File size: 23,939 Bytes
e096282 |
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 |
2023-10-17 15:33:07,867 ----------------------------------------------------------------------------------------------------
2023-10-17 15:33:07,868 Model: "SequenceTagger(
(embeddings): TransformerWordEmbeddings(
(model): ElectraModel(
(embeddings): ElectraEmbeddings(
(word_embeddings): Embedding(32001, 768)
(position_embeddings): Embedding(512, 768)
(token_type_embeddings): Embedding(2, 768)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(encoder): ElectraEncoder(
(layer): ModuleList(
(0-11): 12 x ElectraLayer(
(attention): ElectraAttention(
(self): ElectraSelfAttention(
(query): Linear(in_features=768, out_features=768, bias=True)
(key): Linear(in_features=768, out_features=768, bias=True)
(value): Linear(in_features=768, out_features=768, bias=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(output): ElectraSelfOutput(
(dense): Linear(in_features=768, out_features=768, bias=True)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
(intermediate): ElectraIntermediate(
(dense): Linear(in_features=768, out_features=3072, bias=True)
(intermediate_act_fn): GELUActivation()
)
(output): ElectraOutput(
(dense): Linear(in_features=3072, out_features=768, bias=True)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
)
)
)
(locked_dropout): LockedDropout(p=0.5)
(linear): Linear(in_features=768, out_features=13, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-17 15:33:07,868 ----------------------------------------------------------------------------------------------------
2023-10-17 15:33:07,868 MultiCorpus: 5777 train + 722 dev + 723 test sentences
- NER_ICDAR_EUROPEANA Corpus: 5777 train + 722 dev + 723 test sentences - /root/.flair/datasets/ner_icdar_europeana/nl
2023-10-17 15:33:07,868 ----------------------------------------------------------------------------------------------------
2023-10-17 15:33:07,868 Train: 5777 sentences
2023-10-17 15:33:07,868 (train_with_dev=False, train_with_test=False)
2023-10-17 15:33:07,868 ----------------------------------------------------------------------------------------------------
2023-10-17 15:33:07,868 Training Params:
2023-10-17 15:33:07,869 - learning_rate: "5e-05"
2023-10-17 15:33:07,869 - mini_batch_size: "4"
2023-10-17 15:33:07,869 - max_epochs: "10"
2023-10-17 15:33:07,869 - shuffle: "True"
2023-10-17 15:33:07,869 ----------------------------------------------------------------------------------------------------
2023-10-17 15:33:07,869 Plugins:
2023-10-17 15:33:07,869 - TensorboardLogger
2023-10-17 15:33:07,869 - LinearScheduler | warmup_fraction: '0.1'
2023-10-17 15:33:07,869 ----------------------------------------------------------------------------------------------------
2023-10-17 15:33:07,869 Final evaluation on model from best epoch (best-model.pt)
2023-10-17 15:33:07,869 - metric: "('micro avg', 'f1-score')"
2023-10-17 15:33:07,869 ----------------------------------------------------------------------------------------------------
2023-10-17 15:33:07,869 Computation:
2023-10-17 15:33:07,869 - compute on device: cuda:0
2023-10-17 15:33:07,869 - embedding storage: none
2023-10-17 15:33:07,869 ----------------------------------------------------------------------------------------------------
2023-10-17 15:33:07,869 Model training base path: "hmbench-icdar/nl-hmteams/teams-base-historic-multilingual-discriminator-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-1"
2023-10-17 15:33:07,869 ----------------------------------------------------------------------------------------------------
2023-10-17 15:33:07,869 ----------------------------------------------------------------------------------------------------
2023-10-17 15:33:07,869 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-17 15:33:14,818 epoch 1 - iter 144/1445 - loss 1.92593897 - time (sec): 6.95 - samples/sec: 2670.87 - lr: 0.000005 - momentum: 0.000000
2023-10-17 15:33:21,656 epoch 1 - iter 288/1445 - loss 1.14848659 - time (sec): 13.79 - samples/sec: 2498.28 - lr: 0.000010 - momentum: 0.000000
2023-10-17 15:33:28,548 epoch 1 - iter 432/1445 - loss 0.81417525 - time (sec): 20.68 - samples/sec: 2519.75 - lr: 0.000015 - momentum: 0.000000
2023-10-17 15:33:35,383 epoch 1 - iter 576/1445 - loss 0.65607071 - time (sec): 27.51 - samples/sec: 2481.77 - lr: 0.000020 - momentum: 0.000000
2023-10-17 15:33:42,441 epoch 1 - iter 720/1445 - loss 0.55106262 - time (sec): 34.57 - samples/sec: 2502.18 - lr: 0.000025 - momentum: 0.000000
2023-10-17 15:33:49,471 epoch 1 - iter 864/1445 - loss 0.47665927 - time (sec): 41.60 - samples/sec: 2536.30 - lr: 0.000030 - momentum: 0.000000
2023-10-17 15:33:56,060 epoch 1 - iter 1008/1445 - loss 0.42564629 - time (sec): 48.19 - samples/sec: 2555.27 - lr: 0.000035 - momentum: 0.000000
2023-10-17 15:34:02,950 epoch 1 - iter 1152/1445 - loss 0.38691122 - time (sec): 55.08 - samples/sec: 2562.18 - lr: 0.000040 - momentum: 0.000000
2023-10-17 15:34:09,702 epoch 1 - iter 1296/1445 - loss 0.35796946 - time (sec): 61.83 - samples/sec: 2563.72 - lr: 0.000045 - momentum: 0.000000
2023-10-17 15:34:16,350 epoch 1 - iter 1440/1445 - loss 0.33514967 - time (sec): 68.48 - samples/sec: 2565.26 - lr: 0.000050 - momentum: 0.000000
2023-10-17 15:34:16,569 ----------------------------------------------------------------------------------------------------
2023-10-17 15:34:16,569 EPOCH 1 done: loss 0.3344 - lr: 0.000050
2023-10-17 15:34:19,829 DEV : loss 0.10861274600028992 - f1-score (micro avg) 0.6573
2023-10-17 15:34:19,861 saving best model
2023-10-17 15:34:20,273 ----------------------------------------------------------------------------------------------------
2023-10-17 15:34:27,269 epoch 2 - iter 144/1445 - loss 0.11625048 - time (sec): 6.99 - samples/sec: 2379.25 - lr: 0.000049 - momentum: 0.000000
2023-10-17 15:34:34,119 epoch 2 - iter 288/1445 - loss 0.11444173 - time (sec): 13.84 - samples/sec: 2451.31 - lr: 0.000049 - momentum: 0.000000
2023-10-17 15:34:41,181 epoch 2 - iter 432/1445 - loss 0.10811992 - time (sec): 20.91 - samples/sec: 2455.20 - lr: 0.000048 - momentum: 0.000000
2023-10-17 15:34:48,042 epoch 2 - iter 576/1445 - loss 0.10325462 - time (sec): 27.77 - samples/sec: 2465.58 - lr: 0.000048 - momentum: 0.000000
2023-10-17 15:34:55,055 epoch 2 - iter 720/1445 - loss 0.09805219 - time (sec): 34.78 - samples/sec: 2495.12 - lr: 0.000047 - momentum: 0.000000
2023-10-17 15:35:02,213 epoch 2 - iter 864/1445 - loss 0.09411125 - time (sec): 41.94 - samples/sec: 2524.85 - lr: 0.000047 - momentum: 0.000000
2023-10-17 15:35:08,997 epoch 2 - iter 1008/1445 - loss 0.09339206 - time (sec): 48.72 - samples/sec: 2518.97 - lr: 0.000046 - momentum: 0.000000
2023-10-17 15:35:16,245 epoch 2 - iter 1152/1445 - loss 0.09392903 - time (sec): 55.97 - samples/sec: 2509.82 - lr: 0.000046 - momentum: 0.000000
2023-10-17 15:35:23,150 epoch 2 - iter 1296/1445 - loss 0.09494132 - time (sec): 62.87 - samples/sec: 2504.18 - lr: 0.000045 - momentum: 0.000000
2023-10-17 15:35:30,230 epoch 2 - iter 1440/1445 - loss 0.09822877 - time (sec): 69.95 - samples/sec: 2512.31 - lr: 0.000044 - momentum: 0.000000
2023-10-17 15:35:30,450 ----------------------------------------------------------------------------------------------------
2023-10-17 15:35:30,450 EPOCH 2 done: loss 0.0984 - lr: 0.000044
2023-10-17 15:35:34,251 DEV : loss 0.13345900177955627 - f1-score (micro avg) 0.6763
2023-10-17 15:35:34,270 saving best model
2023-10-17 15:35:34,739 ----------------------------------------------------------------------------------------------------
2023-10-17 15:35:41,733 epoch 3 - iter 144/1445 - loss 0.06944847 - time (sec): 6.99 - samples/sec: 2484.40 - lr: 0.000044 - momentum: 0.000000
2023-10-17 15:35:48,620 epoch 3 - iter 288/1445 - loss 0.06628234 - time (sec): 13.88 - samples/sec: 2494.99 - lr: 0.000043 - momentum: 0.000000
2023-10-17 15:35:55,938 epoch 3 - iter 432/1445 - loss 0.06588793 - time (sec): 21.20 - samples/sec: 2528.44 - lr: 0.000043 - momentum: 0.000000
2023-10-17 15:36:02,862 epoch 3 - iter 576/1445 - loss 0.06519428 - time (sec): 28.12 - samples/sec: 2515.02 - lr: 0.000042 - momentum: 0.000000
2023-10-17 15:36:09,625 epoch 3 - iter 720/1445 - loss 0.06602713 - time (sec): 34.88 - samples/sec: 2507.94 - lr: 0.000042 - momentum: 0.000000
2023-10-17 15:36:16,739 epoch 3 - iter 864/1445 - loss 0.06637663 - time (sec): 42.00 - samples/sec: 2507.44 - lr: 0.000041 - momentum: 0.000000
2023-10-17 15:36:23,756 epoch 3 - iter 1008/1445 - loss 0.06655629 - time (sec): 49.02 - samples/sec: 2489.67 - lr: 0.000041 - momentum: 0.000000
2023-10-17 15:36:30,816 epoch 3 - iter 1152/1445 - loss 0.06748428 - time (sec): 56.08 - samples/sec: 2487.24 - lr: 0.000040 - momentum: 0.000000
2023-10-17 15:36:38,008 epoch 3 - iter 1296/1445 - loss 0.06799725 - time (sec): 63.27 - samples/sec: 2491.50 - lr: 0.000039 - momentum: 0.000000
2023-10-17 15:36:44,928 epoch 3 - iter 1440/1445 - loss 0.06778051 - time (sec): 70.19 - samples/sec: 2504.64 - lr: 0.000039 - momentum: 0.000000
2023-10-17 15:36:45,142 ----------------------------------------------------------------------------------------------------
2023-10-17 15:36:45,143 EPOCH 3 done: loss 0.0681 - lr: 0.000039
2023-10-17 15:36:48,337 DEV : loss 0.09013378620147705 - f1-score (micro avg) 0.8588
2023-10-17 15:36:48,354 saving best model
2023-10-17 15:36:48,816 ----------------------------------------------------------------------------------------------------
2023-10-17 15:36:55,898 epoch 4 - iter 144/1445 - loss 0.04676885 - time (sec): 7.08 - samples/sec: 2579.33 - lr: 0.000038 - momentum: 0.000000
2023-10-17 15:37:02,866 epoch 4 - iter 288/1445 - loss 0.05260260 - time (sec): 14.04 - samples/sec: 2532.21 - lr: 0.000038 - momentum: 0.000000
2023-10-17 15:37:09,745 epoch 4 - iter 432/1445 - loss 0.04637383 - time (sec): 20.92 - samples/sec: 2529.30 - lr: 0.000037 - momentum: 0.000000
2023-10-17 15:37:16,682 epoch 4 - iter 576/1445 - loss 0.04944995 - time (sec): 27.86 - samples/sec: 2523.60 - lr: 0.000037 - momentum: 0.000000
2023-10-17 15:37:23,487 epoch 4 - iter 720/1445 - loss 0.05010050 - time (sec): 34.66 - samples/sec: 2507.18 - lr: 0.000036 - momentum: 0.000000
2023-10-17 15:37:30,605 epoch 4 - iter 864/1445 - loss 0.05081688 - time (sec): 41.78 - samples/sec: 2501.61 - lr: 0.000036 - momentum: 0.000000
2023-10-17 15:37:37,605 epoch 4 - iter 1008/1445 - loss 0.05019035 - time (sec): 48.78 - samples/sec: 2507.35 - lr: 0.000035 - momentum: 0.000000
2023-10-17 15:37:44,603 epoch 4 - iter 1152/1445 - loss 0.04964440 - time (sec): 55.78 - samples/sec: 2508.06 - lr: 0.000034 - momentum: 0.000000
2023-10-17 15:37:51,644 epoch 4 - iter 1296/1445 - loss 0.04892615 - time (sec): 62.82 - samples/sec: 2509.01 - lr: 0.000034 - momentum: 0.000000
2023-10-17 15:37:58,723 epoch 4 - iter 1440/1445 - loss 0.05060487 - time (sec): 69.90 - samples/sec: 2514.90 - lr: 0.000033 - momentum: 0.000000
2023-10-17 15:37:58,946 ----------------------------------------------------------------------------------------------------
2023-10-17 15:37:58,946 EPOCH 4 done: loss 0.0505 - lr: 0.000033
2023-10-17 15:38:02,111 DEV : loss 0.09323444962501526 - f1-score (micro avg) 0.8574
2023-10-17 15:38:02,127 ----------------------------------------------------------------------------------------------------
2023-10-17 15:38:09,095 epoch 5 - iter 144/1445 - loss 0.03072807 - time (sec): 6.97 - samples/sec: 2539.84 - lr: 0.000033 - momentum: 0.000000
2023-10-17 15:38:16,057 epoch 5 - iter 288/1445 - loss 0.03358255 - time (sec): 13.93 - samples/sec: 2580.24 - lr: 0.000032 - momentum: 0.000000
2023-10-17 15:38:23,517 epoch 5 - iter 432/1445 - loss 0.03446332 - time (sec): 21.39 - samples/sec: 2512.59 - lr: 0.000032 - momentum: 0.000000
2023-10-17 15:38:30,301 epoch 5 - iter 576/1445 - loss 0.03497844 - time (sec): 28.17 - samples/sec: 2501.32 - lr: 0.000031 - momentum: 0.000000
2023-10-17 15:38:37,245 epoch 5 - iter 720/1445 - loss 0.03494447 - time (sec): 35.12 - samples/sec: 2500.49 - lr: 0.000031 - momentum: 0.000000
2023-10-17 15:38:44,177 epoch 5 - iter 864/1445 - loss 0.03849124 - time (sec): 42.05 - samples/sec: 2494.59 - lr: 0.000030 - momentum: 0.000000
2023-10-17 15:38:51,092 epoch 5 - iter 1008/1445 - loss 0.03868840 - time (sec): 48.96 - samples/sec: 2492.81 - lr: 0.000029 - momentum: 0.000000
2023-10-17 15:38:58,128 epoch 5 - iter 1152/1445 - loss 0.03863802 - time (sec): 56.00 - samples/sec: 2510.90 - lr: 0.000029 - momentum: 0.000000
2023-10-17 15:39:04,931 epoch 5 - iter 1296/1445 - loss 0.03989823 - time (sec): 62.80 - samples/sec: 2513.11 - lr: 0.000028 - momentum: 0.000000
2023-10-17 15:39:12,028 epoch 5 - iter 1440/1445 - loss 0.03957638 - time (sec): 69.90 - samples/sec: 2512.94 - lr: 0.000028 - momentum: 0.000000
2023-10-17 15:39:12,247 ----------------------------------------------------------------------------------------------------
2023-10-17 15:39:12,248 EPOCH 5 done: loss 0.0395 - lr: 0.000028
2023-10-17 15:39:15,509 DEV : loss 0.11628924310207367 - f1-score (micro avg) 0.8495
2023-10-17 15:39:15,527 ----------------------------------------------------------------------------------------------------
2023-10-17 15:39:22,621 epoch 6 - iter 144/1445 - loss 0.02447880 - time (sec): 7.09 - samples/sec: 2467.72 - lr: 0.000027 - momentum: 0.000000
2023-10-17 15:39:29,423 epoch 6 - iter 288/1445 - loss 0.03137841 - time (sec): 13.89 - samples/sec: 2449.60 - lr: 0.000027 - momentum: 0.000000
2023-10-17 15:39:36,131 epoch 6 - iter 432/1445 - loss 0.03035696 - time (sec): 20.60 - samples/sec: 2504.30 - lr: 0.000026 - momentum: 0.000000
2023-10-17 15:39:43,478 epoch 6 - iter 576/1445 - loss 0.02805603 - time (sec): 27.95 - samples/sec: 2497.07 - lr: 0.000026 - momentum: 0.000000
2023-10-17 15:39:50,654 epoch 6 - iter 720/1445 - loss 0.02854656 - time (sec): 35.13 - samples/sec: 2505.16 - lr: 0.000025 - momentum: 0.000000
2023-10-17 15:39:57,367 epoch 6 - iter 864/1445 - loss 0.02796582 - time (sec): 41.84 - samples/sec: 2496.03 - lr: 0.000024 - momentum: 0.000000
2023-10-17 15:40:04,170 epoch 6 - iter 1008/1445 - loss 0.02775939 - time (sec): 48.64 - samples/sec: 2519.64 - lr: 0.000024 - momentum: 0.000000
2023-10-17 15:40:11,115 epoch 6 - iter 1152/1445 - loss 0.02855916 - time (sec): 55.59 - samples/sec: 2501.88 - lr: 0.000023 - momentum: 0.000000
2023-10-17 15:40:18,248 epoch 6 - iter 1296/1445 - loss 0.02841167 - time (sec): 62.72 - samples/sec: 2499.15 - lr: 0.000023 - momentum: 0.000000
2023-10-17 15:40:25,387 epoch 6 - iter 1440/1445 - loss 0.02790215 - time (sec): 69.86 - samples/sec: 2512.16 - lr: 0.000022 - momentum: 0.000000
2023-10-17 15:40:25,612 ----------------------------------------------------------------------------------------------------
2023-10-17 15:40:25,612 EPOCH 6 done: loss 0.0278 - lr: 0.000022
2023-10-17 15:40:29,051 DEV : loss 0.13435053825378418 - f1-score (micro avg) 0.8472
2023-10-17 15:40:29,073 ----------------------------------------------------------------------------------------------------
2023-10-17 15:40:36,627 epoch 7 - iter 144/1445 - loss 0.03621832 - time (sec): 7.55 - samples/sec: 2286.98 - lr: 0.000022 - momentum: 0.000000
2023-10-17 15:40:43,645 epoch 7 - iter 288/1445 - loss 0.02724953 - time (sec): 14.57 - samples/sec: 2335.39 - lr: 0.000021 - momentum: 0.000000
2023-10-17 15:40:50,674 epoch 7 - iter 432/1445 - loss 0.02754718 - time (sec): 21.60 - samples/sec: 2405.25 - lr: 0.000021 - momentum: 0.000000
2023-10-17 15:40:57,976 epoch 7 - iter 576/1445 - loss 0.02528310 - time (sec): 28.90 - samples/sec: 2422.96 - lr: 0.000020 - momentum: 0.000000
2023-10-17 15:41:04,930 epoch 7 - iter 720/1445 - loss 0.02377881 - time (sec): 35.86 - samples/sec: 2435.71 - lr: 0.000019 - momentum: 0.000000
2023-10-17 15:41:11,853 epoch 7 - iter 864/1445 - loss 0.02330516 - time (sec): 42.78 - samples/sec: 2478.15 - lr: 0.000019 - momentum: 0.000000
2023-10-17 15:41:18,745 epoch 7 - iter 1008/1445 - loss 0.02094244 - time (sec): 49.67 - samples/sec: 2479.00 - lr: 0.000018 - momentum: 0.000000
2023-10-17 15:41:25,574 epoch 7 - iter 1152/1445 - loss 0.01946340 - time (sec): 56.50 - samples/sec: 2481.85 - lr: 0.000018 - momentum: 0.000000
2023-10-17 15:41:33,047 epoch 7 - iter 1296/1445 - loss 0.01934243 - time (sec): 63.97 - samples/sec: 2471.24 - lr: 0.000017 - momentum: 0.000000
2023-10-17 15:41:39,929 epoch 7 - iter 1440/1445 - loss 0.01931728 - time (sec): 70.85 - samples/sec: 2480.95 - lr: 0.000017 - momentum: 0.000000
2023-10-17 15:41:40,152 ----------------------------------------------------------------------------------------------------
2023-10-17 15:41:40,153 EPOCH 7 done: loss 0.0193 - lr: 0.000017
2023-10-17 15:41:43,359 DEV : loss 0.13610993325710297 - f1-score (micro avg) 0.8496
2023-10-17 15:41:43,376 ----------------------------------------------------------------------------------------------------
2023-10-17 15:41:50,179 epoch 8 - iter 144/1445 - loss 0.01156269 - time (sec): 6.80 - samples/sec: 2373.92 - lr: 0.000016 - momentum: 0.000000
2023-10-17 15:41:57,334 epoch 8 - iter 288/1445 - loss 0.01424879 - time (sec): 13.96 - samples/sec: 2475.60 - lr: 0.000016 - momentum: 0.000000
2023-10-17 15:42:04,154 epoch 8 - iter 432/1445 - loss 0.01152694 - time (sec): 20.78 - samples/sec: 2517.25 - lr: 0.000015 - momentum: 0.000000
2023-10-17 15:42:11,076 epoch 8 - iter 576/1445 - loss 0.01252404 - time (sec): 27.70 - samples/sec: 2492.28 - lr: 0.000014 - momentum: 0.000000
2023-10-17 15:42:18,115 epoch 8 - iter 720/1445 - loss 0.01307924 - time (sec): 34.74 - samples/sec: 2478.52 - lr: 0.000014 - momentum: 0.000000
2023-10-17 15:42:25,240 epoch 8 - iter 864/1445 - loss 0.01191157 - time (sec): 41.86 - samples/sec: 2498.03 - lr: 0.000013 - momentum: 0.000000
2023-10-17 15:42:31,873 epoch 8 - iter 1008/1445 - loss 0.01386504 - time (sec): 48.50 - samples/sec: 2523.71 - lr: 0.000013 - momentum: 0.000000
2023-10-17 15:42:38,759 epoch 8 - iter 1152/1445 - loss 0.01425968 - time (sec): 55.38 - samples/sec: 2514.92 - lr: 0.000012 - momentum: 0.000000
2023-10-17 15:42:45,754 epoch 8 - iter 1296/1445 - loss 0.01357515 - time (sec): 62.38 - samples/sec: 2530.49 - lr: 0.000012 - momentum: 0.000000
2023-10-17 15:42:52,599 epoch 8 - iter 1440/1445 - loss 0.01409541 - time (sec): 69.22 - samples/sec: 2534.97 - lr: 0.000011 - momentum: 0.000000
2023-10-17 15:42:52,844 ----------------------------------------------------------------------------------------------------
2023-10-17 15:42:52,844 EPOCH 8 done: loss 0.0140 - lr: 0.000011
2023-10-17 15:42:56,085 DEV : loss 0.16182565689086914 - f1-score (micro avg) 0.8473
2023-10-17 15:42:56,101 ----------------------------------------------------------------------------------------------------
2023-10-17 15:43:03,053 epoch 9 - iter 144/1445 - loss 0.01038750 - time (sec): 6.95 - samples/sec: 2763.07 - lr: 0.000011 - momentum: 0.000000
2023-10-17 15:43:09,940 epoch 9 - iter 288/1445 - loss 0.00853687 - time (sec): 13.84 - samples/sec: 2536.32 - lr: 0.000010 - momentum: 0.000000
2023-10-17 15:43:17,344 epoch 9 - iter 432/1445 - loss 0.00758837 - time (sec): 21.24 - samples/sec: 2549.74 - lr: 0.000009 - momentum: 0.000000
2023-10-17 15:43:24,505 epoch 9 - iter 576/1445 - loss 0.00778959 - time (sec): 28.40 - samples/sec: 2550.27 - lr: 0.000009 - momentum: 0.000000
2023-10-17 15:43:31,649 epoch 9 - iter 720/1445 - loss 0.00866403 - time (sec): 35.55 - samples/sec: 2519.68 - lr: 0.000008 - momentum: 0.000000
2023-10-17 15:43:38,340 epoch 9 - iter 864/1445 - loss 0.00829698 - time (sec): 42.24 - samples/sec: 2498.92 - lr: 0.000008 - momentum: 0.000000
2023-10-17 15:43:45,253 epoch 9 - iter 1008/1445 - loss 0.00824515 - time (sec): 49.15 - samples/sec: 2506.38 - lr: 0.000007 - momentum: 0.000000
2023-10-17 15:43:52,687 epoch 9 - iter 1152/1445 - loss 0.00842807 - time (sec): 56.58 - samples/sec: 2497.57 - lr: 0.000007 - momentum: 0.000000
2023-10-17 15:43:59,636 epoch 9 - iter 1296/1445 - loss 0.00853763 - time (sec): 63.53 - samples/sec: 2501.91 - lr: 0.000006 - momentum: 0.000000
2023-10-17 15:44:06,350 epoch 9 - iter 1440/1445 - loss 0.00875377 - time (sec): 70.25 - samples/sec: 2497.52 - lr: 0.000006 - momentum: 0.000000
2023-10-17 15:44:06,618 ----------------------------------------------------------------------------------------------------
2023-10-17 15:44:06,619 EPOCH 9 done: loss 0.0087 - lr: 0.000006
2023-10-17 15:44:10,219 DEV : loss 0.14013005793094635 - f1-score (micro avg) 0.8622
2023-10-17 15:44:10,236 saving best model
2023-10-17 15:44:10,722 ----------------------------------------------------------------------------------------------------
2023-10-17 15:44:17,455 epoch 10 - iter 144/1445 - loss 0.00579021 - time (sec): 6.73 - samples/sec: 2592.22 - lr: 0.000005 - momentum: 0.000000
2023-10-17 15:44:24,391 epoch 10 - iter 288/1445 - loss 0.00521433 - time (sec): 13.67 - samples/sec: 2564.98 - lr: 0.000004 - momentum: 0.000000
2023-10-17 15:44:31,282 epoch 10 - iter 432/1445 - loss 0.00599340 - time (sec): 20.56 - samples/sec: 2514.52 - lr: 0.000004 - momentum: 0.000000
2023-10-17 15:44:38,028 epoch 10 - iter 576/1445 - loss 0.00550622 - time (sec): 27.30 - samples/sec: 2488.42 - lr: 0.000003 - momentum: 0.000000
2023-10-17 15:44:45,428 epoch 10 - iter 720/1445 - loss 0.00533641 - time (sec): 34.70 - samples/sec: 2497.69 - lr: 0.000003 - momentum: 0.000000
2023-10-17 15:44:52,515 epoch 10 - iter 864/1445 - loss 0.00570968 - time (sec): 41.79 - samples/sec: 2521.83 - lr: 0.000002 - momentum: 0.000000
2023-10-17 15:44:59,374 epoch 10 - iter 1008/1445 - loss 0.00509566 - time (sec): 48.65 - samples/sec: 2508.54 - lr: 0.000002 - momentum: 0.000000
2023-10-17 15:45:06,472 epoch 10 - iter 1152/1445 - loss 0.00510136 - time (sec): 55.75 - samples/sec: 2513.10 - lr: 0.000001 - momentum: 0.000000
2023-10-17 15:45:13,388 epoch 10 - iter 1296/1445 - loss 0.00492028 - time (sec): 62.66 - samples/sec: 2523.18 - lr: 0.000001 - momentum: 0.000000
2023-10-17 15:45:20,396 epoch 10 - iter 1440/1445 - loss 0.00498631 - time (sec): 69.67 - samples/sec: 2522.46 - lr: 0.000000 - momentum: 0.000000
2023-10-17 15:45:20,612 ----------------------------------------------------------------------------------------------------
2023-10-17 15:45:20,612 EPOCH 10 done: loss 0.0050 - lr: 0.000000
2023-10-17 15:45:23,866 DEV : loss 0.16260549426078796 - f1-score (micro avg) 0.8548
2023-10-17 15:45:24,223 ----------------------------------------------------------------------------------------------------
2023-10-17 15:45:24,225 Loading model from best epoch ...
2023-10-17 15:45:25,583 SequenceTagger predicts: Dictionary with 13 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
2023-10-17 15:45:28,550
Results:
- F-score (micro) 0.8439
- F-score (macro) 0.7461
- Accuracy 0.7378
By class:
precision recall f1-score support
PER 0.8556 0.8361 0.8458 482
LOC 0.9392 0.8428 0.8884 458
ORG 0.6000 0.4348 0.5042 69
micro avg 0.8788 0.8117 0.8439 1009
macro avg 0.7983 0.7046 0.7461 1009
weighted avg 0.8761 0.8117 0.8417 1009
2023-10-17 15:45:28,551 ----------------------------------------------------------------------------------------------------
|