File size: 23,968 Bytes
83ae4ac |
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 |
2023-10-14 10:30:31,688 ----------------------------------------------------------------------------------------------------
2023-10-14 10:30:31,689 Model: "SequenceTagger(
(embeddings): TransformerWordEmbeddings(
(model): BertModel(
(embeddings): BertEmbeddings(
(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): BertEncoder(
(layer): ModuleList(
(0-11): 12 x BertLayer(
(attention): BertAttention(
(self): BertSelfAttention(
(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): BertSelfOutput(
(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): BertIntermediate(
(dense): Linear(in_features=768, out_features=3072, bias=True)
(intermediate_act_fn): GELUActivation()
)
(output): BertOutput(
(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)
)
)
)
)
(pooler): BertPooler(
(dense): Linear(in_features=768, out_features=768, bias=True)
(activation): Tanh()
)
)
)
(locked_dropout): LockedDropout(p=0.5)
(linear): Linear(in_features=768, out_features=13, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-14 10:30:31,690 ----------------------------------------------------------------------------------------------------
2023-10-14 10:30:31,690 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
2023-10-14 10:30:31,690 ----------------------------------------------------------------------------------------------------
2023-10-14 10:30:31,690 Train: 5777 sentences
2023-10-14 10:30:31,690 (train_with_dev=False, train_with_test=False)
2023-10-14 10:30:31,690 ----------------------------------------------------------------------------------------------------
2023-10-14 10:30:31,690 Training Params:
2023-10-14 10:30:31,690 - learning_rate: "3e-05"
2023-10-14 10:30:31,690 - mini_batch_size: "4"
2023-10-14 10:30:31,690 - max_epochs: "10"
2023-10-14 10:30:31,690 - shuffle: "True"
2023-10-14 10:30:31,690 ----------------------------------------------------------------------------------------------------
2023-10-14 10:30:31,690 Plugins:
2023-10-14 10:30:31,690 - LinearScheduler | warmup_fraction: '0.1'
2023-10-14 10:30:31,690 ----------------------------------------------------------------------------------------------------
2023-10-14 10:30:31,690 Final evaluation on model from best epoch (best-model.pt)
2023-10-14 10:30:31,690 - metric: "('micro avg', 'f1-score')"
2023-10-14 10:30:31,690 ----------------------------------------------------------------------------------------------------
2023-10-14 10:30:31,690 Computation:
2023-10-14 10:30:31,690 - compute on device: cuda:0
2023-10-14 10:30:31,690 - embedding storage: none
2023-10-14 10:30:31,690 ----------------------------------------------------------------------------------------------------
2023-10-14 10:30:31,691 Model training base path: "hmbench-icdar/nl-dbmdz/bert-base-historic-multilingual-cased-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-4"
2023-10-14 10:30:31,691 ----------------------------------------------------------------------------------------------------
2023-10-14 10:30:31,691 ----------------------------------------------------------------------------------------------------
2023-10-14 10:30:39,193 epoch 1 - iter 144/1445 - loss 1.87631987 - time (sec): 7.50 - samples/sec: 2486.02 - lr: 0.000003 - momentum: 0.000000
2023-10-14 10:30:46,429 epoch 1 - iter 288/1445 - loss 1.10871491 - time (sec): 14.74 - samples/sec: 2450.97 - lr: 0.000006 - momentum: 0.000000
2023-10-14 10:30:53,808 epoch 1 - iter 432/1445 - loss 0.81766779 - time (sec): 22.12 - samples/sec: 2409.53 - lr: 0.000009 - momentum: 0.000000
2023-10-14 10:31:00,941 epoch 1 - iter 576/1445 - loss 0.66179866 - time (sec): 29.25 - samples/sec: 2399.89 - lr: 0.000012 - momentum: 0.000000
2023-10-14 10:31:08,229 epoch 1 - iter 720/1445 - loss 0.56148096 - time (sec): 36.54 - samples/sec: 2416.30 - lr: 0.000015 - momentum: 0.000000
2023-10-14 10:31:15,427 epoch 1 - iter 864/1445 - loss 0.49511789 - time (sec): 43.74 - samples/sec: 2425.95 - lr: 0.000018 - momentum: 0.000000
2023-10-14 10:31:22,615 epoch 1 - iter 1008/1445 - loss 0.44567392 - time (sec): 50.92 - samples/sec: 2427.43 - lr: 0.000021 - momentum: 0.000000
2023-10-14 10:31:30,224 epoch 1 - iter 1152/1445 - loss 0.40668399 - time (sec): 58.53 - samples/sec: 2436.80 - lr: 0.000024 - momentum: 0.000000
2023-10-14 10:31:37,479 epoch 1 - iter 1296/1445 - loss 0.37622430 - time (sec): 65.79 - samples/sec: 2431.07 - lr: 0.000027 - momentum: 0.000000
2023-10-14 10:31:44,262 epoch 1 - iter 1440/1445 - loss 0.35511279 - time (sec): 72.57 - samples/sec: 2420.51 - lr: 0.000030 - momentum: 0.000000
2023-10-14 10:31:44,497 ----------------------------------------------------------------------------------------------------
2023-10-14 10:31:44,497 EPOCH 1 done: loss 0.3543 - lr: 0.000030
2023-10-14 10:31:47,491 DEV : loss 0.12729743123054504 - f1-score (micro avg) 0.656
2023-10-14 10:31:47,518 saving best model
2023-10-14 10:31:47,917 ----------------------------------------------------------------------------------------------------
2023-10-14 10:31:55,504 epoch 2 - iter 144/1445 - loss 0.12159020 - time (sec): 7.59 - samples/sec: 2138.39 - lr: 0.000030 - momentum: 0.000000
2023-10-14 10:32:02,751 epoch 2 - iter 288/1445 - loss 0.11080145 - time (sec): 14.83 - samples/sec: 2278.16 - lr: 0.000029 - momentum: 0.000000
2023-10-14 10:32:10,121 epoch 2 - iter 432/1445 - loss 0.11375360 - time (sec): 22.20 - samples/sec: 2331.27 - lr: 0.000029 - momentum: 0.000000
2023-10-14 10:32:17,758 epoch 2 - iter 576/1445 - loss 0.10963141 - time (sec): 29.84 - samples/sec: 2375.48 - lr: 0.000029 - momentum: 0.000000
2023-10-14 10:32:25,047 epoch 2 - iter 720/1445 - loss 0.10826773 - time (sec): 37.13 - samples/sec: 2392.41 - lr: 0.000028 - momentum: 0.000000
2023-10-14 10:32:32,192 epoch 2 - iter 864/1445 - loss 0.10632474 - time (sec): 44.27 - samples/sec: 2384.95 - lr: 0.000028 - momentum: 0.000000
2023-10-14 10:32:39,186 epoch 2 - iter 1008/1445 - loss 0.10768588 - time (sec): 51.27 - samples/sec: 2382.07 - lr: 0.000028 - momentum: 0.000000
2023-10-14 10:32:46,323 epoch 2 - iter 1152/1445 - loss 0.10485484 - time (sec): 58.40 - samples/sec: 2387.46 - lr: 0.000027 - momentum: 0.000000
2023-10-14 10:32:53,665 epoch 2 - iter 1296/1445 - loss 0.10330686 - time (sec): 65.75 - samples/sec: 2391.73 - lr: 0.000027 - momentum: 0.000000
2023-10-14 10:33:01,061 epoch 2 - iter 1440/1445 - loss 0.10268253 - time (sec): 73.14 - samples/sec: 2402.14 - lr: 0.000027 - momentum: 0.000000
2023-10-14 10:33:01,320 ----------------------------------------------------------------------------------------------------
2023-10-14 10:33:01,321 EPOCH 2 done: loss 0.1025 - lr: 0.000027
2023-10-14 10:33:04,810 DEV : loss 0.1065981537103653 - f1-score (micro avg) 0.7711
2023-10-14 10:33:04,827 saving best model
2023-10-14 10:33:05,296 ----------------------------------------------------------------------------------------------------
2023-10-14 10:33:12,939 epoch 3 - iter 144/1445 - loss 0.06396374 - time (sec): 7.64 - samples/sec: 2373.16 - lr: 0.000026 - momentum: 0.000000
2023-10-14 10:33:20,190 epoch 3 - iter 288/1445 - loss 0.05986877 - time (sec): 14.89 - samples/sec: 2391.48 - lr: 0.000026 - momentum: 0.000000
2023-10-14 10:33:27,268 epoch 3 - iter 432/1445 - loss 0.06439752 - time (sec): 21.97 - samples/sec: 2363.30 - lr: 0.000026 - momentum: 0.000000
2023-10-14 10:33:34,315 epoch 3 - iter 576/1445 - loss 0.06506263 - time (sec): 29.02 - samples/sec: 2386.42 - lr: 0.000025 - momentum: 0.000000
2023-10-14 10:33:41,658 epoch 3 - iter 720/1445 - loss 0.06609539 - time (sec): 36.36 - samples/sec: 2411.92 - lr: 0.000025 - momentum: 0.000000
2023-10-14 10:33:48,841 epoch 3 - iter 864/1445 - loss 0.06894766 - time (sec): 43.54 - samples/sec: 2414.80 - lr: 0.000025 - momentum: 0.000000
2023-10-14 10:33:56,286 epoch 3 - iter 1008/1445 - loss 0.06961197 - time (sec): 50.99 - samples/sec: 2429.56 - lr: 0.000024 - momentum: 0.000000
2023-10-14 10:34:03,260 epoch 3 - iter 1152/1445 - loss 0.06912057 - time (sec): 57.96 - samples/sec: 2421.10 - lr: 0.000024 - momentum: 0.000000
2023-10-14 10:34:10,460 epoch 3 - iter 1296/1445 - loss 0.06848518 - time (sec): 65.16 - samples/sec: 2416.48 - lr: 0.000024 - momentum: 0.000000
2023-10-14 10:34:17,762 epoch 3 - iter 1440/1445 - loss 0.06940007 - time (sec): 72.46 - samples/sec: 2420.46 - lr: 0.000023 - momentum: 0.000000
2023-10-14 10:34:18,075 ----------------------------------------------------------------------------------------------------
2023-10-14 10:34:18,075 EPOCH 3 done: loss 0.0693 - lr: 0.000023
2023-10-14 10:34:22,064 DEV : loss 0.10008460283279419 - f1-score (micro avg) 0.8091
2023-10-14 10:34:22,083 saving best model
2023-10-14 10:34:22,596 ----------------------------------------------------------------------------------------------------
2023-10-14 10:34:30,131 epoch 4 - iter 144/1445 - loss 0.04151479 - time (sec): 7.53 - samples/sec: 2332.12 - lr: 0.000023 - momentum: 0.000000
2023-10-14 10:34:37,775 epoch 4 - iter 288/1445 - loss 0.05738197 - time (sec): 15.18 - samples/sec: 2362.39 - lr: 0.000023 - momentum: 0.000000
2023-10-14 10:34:44,979 epoch 4 - iter 432/1445 - loss 0.05769172 - time (sec): 22.38 - samples/sec: 2358.88 - lr: 0.000022 - momentum: 0.000000
2023-10-14 10:34:52,330 epoch 4 - iter 576/1445 - loss 0.05226711 - time (sec): 29.73 - samples/sec: 2376.58 - lr: 0.000022 - momentum: 0.000000
2023-10-14 10:34:59,385 epoch 4 - iter 720/1445 - loss 0.05024125 - time (sec): 36.79 - samples/sec: 2369.38 - lr: 0.000022 - momentum: 0.000000
2023-10-14 10:35:06,722 epoch 4 - iter 864/1445 - loss 0.04812138 - time (sec): 44.12 - samples/sec: 2395.51 - lr: 0.000021 - momentum: 0.000000
2023-10-14 10:35:13,970 epoch 4 - iter 1008/1445 - loss 0.04874861 - time (sec): 51.37 - samples/sec: 2389.48 - lr: 0.000021 - momentum: 0.000000
2023-10-14 10:35:21,254 epoch 4 - iter 1152/1445 - loss 0.04918823 - time (sec): 58.66 - samples/sec: 2395.34 - lr: 0.000021 - momentum: 0.000000
2023-10-14 10:35:28,544 epoch 4 - iter 1296/1445 - loss 0.04838955 - time (sec): 65.95 - samples/sec: 2404.60 - lr: 0.000020 - momentum: 0.000000
2023-10-14 10:35:35,780 epoch 4 - iter 1440/1445 - loss 0.04831869 - time (sec): 73.18 - samples/sec: 2399.55 - lr: 0.000020 - momentum: 0.000000
2023-10-14 10:35:36,022 ----------------------------------------------------------------------------------------------------
2023-10-14 10:35:36,023 EPOCH 4 done: loss 0.0484 - lr: 0.000020
2023-10-14 10:35:39,548 DEV : loss 0.11963574588298798 - f1-score (micro avg) 0.8123
2023-10-14 10:35:39,565 saving best model
2023-10-14 10:35:40,073 ----------------------------------------------------------------------------------------------------
2023-10-14 10:35:47,545 epoch 5 - iter 144/1445 - loss 0.04127449 - time (sec): 7.47 - samples/sec: 2465.38 - lr: 0.000020 - momentum: 0.000000
2023-10-14 10:35:54,619 epoch 5 - iter 288/1445 - loss 0.04126193 - time (sec): 14.54 - samples/sec: 2436.83 - lr: 0.000019 - momentum: 0.000000
2023-10-14 10:36:02,056 epoch 5 - iter 432/1445 - loss 0.03693959 - time (sec): 21.98 - samples/sec: 2462.92 - lr: 0.000019 - momentum: 0.000000
2023-10-14 10:36:09,257 epoch 5 - iter 576/1445 - loss 0.03990160 - time (sec): 29.18 - samples/sec: 2427.30 - lr: 0.000019 - momentum: 0.000000
2023-10-14 10:36:16,307 epoch 5 - iter 720/1445 - loss 0.03748764 - time (sec): 36.23 - samples/sec: 2419.21 - lr: 0.000018 - momentum: 0.000000
2023-10-14 10:36:23,621 epoch 5 - iter 864/1445 - loss 0.03765664 - time (sec): 43.54 - samples/sec: 2383.06 - lr: 0.000018 - momentum: 0.000000
2023-10-14 10:36:30,856 epoch 5 - iter 1008/1445 - loss 0.03711077 - time (sec): 50.78 - samples/sec: 2401.78 - lr: 0.000018 - momentum: 0.000000
2023-10-14 10:36:38,122 epoch 5 - iter 1152/1445 - loss 0.03800064 - time (sec): 58.05 - samples/sec: 2407.38 - lr: 0.000017 - momentum: 0.000000
2023-10-14 10:36:45,564 epoch 5 - iter 1296/1445 - loss 0.03872554 - time (sec): 65.49 - samples/sec: 2416.72 - lr: 0.000017 - momentum: 0.000000
2023-10-14 10:36:52,852 epoch 5 - iter 1440/1445 - loss 0.03722040 - time (sec): 72.78 - samples/sec: 2414.26 - lr: 0.000017 - momentum: 0.000000
2023-10-14 10:36:53,079 ----------------------------------------------------------------------------------------------------
2023-10-14 10:36:53,079 EPOCH 5 done: loss 0.0373 - lr: 0.000017
2023-10-14 10:36:56,597 DEV : loss 0.18025827407836914 - f1-score (micro avg) 0.7727
2023-10-14 10:36:56,614 ----------------------------------------------------------------------------------------------------
2023-10-14 10:37:03,761 epoch 6 - iter 144/1445 - loss 0.02549194 - time (sec): 7.15 - samples/sec: 2428.23 - lr: 0.000016 - momentum: 0.000000
2023-10-14 10:37:11,095 epoch 6 - iter 288/1445 - loss 0.02825162 - time (sec): 14.48 - samples/sec: 2422.36 - lr: 0.000016 - momentum: 0.000000
2023-10-14 10:37:18,192 epoch 6 - iter 432/1445 - loss 0.02593624 - time (sec): 21.58 - samples/sec: 2432.82 - lr: 0.000016 - momentum: 0.000000
2023-10-14 10:37:25,601 epoch 6 - iter 576/1445 - loss 0.02696454 - time (sec): 28.99 - samples/sec: 2433.41 - lr: 0.000015 - momentum: 0.000000
2023-10-14 10:37:32,807 epoch 6 - iter 720/1445 - loss 0.02662243 - time (sec): 36.19 - samples/sec: 2415.50 - lr: 0.000015 - momentum: 0.000000
2023-10-14 10:37:40,250 epoch 6 - iter 864/1445 - loss 0.02600658 - time (sec): 43.64 - samples/sec: 2407.28 - lr: 0.000015 - momentum: 0.000000
2023-10-14 10:37:47,675 epoch 6 - iter 1008/1445 - loss 0.02574166 - time (sec): 51.06 - samples/sec: 2406.40 - lr: 0.000014 - momentum: 0.000000
2023-10-14 10:37:55,224 epoch 6 - iter 1152/1445 - loss 0.02422581 - time (sec): 58.61 - samples/sec: 2427.06 - lr: 0.000014 - momentum: 0.000000
2023-10-14 10:38:02,318 epoch 6 - iter 1296/1445 - loss 0.02622967 - time (sec): 65.70 - samples/sec: 2421.76 - lr: 0.000014 - momentum: 0.000000
2023-10-14 10:38:09,361 epoch 6 - iter 1440/1445 - loss 0.02636942 - time (sec): 72.75 - samples/sec: 2416.00 - lr: 0.000013 - momentum: 0.000000
2023-10-14 10:38:09,579 ----------------------------------------------------------------------------------------------------
2023-10-14 10:38:09,580 EPOCH 6 done: loss 0.0264 - lr: 0.000013
2023-10-14 10:38:13,698 DEV : loss 0.1579323559999466 - f1-score (micro avg) 0.7854
2023-10-14 10:38:13,726 ----------------------------------------------------------------------------------------------------
2023-10-14 10:38:21,834 epoch 7 - iter 144/1445 - loss 0.01472983 - time (sec): 8.11 - samples/sec: 2164.44 - lr: 0.000013 - momentum: 0.000000
2023-10-14 10:38:30,227 epoch 7 - iter 288/1445 - loss 0.01529246 - time (sec): 16.50 - samples/sec: 2221.27 - lr: 0.000013 - momentum: 0.000000
2023-10-14 10:38:37,324 epoch 7 - iter 432/1445 - loss 0.01675210 - time (sec): 23.60 - samples/sec: 2268.28 - lr: 0.000012 - momentum: 0.000000
2023-10-14 10:38:44,693 epoch 7 - iter 576/1445 - loss 0.01805192 - time (sec): 30.97 - samples/sec: 2322.73 - lr: 0.000012 - momentum: 0.000000
2023-10-14 10:38:51,730 epoch 7 - iter 720/1445 - loss 0.01822939 - time (sec): 38.00 - samples/sec: 2328.47 - lr: 0.000012 - momentum: 0.000000
2023-10-14 10:38:58,764 epoch 7 - iter 864/1445 - loss 0.01917789 - time (sec): 45.04 - samples/sec: 2333.32 - lr: 0.000011 - momentum: 0.000000
2023-10-14 10:39:06,276 epoch 7 - iter 1008/1445 - loss 0.02084014 - time (sec): 52.55 - samples/sec: 2356.77 - lr: 0.000011 - momentum: 0.000000
2023-10-14 10:39:13,611 epoch 7 - iter 1152/1445 - loss 0.02033218 - time (sec): 59.88 - samples/sec: 2369.19 - lr: 0.000011 - momentum: 0.000000
2023-10-14 10:39:20,678 epoch 7 - iter 1296/1445 - loss 0.02070489 - time (sec): 66.95 - samples/sec: 2376.83 - lr: 0.000010 - momentum: 0.000000
2023-10-14 10:39:27,817 epoch 7 - iter 1440/1445 - loss 0.02027042 - time (sec): 74.09 - samples/sec: 2372.88 - lr: 0.000010 - momentum: 0.000000
2023-10-14 10:39:28,051 ----------------------------------------------------------------------------------------------------
2023-10-14 10:39:28,051 EPOCH 7 done: loss 0.0206 - lr: 0.000010
2023-10-14 10:39:31,558 DEV : loss 0.15825796127319336 - f1-score (micro avg) 0.8171
2023-10-14 10:39:31,574 saving best model
2023-10-14 10:39:32,037 ----------------------------------------------------------------------------------------------------
2023-10-14 10:39:39,139 epoch 8 - iter 144/1445 - loss 0.02396519 - time (sec): 7.10 - samples/sec: 2354.23 - lr: 0.000010 - momentum: 0.000000
2023-10-14 10:39:46,793 epoch 8 - iter 288/1445 - loss 0.01698943 - time (sec): 14.75 - samples/sec: 2344.64 - lr: 0.000009 - momentum: 0.000000
2023-10-14 10:39:54,032 epoch 8 - iter 432/1445 - loss 0.01654060 - time (sec): 21.99 - samples/sec: 2379.55 - lr: 0.000009 - momentum: 0.000000
2023-10-14 10:40:01,390 epoch 8 - iter 576/1445 - loss 0.01471133 - time (sec): 29.35 - samples/sec: 2374.05 - lr: 0.000009 - momentum: 0.000000
2023-10-14 10:40:08,625 epoch 8 - iter 720/1445 - loss 0.01430743 - time (sec): 36.59 - samples/sec: 2403.83 - lr: 0.000008 - momentum: 0.000000
2023-10-14 10:40:15,611 epoch 8 - iter 864/1445 - loss 0.01506483 - time (sec): 43.57 - samples/sec: 2399.74 - lr: 0.000008 - momentum: 0.000000
2023-10-14 10:40:23,044 epoch 8 - iter 1008/1445 - loss 0.01564489 - time (sec): 51.01 - samples/sec: 2406.42 - lr: 0.000008 - momentum: 0.000000
2023-10-14 10:40:30,274 epoch 8 - iter 1152/1445 - loss 0.01510179 - time (sec): 58.24 - samples/sec: 2410.96 - lr: 0.000007 - momentum: 0.000000
2023-10-14 10:40:37,318 epoch 8 - iter 1296/1445 - loss 0.01417517 - time (sec): 65.28 - samples/sec: 2413.79 - lr: 0.000007 - momentum: 0.000000
2023-10-14 10:40:44,723 epoch 8 - iter 1440/1445 - loss 0.01465237 - time (sec): 72.68 - samples/sec: 2413.51 - lr: 0.000007 - momentum: 0.000000
2023-10-14 10:40:44,975 ----------------------------------------------------------------------------------------------------
2023-10-14 10:40:44,975 EPOCH 8 done: loss 0.0146 - lr: 0.000007
2023-10-14 10:40:48,530 DEV : loss 0.16648930311203003 - f1-score (micro avg) 0.8246
2023-10-14 10:40:48,550 saving best model
2023-10-14 10:40:49,056 ----------------------------------------------------------------------------------------------------
2023-10-14 10:40:56,452 epoch 9 - iter 144/1445 - loss 0.00629566 - time (sec): 7.39 - samples/sec: 2457.52 - lr: 0.000006 - momentum: 0.000000
2023-10-14 10:41:03,691 epoch 9 - iter 288/1445 - loss 0.00790835 - time (sec): 14.63 - samples/sec: 2414.93 - lr: 0.000006 - momentum: 0.000000
2023-10-14 10:41:11,244 epoch 9 - iter 432/1445 - loss 0.00782712 - time (sec): 22.18 - samples/sec: 2349.41 - lr: 0.000006 - momentum: 0.000000
2023-10-14 10:41:18,904 epoch 9 - iter 576/1445 - loss 0.00916480 - time (sec): 29.84 - samples/sec: 2391.33 - lr: 0.000005 - momentum: 0.000000
2023-10-14 10:41:26,049 epoch 9 - iter 720/1445 - loss 0.00886500 - time (sec): 36.99 - samples/sec: 2399.86 - lr: 0.000005 - momentum: 0.000000
2023-10-14 10:41:33,379 epoch 9 - iter 864/1445 - loss 0.00941845 - time (sec): 44.32 - samples/sec: 2406.98 - lr: 0.000005 - momentum: 0.000000
2023-10-14 10:41:40,390 epoch 9 - iter 1008/1445 - loss 0.00912052 - time (sec): 51.33 - samples/sec: 2406.29 - lr: 0.000004 - momentum: 0.000000
2023-10-14 10:41:47,769 epoch 9 - iter 1152/1445 - loss 0.00928227 - time (sec): 58.71 - samples/sec: 2409.76 - lr: 0.000004 - momentum: 0.000000
2023-10-14 10:41:54,865 epoch 9 - iter 1296/1445 - loss 0.00960617 - time (sec): 65.81 - samples/sec: 2408.28 - lr: 0.000004 - momentum: 0.000000
2023-10-14 10:42:02,054 epoch 9 - iter 1440/1445 - loss 0.00965262 - time (sec): 72.99 - samples/sec: 2407.34 - lr: 0.000003 - momentum: 0.000000
2023-10-14 10:42:02,314 ----------------------------------------------------------------------------------------------------
2023-10-14 10:42:02,314 EPOCH 9 done: loss 0.0097 - lr: 0.000003
2023-10-14 10:42:06,175 DEV : loss 0.1673649400472641 - f1-score (micro avg) 0.8239
2023-10-14 10:42:06,191 ----------------------------------------------------------------------------------------------------
2023-10-14 10:42:13,336 epoch 10 - iter 144/1445 - loss 0.00886764 - time (sec): 7.14 - samples/sec: 2336.97 - lr: 0.000003 - momentum: 0.000000
2023-10-14 10:42:20,685 epoch 10 - iter 288/1445 - loss 0.00772641 - time (sec): 14.49 - samples/sec: 2396.61 - lr: 0.000003 - momentum: 0.000000
2023-10-14 10:42:28,082 epoch 10 - iter 432/1445 - loss 0.00818643 - time (sec): 21.89 - samples/sec: 2393.56 - lr: 0.000002 - momentum: 0.000000
2023-10-14 10:42:35,619 epoch 10 - iter 576/1445 - loss 0.00943870 - time (sec): 29.43 - samples/sec: 2417.73 - lr: 0.000002 - momentum: 0.000000
2023-10-14 10:42:43,034 epoch 10 - iter 720/1445 - loss 0.00885679 - time (sec): 36.84 - samples/sec: 2432.46 - lr: 0.000002 - momentum: 0.000000
2023-10-14 10:42:50,269 epoch 10 - iter 864/1445 - loss 0.00944317 - time (sec): 44.08 - samples/sec: 2424.76 - lr: 0.000001 - momentum: 0.000000
2023-10-14 10:42:57,240 epoch 10 - iter 1008/1445 - loss 0.00869228 - time (sec): 51.05 - samples/sec: 2401.38 - lr: 0.000001 - momentum: 0.000000
2023-10-14 10:43:04,381 epoch 10 - iter 1152/1445 - loss 0.00816640 - time (sec): 58.19 - samples/sec: 2394.05 - lr: 0.000001 - momentum: 0.000000
2023-10-14 10:43:11,807 epoch 10 - iter 1296/1445 - loss 0.00783806 - time (sec): 65.61 - samples/sec: 2404.43 - lr: 0.000000 - momentum: 0.000000
2023-10-14 10:43:19,049 epoch 10 - iter 1440/1445 - loss 0.00751126 - time (sec): 72.86 - samples/sec: 2408.74 - lr: 0.000000 - momentum: 0.000000
2023-10-14 10:43:19,315 ----------------------------------------------------------------------------------------------------
2023-10-14 10:43:19,316 EPOCH 10 done: loss 0.0075 - lr: 0.000000
2023-10-14 10:43:22,979 DEV : loss 0.17992718517780304 - f1-score (micro avg) 0.8234
2023-10-14 10:43:23,439 ----------------------------------------------------------------------------------------------------
2023-10-14 10:43:23,441 Loading model from best epoch ...
2023-10-14 10:43:25,159 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
2023-10-14 10:43:28,390
Results:
- F-score (micro) 0.8056
- F-score (macro) 0.7149
- Accuracy 0.6863
By class:
precision recall f1-score support
PER 0.8326 0.8050 0.8186 482
LOC 0.8916 0.7904 0.8380 458
ORG 0.5345 0.4493 0.4882 69
micro avg 0.8398 0.7740 0.8056 1009
macro avg 0.7529 0.6815 0.7149 1009
weighted avg 0.8390 0.7740 0.8048 1009
2023-10-14 10:43:28,390 ----------------------------------------------------------------------------------------------------
|