,030 epoch 8 - iter 65/136 - loss 0.17843550 - time (sec): 42.47 - samples/sec: 579.20 - lr: 0.000043 - momentum: 0.000000
2023-10-11 09:17:08,509 epoch 8 - iter 78/136 - loss 0.18239003 - time (sec): 50.95 - samples/sec: 579.74 - lr: 0.000041 - momentum: 0.000000
2023-10-11 09:17:17,131 epoch 8 - iter 91/136 - loss 0.17848480 - time (sec): 59.57 - samples/sec: 583.57 - lr: 0.000039 - momentum: 0.000000
2023-10-11 09:17:26,021 epoch 8 - iter 104/136 - loss 0.17535801 - time (sec): 68.46 - samples/sec: 589.57 - lr: 0.000038 - momentum: 0.000000
2023-10-11 09:17:34,586 epoch 8 - iter 117/136 - loss 0.17156513 - time (sec): 77.03 - samples/sec: 588.99 - lr: 0.000036 - momentum: 0.000000
2023-10-11 09:17:42,440 epoch 8 - iter 130/136 - loss 0.17067507 - time (sec): 84.88 - samples/sec: 586.87 - lr: 0.000035 - momentum: 0.000000
2023-10-11 09:17:46,087 ----------------------------------------------------------------------------------------------------
2023-10-11 09:17:46,088 EPOCH 8 done: loss 0.1690 - lr: 0.000035
2023-10-11 09:17:51,758 DEV : loss 0.2015739232301712 - f1-score (micro avg) 0.5357
2023-10-11 09:17:51,766 saving best model
2023-10-11 09:17:54,288 ----------------------------------------------------------------------------------------------------
2023-10-11 09:18:02,209 epoch 9 - iter 13/136 - loss 0.19176870 - time (sec): 7.92 - samples/sec: 521.35 - lr: 0.000032 - momentum: 0.000000
2023-10-11 09:18:10,535 epoch 9 - iter 26/136 - loss 0.17373868 - time (sec): 16.24 - samples/sec: 544.69 - lr: 0.000031 - momentum: 0.000000
2023-10-11 09:18:19,288 epoch 9 - iter 39/136 - loss 0.16354683 - time (sec): 25.00 - samples/sec: 560.01 - lr: 0.000029 - momentum: 0.000000
2023-10-11 09:18:28,315 epoch 9 - iter 52/136 - loss 0.17111041 - time (sec): 34.02 - samples/sec: 549.24 - lr: 0.000028 - momentum: 0.000000
2023-10-11 09:18:37,045 epoch 9 - iter 65/136 - loss 0.17730382 - time (sec): 42.75 - samples/sec: 543.18 - lr: 0.000026 - momentum: 0.000000
2023-10-11 09:18:46,344 epoch 9 - iter 78/136 - loss 0.16675119 - time (sec): 52.05 - samples/sec: 550.14 - lr: 0.000024 - momentum: 0.000000
2023-10-11 09:18:54,829 epoch 9 - iter 91/136 - loss 0.16491876 - time (sec): 60.54 - samples/sec: 552.88 - lr: 0.000023 - momentum: 0.000000
2023-10-11 09:19:04,115 epoch 9 - iter 104/136 - loss 0.16242229 - time (sec): 69.82 - samples/sec: 562.19 - lr: 0.000021 - momentum: 0.000000
2023-10-11 09:19:13,031 epoch 9 - iter 117/136 - loss 0.15534520 - time (sec): 78.74 - samples/sec: 565.64 - lr: 0.000020 - momentum: 0.000000
2023-10-11 09:19:21,897 epoch 9 - iter 130/136 - loss 0.15219382 - time (sec): 87.61 - samples/sec: 566.79 - lr: 0.000018 - momentum: 0.000000
2023-10-11 09:19:25,761 ----------------------------------------------------------------------------------------------------
2023-10-11 09:19:25,762 EPOCH 9 done: loss 0.1517 - lr: 0.000018
2023-10-11 09:19:31,594 DEV : loss 0.1982397586107254 - f1-score (micro avg) 0.5445
2023-10-11 09:19:31,603 saving best model
2023-10-11 09:19:34,155 ----------------------------------------------------------------------------------------------------
2023-10-11 09:19:43,245 epoch 10 - iter 13/136 - loss 0.16390125 - time (sec): 9.08 - samples/sec: 564.67 - lr: 0.000016 - momentum: 0.000000
2023-10-11 09:19:51,864 epoch 10 - iter 26/136 - loss 0.16014339 - time (sec): 17.70 - samples/sec: 552.92 - lr: 0.000014 - momentum: 0.000000
2023-10-11 09:20:00,312 epoch 10 - iter 39/136 - loss 0.15538690 - time (sec): 26.15 - samples/sec: 553.66 - lr: 0.000013 - momentum: 0.000000
2023-10-11 09:20:08,630 epoch 10 - iter 52/136 - loss 0.15425086 - time (sec): 34.47 - samples/sec: 549.87 - lr: 0.000011 - momentum: 0.000000
2023-10-11 09:20:17,912 epoch 10 - iter 65/136 - loss 0.15381241 - time (sec): 43.75 - samples/sec: 565.00 - lr: 0.000009 - momentum: 0.000000
2023-10-11 09:20:26,756 epoch 10 - iter 78/136 - loss 0.14518053 - time (sec): 52.60 - samples/sec: 567.82 - lr: 0.000008 - momentum: 0.000000
2023-10-11 09:20:35,335 epoch 10 - iter 91/136 - loss 0.14825107 - time (sec): 61.17 - samples/sec: 566.97 - lr: 0.000006 - momentum: 0.000000
2023-10-11 09:20:44,119 epoch 10 - iter 104/136 - loss 0.14892497 - time (sec): 69.96 - samples/sec: 566.39 - lr: 0.000005 - momentum: 0.000000
2023-10-11 09:20:52,990 epoch 10 - iter 117/136 - loss 0.14425945 - time (sec): 78.83 - samples/sec: 568.35 - lr: 0.000003 - momentum: 0.000000
2023-10-11 09:21:01,634 epoch 10 - iter 130/136 - loss 0.14459632 - time (sec): 87.47 - samples/sec: 567.41 - lr: 0.000001 - momentum: 0.000000
2023-10-11 09:21:05,521 ----------------------------------------------------------------------------------------------------
2023-10-11 09:21:05,521 EPOCH 10 done: loss 0.1445 - lr: 0.000001
2023-10-11 09:21:11,286 DEV : loss 0.19525763392448425 - f1-score (micro avg) 0.5471
2023-10-11 09:21:11,295 saving best model
2023-10-11 09:21:14,711 ----------------------------------------------------------------------------------------------------
2023-10-11 09:21:14,713 Loading model from best epoch ...
2023-10-11 09:21:18,624 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 09:21:30,954
Results:
- F-score (micro) 0.5107
- F-score (macro) 0.3264
- Accuracy 0.3794
By class:
precision recall f1-score support
LOC 0.5385 0.7179 0.6154 312
PER 0.4023 0.4952 0.4440 208
HumanProd 0.1860 0.3636 0.2462 22
ORG 0.0000 0.0000 0.0000 55
micro avg 0.4685 0.5611 0.5107 597
macro avg 0.2817 0.3942 0.3264 597
weighted avg 0.4284 0.5611 0.4854 597
2023-10-11 09:21:30,954 ----------------------------------------------------------------------------------------------------
|