File size: 24,314 Bytes
22b39ea |
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 |
2023-10-20 09:41:09,199 ----------------------------------------------------------------------------------------------------
2023-10-20 09:41:09,200 Model: "SequenceTagger(
(embeddings): TransformerWordEmbeddings(
(model): BertModel(
(embeddings): BertEmbeddings(
(word_embeddings): Embedding(32001, 128)
(position_embeddings): Embedding(512, 128)
(token_type_embeddings): Embedding(2, 128)
(LayerNorm): LayerNorm((128,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(encoder): BertEncoder(
(layer): ModuleList(
(0-1): 2 x BertLayer(
(attention): BertAttention(
(self): BertSelfAttention(
(query): Linear(in_features=128, out_features=128, bias=True)
(key): Linear(in_features=128, out_features=128, bias=True)
(value): Linear(in_features=128, out_features=128, bias=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(output): BertSelfOutput(
(dense): Linear(in_features=128, out_features=128, bias=True)
(LayerNorm): LayerNorm((128,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
(intermediate): BertIntermediate(
(dense): Linear(in_features=128, out_features=512, bias=True)
(intermediate_act_fn): GELUActivation()
)
(output): BertOutput(
(dense): Linear(in_features=512, out_features=128, bias=True)
(LayerNorm): LayerNorm((128,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
)
(pooler): BertPooler(
(dense): Linear(in_features=128, out_features=128, bias=True)
(activation): Tanh()
)
)
)
(locked_dropout): LockedDropout(p=0.5)
(linear): Linear(in_features=128, out_features=13, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-20 09:41:09,200 ----------------------------------------------------------------------------------------------------
2023-10-20 09:41:09,200 MultiCorpus: 6183 train + 680 dev + 2113 test sentences
- NER_HIPE_2022 Corpus: 6183 train + 680 dev + 2113 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/topres19th/en/with_doc_seperator
2023-10-20 09:41:09,200 ----------------------------------------------------------------------------------------------------
2023-10-20 09:41:09,200 Train: 6183 sentences
2023-10-20 09:41:09,200 (train_with_dev=False, train_with_test=False)
2023-10-20 09:41:09,200 ----------------------------------------------------------------------------------------------------
2023-10-20 09:41:09,200 Training Params:
2023-10-20 09:41:09,200 - learning_rate: "5e-05"
2023-10-20 09:41:09,200 - mini_batch_size: "4"
2023-10-20 09:41:09,200 - max_epochs: "10"
2023-10-20 09:41:09,200 - shuffle: "True"
2023-10-20 09:41:09,200 ----------------------------------------------------------------------------------------------------
2023-10-20 09:41:09,200 Plugins:
2023-10-20 09:41:09,200 - TensorboardLogger
2023-10-20 09:41:09,200 - LinearScheduler | warmup_fraction: '0.1'
2023-10-20 09:41:09,200 ----------------------------------------------------------------------------------------------------
2023-10-20 09:41:09,200 Final evaluation on model from best epoch (best-model.pt)
2023-10-20 09:41:09,200 - metric: "('micro avg', 'f1-score')"
2023-10-20 09:41:09,200 ----------------------------------------------------------------------------------------------------
2023-10-20 09:41:09,200 Computation:
2023-10-20 09:41:09,200 - compute on device: cuda:0
2023-10-20 09:41:09,200 - embedding storage: none
2023-10-20 09:41:09,200 ----------------------------------------------------------------------------------------------------
2023-10-20 09:41:09,200 Model training base path: "hmbench-topres19th/en-dbmdz/bert-tiny-historic-multilingual-cased-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-3"
2023-10-20 09:41:09,201 ----------------------------------------------------------------------------------------------------
2023-10-20 09:41:09,201 ----------------------------------------------------------------------------------------------------
2023-10-20 09:41:09,201 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-20 09:41:11,581 epoch 1 - iter 154/1546 - loss 2.73201796 - time (sec): 2.38 - samples/sec: 5211.62 - lr: 0.000005 - momentum: 0.000000
2023-10-20 09:41:13,997 epoch 1 - iter 308/1546 - loss 2.28602428 - time (sec): 4.80 - samples/sec: 5370.04 - lr: 0.000010 - momentum: 0.000000
2023-10-20 09:41:16,612 epoch 1 - iter 462/1546 - loss 1.80447606 - time (sec): 7.41 - samples/sec: 5025.78 - lr: 0.000015 - momentum: 0.000000
2023-10-20 09:41:19,191 epoch 1 - iter 616/1546 - loss 1.45860759 - time (sec): 9.99 - samples/sec: 4859.62 - lr: 0.000020 - momentum: 0.000000
2023-10-20 09:41:21,393 epoch 1 - iter 770/1546 - loss 1.21626669 - time (sec): 12.19 - samples/sec: 4991.92 - lr: 0.000025 - momentum: 0.000000
2023-10-20 09:41:23,508 epoch 1 - iter 924/1546 - loss 1.05743701 - time (sec): 14.31 - samples/sec: 5116.89 - lr: 0.000030 - momentum: 0.000000
2023-10-20 09:41:25,915 epoch 1 - iter 1078/1546 - loss 0.93364531 - time (sec): 16.71 - samples/sec: 5203.28 - lr: 0.000035 - momentum: 0.000000
2023-10-20 09:41:28,282 epoch 1 - iter 1232/1546 - loss 0.85352136 - time (sec): 19.08 - samples/sec: 5178.04 - lr: 0.000040 - momentum: 0.000000
2023-10-20 09:41:30,632 epoch 1 - iter 1386/1546 - loss 0.78183431 - time (sec): 21.43 - samples/sec: 5187.56 - lr: 0.000045 - momentum: 0.000000
2023-10-20 09:41:32,915 epoch 1 - iter 1540/1546 - loss 0.72103837 - time (sec): 23.71 - samples/sec: 5221.53 - lr: 0.000050 - momentum: 0.000000
2023-10-20 09:41:33,007 ----------------------------------------------------------------------------------------------------
2023-10-20 09:41:33,008 EPOCH 1 done: loss 0.7187 - lr: 0.000050
2023-10-20 09:41:33,999 DEV : loss 0.12611015141010284 - f1-score (micro avg) 0.0923
2023-10-20 09:41:34,010 saving best model
2023-10-20 09:41:34,039 ----------------------------------------------------------------------------------------------------
2023-10-20 09:41:36,381 epoch 2 - iter 154/1546 - loss 0.19147722 - time (sec): 2.34 - samples/sec: 5814.62 - lr: 0.000049 - momentum: 0.000000
2023-10-20 09:41:38,793 epoch 2 - iter 308/1546 - loss 0.18883533 - time (sec): 4.75 - samples/sec: 5458.41 - lr: 0.000049 - momentum: 0.000000
2023-10-20 09:41:41,367 epoch 2 - iter 462/1546 - loss 0.19117660 - time (sec): 7.33 - samples/sec: 5234.09 - lr: 0.000048 - momentum: 0.000000
2023-10-20 09:41:43,731 epoch 2 - iter 616/1546 - loss 0.18776788 - time (sec): 9.69 - samples/sec: 5212.68 - lr: 0.000048 - momentum: 0.000000
2023-10-20 09:41:46,014 epoch 2 - iter 770/1546 - loss 0.19099004 - time (sec): 11.98 - samples/sec: 5248.49 - lr: 0.000047 - momentum: 0.000000
2023-10-20 09:41:48,419 epoch 2 - iter 924/1546 - loss 0.18616969 - time (sec): 14.38 - samples/sec: 5262.06 - lr: 0.000047 - momentum: 0.000000
2023-10-20 09:41:50,714 epoch 2 - iter 1078/1546 - loss 0.18577369 - time (sec): 16.67 - samples/sec: 5266.37 - lr: 0.000046 - momentum: 0.000000
2023-10-20 09:41:53,055 epoch 2 - iter 1232/1546 - loss 0.18463930 - time (sec): 19.02 - samples/sec: 5225.70 - lr: 0.000046 - momentum: 0.000000
2023-10-20 09:41:55,521 epoch 2 - iter 1386/1546 - loss 0.18415167 - time (sec): 21.48 - samples/sec: 5196.25 - lr: 0.000045 - momentum: 0.000000
2023-10-20 09:41:57,913 epoch 2 - iter 1540/1546 - loss 0.18017139 - time (sec): 23.87 - samples/sec: 5186.12 - lr: 0.000044 - momentum: 0.000000
2023-10-20 09:41:58,001 ----------------------------------------------------------------------------------------------------
2023-10-20 09:41:58,002 EPOCH 2 done: loss 0.1802 - lr: 0.000044
2023-10-20 09:41:59,100 DEV : loss 0.09276499599218369 - f1-score (micro avg) 0.463
2023-10-20 09:41:59,115 saving best model
2023-10-20 09:41:59,156 ----------------------------------------------------------------------------------------------------
2023-10-20 09:42:01,495 epoch 3 - iter 154/1546 - loss 0.17659999 - time (sec): 2.34 - samples/sec: 4765.94 - lr: 0.000044 - momentum: 0.000000
2023-10-20 09:42:03,992 epoch 3 - iter 308/1546 - loss 0.15431148 - time (sec): 4.84 - samples/sec: 5034.38 - lr: 0.000043 - momentum: 0.000000
2023-10-20 09:42:06,185 epoch 3 - iter 462/1546 - loss 0.15873776 - time (sec): 7.03 - samples/sec: 5224.34 - lr: 0.000043 - momentum: 0.000000
2023-10-20 09:42:08,363 epoch 3 - iter 616/1546 - loss 0.15154501 - time (sec): 9.21 - samples/sec: 5250.01 - lr: 0.000042 - momentum: 0.000000
2023-10-20 09:42:10,521 epoch 3 - iter 770/1546 - loss 0.15058689 - time (sec): 11.36 - samples/sec: 5370.97 - lr: 0.000042 - momentum: 0.000000
2023-10-20 09:42:12,875 epoch 3 - iter 924/1546 - loss 0.14854044 - time (sec): 13.72 - samples/sec: 5401.12 - lr: 0.000041 - momentum: 0.000000
2023-10-20 09:42:15,237 epoch 3 - iter 1078/1546 - loss 0.14970216 - time (sec): 16.08 - samples/sec: 5374.95 - lr: 0.000041 - momentum: 0.000000
2023-10-20 09:42:17,568 epoch 3 - iter 1232/1546 - loss 0.15198622 - time (sec): 18.41 - samples/sec: 5357.42 - lr: 0.000040 - momentum: 0.000000
2023-10-20 09:42:19,976 epoch 3 - iter 1386/1546 - loss 0.15043598 - time (sec): 20.82 - samples/sec: 5361.18 - lr: 0.000039 - momentum: 0.000000
2023-10-20 09:42:22,330 epoch 3 - iter 1540/1546 - loss 0.15010108 - time (sec): 23.17 - samples/sec: 5342.90 - lr: 0.000039 - momentum: 0.000000
2023-10-20 09:42:22,421 ----------------------------------------------------------------------------------------------------
2023-10-20 09:42:22,421 EPOCH 3 done: loss 0.1507 - lr: 0.000039
2023-10-20 09:42:23,519 DEV : loss 0.09011607617139816 - f1-score (micro avg) 0.5241
2023-10-20 09:42:23,531 saving best model
2023-10-20 09:42:23,572 ----------------------------------------------------------------------------------------------------
2023-10-20 09:42:26,002 epoch 4 - iter 154/1546 - loss 0.13234173 - time (sec): 2.43 - samples/sec: 5609.47 - lr: 0.000038 - momentum: 0.000000
2023-10-20 09:42:28,290 epoch 4 - iter 308/1546 - loss 0.13714759 - time (sec): 4.72 - samples/sec: 5340.94 - lr: 0.000038 - momentum: 0.000000
2023-10-20 09:42:30,599 epoch 4 - iter 462/1546 - loss 0.12728020 - time (sec): 7.03 - samples/sec: 5327.92 - lr: 0.000037 - momentum: 0.000000
2023-10-20 09:42:32,941 epoch 4 - iter 616/1546 - loss 0.13080542 - time (sec): 9.37 - samples/sec: 5291.06 - lr: 0.000037 - momentum: 0.000000
2023-10-20 09:42:35,381 epoch 4 - iter 770/1546 - loss 0.12845508 - time (sec): 11.81 - samples/sec: 5301.24 - lr: 0.000036 - momentum: 0.000000
2023-10-20 09:42:37,710 epoch 4 - iter 924/1546 - loss 0.13383808 - time (sec): 14.14 - samples/sec: 5289.52 - lr: 0.000036 - momentum: 0.000000
2023-10-20 09:42:40,090 epoch 4 - iter 1078/1546 - loss 0.13508658 - time (sec): 16.52 - samples/sec: 5292.52 - lr: 0.000035 - momentum: 0.000000
2023-10-20 09:42:42,510 epoch 4 - iter 1232/1546 - loss 0.13532445 - time (sec): 18.94 - samples/sec: 5289.66 - lr: 0.000034 - momentum: 0.000000
2023-10-20 09:42:44,865 epoch 4 - iter 1386/1546 - loss 0.13315587 - time (sec): 21.29 - samples/sec: 5283.64 - lr: 0.000034 - momentum: 0.000000
2023-10-20 09:42:47,186 epoch 4 - iter 1540/1546 - loss 0.13381815 - time (sec): 23.61 - samples/sec: 5244.78 - lr: 0.000033 - momentum: 0.000000
2023-10-20 09:42:47,277 ----------------------------------------------------------------------------------------------------
2023-10-20 09:42:47,277 EPOCH 4 done: loss 0.1337 - lr: 0.000033
2023-10-20 09:42:48,357 DEV : loss 0.09306028485298157 - f1-score (micro avg) 0.5244
2023-10-20 09:42:48,368 saving best model
2023-10-20 09:42:48,408 ----------------------------------------------------------------------------------------------------
2023-10-20 09:42:50,715 epoch 5 - iter 154/1546 - loss 0.11681683 - time (sec): 2.31 - samples/sec: 5404.86 - lr: 0.000033 - momentum: 0.000000
2023-10-20 09:42:52,960 epoch 5 - iter 308/1546 - loss 0.11485080 - time (sec): 4.55 - samples/sec: 5433.24 - lr: 0.000032 - momentum: 0.000000
2023-10-20 09:42:55,341 epoch 5 - iter 462/1546 - loss 0.12181429 - time (sec): 6.93 - samples/sec: 5293.09 - lr: 0.000032 - momentum: 0.000000
2023-10-20 09:42:57,788 epoch 5 - iter 616/1546 - loss 0.11977320 - time (sec): 9.38 - samples/sec: 5288.81 - lr: 0.000031 - momentum: 0.000000
2023-10-20 09:43:00,174 epoch 5 - iter 770/1546 - loss 0.11612074 - time (sec): 11.77 - samples/sec: 5309.53 - lr: 0.000031 - momentum: 0.000000
2023-10-20 09:43:02,601 epoch 5 - iter 924/1546 - loss 0.12322243 - time (sec): 14.19 - samples/sec: 5285.12 - lr: 0.000030 - momentum: 0.000000
2023-10-20 09:43:04,977 epoch 5 - iter 1078/1546 - loss 0.12454524 - time (sec): 16.57 - samples/sec: 5260.58 - lr: 0.000029 - momentum: 0.000000
2023-10-20 09:43:07,312 epoch 5 - iter 1232/1546 - loss 0.12565251 - time (sec): 18.90 - samples/sec: 5246.75 - lr: 0.000029 - momentum: 0.000000
2023-10-20 09:43:09,670 epoch 5 - iter 1386/1546 - loss 0.12496953 - time (sec): 21.26 - samples/sec: 5245.43 - lr: 0.000028 - momentum: 0.000000
2023-10-20 09:43:12,064 epoch 5 - iter 1540/1546 - loss 0.12186724 - time (sec): 23.66 - samples/sec: 5235.16 - lr: 0.000028 - momentum: 0.000000
2023-10-20 09:43:12,158 ----------------------------------------------------------------------------------------------------
2023-10-20 09:43:12,158 EPOCH 5 done: loss 0.1221 - lr: 0.000028
2023-10-20 09:43:13,241 DEV : loss 0.09604145586490631 - f1-score (micro avg) 0.5656
2023-10-20 09:43:13,254 saving best model
2023-10-20 09:43:13,286 ----------------------------------------------------------------------------------------------------
2023-10-20 09:43:15,726 epoch 6 - iter 154/1546 - loss 0.14501961 - time (sec): 2.44 - samples/sec: 5098.88 - lr: 0.000027 - momentum: 0.000000
2023-10-20 09:43:18,081 epoch 6 - iter 308/1546 - loss 0.12419256 - time (sec): 4.79 - samples/sec: 5241.47 - lr: 0.000027 - momentum: 0.000000
2023-10-20 09:43:20,410 epoch 6 - iter 462/1546 - loss 0.11952545 - time (sec): 7.12 - samples/sec: 5247.84 - lr: 0.000026 - momentum: 0.000000
2023-10-20 09:43:22,764 epoch 6 - iter 616/1546 - loss 0.12112158 - time (sec): 9.48 - samples/sec: 5154.77 - lr: 0.000026 - momentum: 0.000000
2023-10-20 09:43:25,146 epoch 6 - iter 770/1546 - loss 0.11804613 - time (sec): 11.86 - samples/sec: 5239.33 - lr: 0.000025 - momentum: 0.000000
2023-10-20 09:43:27,511 epoch 6 - iter 924/1546 - loss 0.11766846 - time (sec): 14.22 - samples/sec: 5215.77 - lr: 0.000024 - momentum: 0.000000
2023-10-20 09:43:29,890 epoch 6 - iter 1078/1546 - loss 0.11754179 - time (sec): 16.60 - samples/sec: 5214.73 - lr: 0.000024 - momentum: 0.000000
2023-10-20 09:43:32,248 epoch 6 - iter 1232/1546 - loss 0.11564541 - time (sec): 18.96 - samples/sec: 5218.61 - lr: 0.000023 - momentum: 0.000000
2023-10-20 09:43:34,677 epoch 6 - iter 1386/1546 - loss 0.11640739 - time (sec): 21.39 - samples/sec: 5238.26 - lr: 0.000023 - momentum: 0.000000
2023-10-20 09:43:36,968 epoch 6 - iter 1540/1546 - loss 0.11549879 - time (sec): 23.68 - samples/sec: 5228.42 - lr: 0.000022 - momentum: 0.000000
2023-10-20 09:43:37,050 ----------------------------------------------------------------------------------------------------
2023-10-20 09:43:37,050 EPOCH 6 done: loss 0.1151 - lr: 0.000022
2023-10-20 09:43:38,136 DEV : loss 0.10433212667703629 - f1-score (micro avg) 0.5727
2023-10-20 09:43:38,149 saving best model
2023-10-20 09:43:38,190 ----------------------------------------------------------------------------------------------------
2023-10-20 09:43:40,570 epoch 7 - iter 154/1546 - loss 0.12205843 - time (sec): 2.38 - samples/sec: 5053.28 - lr: 0.000022 - momentum: 0.000000
2023-10-20 09:43:42,919 epoch 7 - iter 308/1546 - loss 0.10630383 - time (sec): 4.73 - samples/sec: 5325.75 - lr: 0.000021 - momentum: 0.000000
2023-10-20 09:43:45,240 epoch 7 - iter 462/1546 - loss 0.11247321 - time (sec): 7.05 - samples/sec: 5240.98 - lr: 0.000021 - momentum: 0.000000
2023-10-20 09:43:47,691 epoch 7 - iter 616/1546 - loss 0.10731717 - time (sec): 9.50 - samples/sec: 5238.98 - lr: 0.000020 - momentum: 0.000000
2023-10-20 09:43:50,026 epoch 7 - iter 770/1546 - loss 0.10659080 - time (sec): 11.84 - samples/sec: 5134.02 - lr: 0.000019 - momentum: 0.000000
2023-10-20 09:43:52,380 epoch 7 - iter 924/1546 - loss 0.10675804 - time (sec): 14.19 - samples/sec: 5129.08 - lr: 0.000019 - momentum: 0.000000
2023-10-20 09:43:54,757 epoch 7 - iter 1078/1546 - loss 0.10617289 - time (sec): 16.57 - samples/sec: 5171.36 - lr: 0.000018 - momentum: 0.000000
2023-10-20 09:43:57,215 epoch 7 - iter 1232/1546 - loss 0.10595788 - time (sec): 19.02 - samples/sec: 5179.16 - lr: 0.000018 - momentum: 0.000000
2023-10-20 09:43:59,544 epoch 7 - iter 1386/1546 - loss 0.10475076 - time (sec): 21.35 - samples/sec: 5221.19 - lr: 0.000017 - momentum: 0.000000
2023-10-20 09:44:01,661 epoch 7 - iter 1540/1546 - loss 0.10684330 - time (sec): 23.47 - samples/sec: 5274.67 - lr: 0.000017 - momentum: 0.000000
2023-10-20 09:44:01,740 ----------------------------------------------------------------------------------------------------
2023-10-20 09:44:01,741 EPOCH 7 done: loss 0.1067 - lr: 0.000017
2023-10-20 09:44:02,845 DEV : loss 0.10244878381490707 - f1-score (micro avg) 0.5978
2023-10-20 09:44:02,858 saving best model
2023-10-20 09:44:02,900 ----------------------------------------------------------------------------------------------------
2023-10-20 09:44:05,137 epoch 8 - iter 154/1546 - loss 0.07782089 - time (sec): 2.24 - samples/sec: 5379.85 - lr: 0.000016 - momentum: 0.000000
2023-10-20 09:44:07,488 epoch 8 - iter 308/1546 - loss 0.09938386 - time (sec): 4.59 - samples/sec: 5205.84 - lr: 0.000016 - momentum: 0.000000
2023-10-20 09:44:09,760 epoch 8 - iter 462/1546 - loss 0.10503251 - time (sec): 6.86 - samples/sec: 5325.75 - lr: 0.000015 - momentum: 0.000000
2023-10-20 09:44:12,071 epoch 8 - iter 616/1546 - loss 0.09702954 - time (sec): 9.17 - samples/sec: 5452.15 - lr: 0.000014 - momentum: 0.000000
2023-10-20 09:44:14,469 epoch 8 - iter 770/1546 - loss 0.09883319 - time (sec): 11.57 - samples/sec: 5429.20 - lr: 0.000014 - momentum: 0.000000
2023-10-20 09:44:16,911 epoch 8 - iter 924/1546 - loss 0.10089289 - time (sec): 14.01 - samples/sec: 5368.77 - lr: 0.000013 - momentum: 0.000000
2023-10-20 09:44:19,253 epoch 8 - iter 1078/1546 - loss 0.10490967 - time (sec): 16.35 - samples/sec: 5317.93 - lr: 0.000013 - momentum: 0.000000
2023-10-20 09:44:21,864 epoch 8 - iter 1232/1546 - loss 0.10403049 - time (sec): 18.96 - samples/sec: 5252.66 - lr: 0.000012 - momentum: 0.000000
2023-10-20 09:44:24,326 epoch 8 - iter 1386/1546 - loss 0.10516244 - time (sec): 21.43 - samples/sec: 5188.95 - lr: 0.000012 - momentum: 0.000000
2023-10-20 09:44:26,789 epoch 8 - iter 1540/1546 - loss 0.10342984 - time (sec): 23.89 - samples/sec: 5181.70 - lr: 0.000011 - momentum: 0.000000
2023-10-20 09:44:26,877 ----------------------------------------------------------------------------------------------------
2023-10-20 09:44:26,878 EPOCH 8 done: loss 0.1032 - lr: 0.000011
2023-10-20 09:44:27,986 DEV : loss 0.10922187566757202 - f1-score (micro avg) 0.6057
2023-10-20 09:44:27,998 saving best model
2023-10-20 09:44:28,038 ----------------------------------------------------------------------------------------------------
2023-10-20 09:44:30,360 epoch 9 - iter 154/1546 - loss 0.08640753 - time (sec): 2.32 - samples/sec: 5049.90 - lr: 0.000011 - momentum: 0.000000
2023-10-20 09:44:32,771 epoch 9 - iter 308/1546 - loss 0.10139803 - time (sec): 4.73 - samples/sec: 5185.55 - lr: 0.000010 - momentum: 0.000000
2023-10-20 09:44:35,161 epoch 9 - iter 462/1546 - loss 0.10459999 - time (sec): 7.12 - samples/sec: 5147.67 - lr: 0.000009 - momentum: 0.000000
2023-10-20 09:44:37,562 epoch 9 - iter 616/1546 - loss 0.10331605 - time (sec): 9.52 - samples/sec: 5178.00 - lr: 0.000009 - momentum: 0.000000
2023-10-20 09:44:39,934 epoch 9 - iter 770/1546 - loss 0.10432797 - time (sec): 11.90 - samples/sec: 5263.07 - lr: 0.000008 - momentum: 0.000000
2023-10-20 09:44:42,272 epoch 9 - iter 924/1546 - loss 0.10164891 - time (sec): 14.23 - samples/sec: 5224.54 - lr: 0.000008 - momentum: 0.000000
2023-10-20 09:44:44,608 epoch 9 - iter 1078/1546 - loss 0.10042381 - time (sec): 16.57 - samples/sec: 5216.95 - lr: 0.000007 - momentum: 0.000000
2023-10-20 09:44:47,009 epoch 9 - iter 1232/1546 - loss 0.09913846 - time (sec): 18.97 - samples/sec: 5227.53 - lr: 0.000007 - momentum: 0.000000
2023-10-20 09:44:49,379 epoch 9 - iter 1386/1546 - loss 0.09843496 - time (sec): 21.34 - samples/sec: 5225.60 - lr: 0.000006 - momentum: 0.000000
2023-10-20 09:44:51,770 epoch 9 - iter 1540/1546 - loss 0.09869379 - time (sec): 23.73 - samples/sec: 5218.24 - lr: 0.000006 - momentum: 0.000000
2023-10-20 09:44:51,858 ----------------------------------------------------------------------------------------------------
2023-10-20 09:44:51,859 EPOCH 9 done: loss 0.0986 - lr: 0.000006
2023-10-20 09:44:52,949 DEV : loss 0.10991593450307846 - f1-score (micro avg) 0.6157
2023-10-20 09:44:52,962 saving best model
2023-10-20 09:44:52,995 ----------------------------------------------------------------------------------------------------
2023-10-20 09:44:55,276 epoch 10 - iter 154/1546 - loss 0.08007322 - time (sec): 2.28 - samples/sec: 5351.05 - lr: 0.000005 - momentum: 0.000000
2023-10-20 09:44:57,614 epoch 10 - iter 308/1546 - loss 0.07876762 - time (sec): 4.62 - samples/sec: 4959.39 - lr: 0.000004 - momentum: 0.000000
2023-10-20 09:45:00,139 epoch 10 - iter 462/1546 - loss 0.08637072 - time (sec): 7.14 - samples/sec: 5137.18 - lr: 0.000004 - momentum: 0.000000
2023-10-20 09:45:02,578 epoch 10 - iter 616/1546 - loss 0.09043747 - time (sec): 9.58 - samples/sec: 5157.41 - lr: 0.000003 - momentum: 0.000000
2023-10-20 09:45:04,930 epoch 10 - iter 770/1546 - loss 0.09135879 - time (sec): 11.93 - samples/sec: 5188.74 - lr: 0.000003 - momentum: 0.000000
2023-10-20 09:45:07,334 epoch 10 - iter 924/1546 - loss 0.09566442 - time (sec): 14.34 - samples/sec: 5196.85 - lr: 0.000002 - momentum: 0.000000
2023-10-20 09:45:09,778 epoch 10 - iter 1078/1546 - loss 0.09739706 - time (sec): 16.78 - samples/sec: 5147.91 - lr: 0.000002 - momentum: 0.000000
2023-10-20 09:45:12,353 epoch 10 - iter 1232/1546 - loss 0.09546748 - time (sec): 19.36 - samples/sec: 5149.80 - lr: 0.000001 - momentum: 0.000000
2023-10-20 09:45:14,711 epoch 10 - iter 1386/1546 - loss 0.09546014 - time (sec): 21.71 - samples/sec: 5127.46 - lr: 0.000001 - momentum: 0.000000
2023-10-20 09:45:17,110 epoch 10 - iter 1540/1546 - loss 0.09650173 - time (sec): 24.11 - samples/sec: 5137.09 - lr: 0.000000 - momentum: 0.000000
2023-10-20 09:45:17,203 ----------------------------------------------------------------------------------------------------
2023-10-20 09:45:17,203 EPOCH 10 done: loss 0.0964 - lr: 0.000000
2023-10-20 09:45:18,278 DEV : loss 0.11002404242753983 - f1-score (micro avg) 0.628
2023-10-20 09:45:18,291 saving best model
2023-10-20 09:45:18,358 ----------------------------------------------------------------------------------------------------
2023-10-20 09:45:18,359 Loading model from best epoch ...
2023-10-20 09:45:18,432 SequenceTagger predicts: Dictionary with 13 tags: O, S-LOC, B-LOC, E-LOC, I-LOC, S-BUILDING, B-BUILDING, E-BUILDING, I-BUILDING, S-STREET, B-STREET, E-STREET, I-STREET
2023-10-20 09:45:21,376
Results:
- F-score (micro) 0.5802
- F-score (macro) 0.3927
- Accuracy 0.4193
By class:
precision recall f1-score support
LOC 0.6257 0.6681 0.6462 946
BUILDING 0.2710 0.1568 0.1986 185
STREET 0.7500 0.2143 0.3333 56
micro avg 0.5940 0.5670 0.5802 1187
macro avg 0.5489 0.3464 0.3927 1187
weighted avg 0.5763 0.5670 0.5617 1187
2023-10-20 09:45:21,376 ----------------------------------------------------------------------------------------------------
|