File size: 24,105 Bytes
57e4064 |
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 |
2023-10-25 10:55:20,713 ----------------------------------------------------------------------------------------------------
2023-10-25 10:55:20,714 Model: "SequenceTagger(
(embeddings): TransformerWordEmbeddings(
(model): BertModel(
(embeddings): BertEmbeddings(
(word_embeddings): Embedding(64001, 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()
)"
2023-10-25 10:55:20,715 ----------------------------------------------------------------------------------------------------
2023-10-25 10:55:20,715 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
2023-10-25 10:55:20,715 ----------------------------------------------------------------------------------------------------
2023-10-25 10:55:20,715 Train: 6183 sentences
2023-10-25 10:55:20,715 (train_with_dev=False, train_with_test=False)
2023-10-25 10:55:20,715 ----------------------------------------------------------------------------------------------------
2023-10-25 10:55:20,715 Training Params:
2023-10-25 10:55:20,715 - learning_rate: "3e-05"
2023-10-25 10:55:20,715 - mini_batch_size: "4"
2023-10-25 10:55:20,715 - max_epochs: "10"
2023-10-25 10:55:20,715 - shuffle: "True"
2023-10-25 10:55:20,715 ----------------------------------------------------------------------------------------------------
2023-10-25 10:55:20,715 Plugins:
2023-10-25 10:55:20,715 - TensorboardLogger
2023-10-25 10:55:20,715 - LinearScheduler | warmup_fraction: '0.1'
2023-10-25 10:55:20,715 ----------------------------------------------------------------------------------------------------
2023-10-25 10:55:20,715 Final evaluation on model from best epoch (best-model.pt)
2023-10-25 10:55:20,715 - metric: "('micro avg', 'f1-score')"
2023-10-25 10:55:20,715 ----------------------------------------------------------------------------------------------------
2023-10-25 10:55:20,716 Computation:
2023-10-25 10:55:20,716 - compute on device: cuda:0
2023-10-25 10:55:20,716 - embedding storage: none
2023-10-25 10:55:20,716 ----------------------------------------------------------------------------------------------------
2023-10-25 10:55:20,716 Model training base path: "hmbench-topres19th/en-dbmdz/bert-base-historic-multilingual-64k-td-cased-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-2"
2023-10-25 10:55:20,716 ----------------------------------------------------------------------------------------------------
2023-10-25 10:55:20,716 ----------------------------------------------------------------------------------------------------
2023-10-25 10:55:20,716 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-25 10:55:28,912 epoch 1 - iter 154/1546 - loss 1.65969452 - time (sec): 8.20 - samples/sec: 1555.63 - lr: 0.000003 - momentum: 0.000000
2023-10-25 10:55:36,950 epoch 1 - iter 308/1546 - loss 0.93754585 - time (sec): 16.23 - samples/sec: 1534.93 - lr: 0.000006 - momentum: 0.000000
2023-10-25 10:55:44,983 epoch 1 - iter 462/1546 - loss 0.68571099 - time (sec): 24.27 - samples/sec: 1522.86 - lr: 0.000009 - momentum: 0.000000
2023-10-25 10:55:53,052 epoch 1 - iter 616/1546 - loss 0.54254456 - time (sec): 32.33 - samples/sec: 1540.72 - lr: 0.000012 - momentum: 0.000000
2023-10-25 10:56:01,033 epoch 1 - iter 770/1546 - loss 0.45649672 - time (sec): 40.32 - samples/sec: 1542.76 - lr: 0.000015 - momentum: 0.000000
2023-10-25 10:56:09,335 epoch 1 - iter 924/1546 - loss 0.40353128 - time (sec): 48.62 - samples/sec: 1521.17 - lr: 0.000018 - momentum: 0.000000
2023-10-25 10:56:17,417 epoch 1 - iter 1078/1546 - loss 0.36171769 - time (sec): 56.70 - samples/sec: 1531.40 - lr: 0.000021 - momentum: 0.000000
2023-10-25 10:56:25,605 epoch 1 - iter 1232/1546 - loss 0.32774519 - time (sec): 64.89 - samples/sec: 1535.03 - lr: 0.000024 - momentum: 0.000000
2023-10-25 10:56:33,774 epoch 1 - iter 1386/1546 - loss 0.30231800 - time (sec): 73.06 - samples/sec: 1531.32 - lr: 0.000027 - momentum: 0.000000
2023-10-25 10:56:41,853 epoch 1 - iter 1540/1546 - loss 0.28338013 - time (sec): 81.14 - samples/sec: 1526.05 - lr: 0.000030 - momentum: 0.000000
2023-10-25 10:56:42,169 ----------------------------------------------------------------------------------------------------
2023-10-25 10:56:42,170 EPOCH 1 done: loss 0.2827 - lr: 0.000030
2023-10-25 10:56:45,326 DEV : loss 0.062169428914785385 - f1-score (micro avg) 0.7459
2023-10-25 10:56:45,343 saving best model
2023-10-25 10:56:45,846 ----------------------------------------------------------------------------------------------------
2023-10-25 10:56:53,944 epoch 2 - iter 154/1546 - loss 0.08228178 - time (sec): 8.10 - samples/sec: 1517.82 - lr: 0.000030 - momentum: 0.000000
2023-10-25 10:57:02,045 epoch 2 - iter 308/1546 - loss 0.09250857 - time (sec): 16.20 - samples/sec: 1529.83 - lr: 0.000029 - momentum: 0.000000
2023-10-25 10:57:10,520 epoch 2 - iter 462/1546 - loss 0.09380708 - time (sec): 24.67 - samples/sec: 1461.41 - lr: 0.000029 - momentum: 0.000000
2023-10-25 10:57:18,800 epoch 2 - iter 616/1546 - loss 0.09259481 - time (sec): 32.95 - samples/sec: 1485.60 - lr: 0.000029 - momentum: 0.000000
2023-10-25 10:57:27,133 epoch 2 - iter 770/1546 - loss 0.08939337 - time (sec): 41.29 - samples/sec: 1492.76 - lr: 0.000028 - momentum: 0.000000
2023-10-25 10:57:35,482 epoch 2 - iter 924/1546 - loss 0.08678052 - time (sec): 49.63 - samples/sec: 1496.72 - lr: 0.000028 - momentum: 0.000000
2023-10-25 10:57:43,622 epoch 2 - iter 1078/1546 - loss 0.08519177 - time (sec): 57.77 - samples/sec: 1506.85 - lr: 0.000028 - momentum: 0.000000
2023-10-25 10:57:51,769 epoch 2 - iter 1232/1546 - loss 0.08376653 - time (sec): 65.92 - samples/sec: 1497.63 - lr: 0.000027 - momentum: 0.000000
2023-10-25 10:58:00,015 epoch 2 - iter 1386/1546 - loss 0.08475831 - time (sec): 74.17 - samples/sec: 1489.46 - lr: 0.000027 - momentum: 0.000000
2023-10-25 10:58:08,149 epoch 2 - iter 1540/1546 - loss 0.08472997 - time (sec): 82.30 - samples/sec: 1503.75 - lr: 0.000027 - momentum: 0.000000
2023-10-25 10:58:08,448 ----------------------------------------------------------------------------------------------------
2023-10-25 10:58:08,449 EPOCH 2 done: loss 0.0846 - lr: 0.000027
2023-10-25 10:58:11,612 DEV : loss 0.07261277735233307 - f1-score (micro avg) 0.7277
2023-10-25 10:58:11,628 ----------------------------------------------------------------------------------------------------
2023-10-25 10:58:19,774 epoch 3 - iter 154/1546 - loss 0.05836815 - time (sec): 8.14 - samples/sec: 1498.97 - lr: 0.000026 - momentum: 0.000000
2023-10-25 10:58:27,451 epoch 3 - iter 308/1546 - loss 0.05750916 - time (sec): 15.82 - samples/sec: 1507.96 - lr: 0.000026 - momentum: 0.000000
2023-10-25 10:58:35,457 epoch 3 - iter 462/1546 - loss 0.05789807 - time (sec): 23.83 - samples/sec: 1501.32 - lr: 0.000026 - momentum: 0.000000
2023-10-25 10:58:43,631 epoch 3 - iter 616/1546 - loss 0.05366765 - time (sec): 32.00 - samples/sec: 1510.95 - lr: 0.000025 - momentum: 0.000000
2023-10-25 10:58:51,877 epoch 3 - iter 770/1546 - loss 0.05391996 - time (sec): 40.25 - samples/sec: 1503.99 - lr: 0.000025 - momentum: 0.000000
2023-10-25 10:59:00,343 epoch 3 - iter 924/1546 - loss 0.05489213 - time (sec): 48.71 - samples/sec: 1510.07 - lr: 0.000025 - momentum: 0.000000
2023-10-25 10:59:08,613 epoch 3 - iter 1078/1546 - loss 0.05440378 - time (sec): 56.98 - samples/sec: 1510.99 - lr: 0.000024 - momentum: 0.000000
2023-10-25 10:59:16,017 epoch 3 - iter 1232/1546 - loss 0.05523524 - time (sec): 64.39 - samples/sec: 1533.04 - lr: 0.000024 - momentum: 0.000000
2023-10-25 10:59:23,782 epoch 3 - iter 1386/1546 - loss 0.05498877 - time (sec): 72.15 - samples/sec: 1546.14 - lr: 0.000024 - momentum: 0.000000
2023-10-25 10:59:31,809 epoch 3 - iter 1540/1546 - loss 0.05337318 - time (sec): 80.18 - samples/sec: 1545.06 - lr: 0.000023 - momentum: 0.000000
2023-10-25 10:59:32,114 ----------------------------------------------------------------------------------------------------
2023-10-25 10:59:32,114 EPOCH 3 done: loss 0.0533 - lr: 0.000023
2023-10-25 10:59:34,713 DEV : loss 0.09182113409042358 - f1-score (micro avg) 0.7679
2023-10-25 10:59:34,735 saving best model
2023-10-25 10:59:35,483 ----------------------------------------------------------------------------------------------------
2023-10-25 10:59:43,457 epoch 4 - iter 154/1546 - loss 0.02507648 - time (sec): 7.97 - samples/sec: 1565.78 - lr: 0.000023 - momentum: 0.000000
2023-10-25 10:59:51,780 epoch 4 - iter 308/1546 - loss 0.03175758 - time (sec): 16.29 - samples/sec: 1550.68 - lr: 0.000023 - momentum: 0.000000
2023-10-25 10:59:59,992 epoch 4 - iter 462/1546 - loss 0.03194911 - time (sec): 24.51 - samples/sec: 1540.95 - lr: 0.000022 - momentum: 0.000000
2023-10-25 11:00:08,358 epoch 4 - iter 616/1546 - loss 0.03245614 - time (sec): 32.87 - samples/sec: 1530.37 - lr: 0.000022 - momentum: 0.000000
2023-10-25 11:00:16,776 epoch 4 - iter 770/1546 - loss 0.03363787 - time (sec): 41.29 - samples/sec: 1507.55 - lr: 0.000022 - momentum: 0.000000
2023-10-25 11:00:25,086 epoch 4 - iter 924/1546 - loss 0.03411037 - time (sec): 49.60 - samples/sec: 1494.55 - lr: 0.000021 - momentum: 0.000000
2023-10-25 11:00:33,183 epoch 4 - iter 1078/1546 - loss 0.03460341 - time (sec): 57.70 - samples/sec: 1505.14 - lr: 0.000021 - momentum: 0.000000
2023-10-25 11:00:41,566 epoch 4 - iter 1232/1546 - loss 0.03464367 - time (sec): 66.08 - samples/sec: 1507.10 - lr: 0.000021 - momentum: 0.000000
2023-10-25 11:00:50,390 epoch 4 - iter 1386/1546 - loss 0.03534451 - time (sec): 74.90 - samples/sec: 1499.75 - lr: 0.000020 - momentum: 0.000000
2023-10-25 11:00:59,048 epoch 4 - iter 1540/1546 - loss 0.03677995 - time (sec): 83.56 - samples/sec: 1481.68 - lr: 0.000020 - momentum: 0.000000
2023-10-25 11:00:59,372 ----------------------------------------------------------------------------------------------------
2023-10-25 11:00:59,372 EPOCH 4 done: loss 0.0367 - lr: 0.000020
2023-10-25 11:01:02,425 DEV : loss 0.09009607881307602 - f1-score (micro avg) 0.7484
2023-10-25 11:01:02,448 ----------------------------------------------------------------------------------------------------
2023-10-25 11:01:11,044 epoch 5 - iter 154/1546 - loss 0.02548701 - time (sec): 8.59 - samples/sec: 1432.16 - lr: 0.000020 - momentum: 0.000000
2023-10-25 11:01:19,545 epoch 5 - iter 308/1546 - loss 0.02652036 - time (sec): 17.10 - samples/sec: 1443.33 - lr: 0.000019 - momentum: 0.000000
2023-10-25 11:01:28,336 epoch 5 - iter 462/1546 - loss 0.02539993 - time (sec): 25.89 - samples/sec: 1452.56 - lr: 0.000019 - momentum: 0.000000
2023-10-25 11:01:36,776 epoch 5 - iter 616/1546 - loss 0.02612452 - time (sec): 34.33 - samples/sec: 1449.22 - lr: 0.000019 - momentum: 0.000000
2023-10-25 11:01:45,209 epoch 5 - iter 770/1546 - loss 0.02524168 - time (sec): 42.76 - samples/sec: 1466.66 - lr: 0.000018 - momentum: 0.000000
2023-10-25 11:01:53,861 epoch 5 - iter 924/1546 - loss 0.02572133 - time (sec): 51.41 - samples/sec: 1467.85 - lr: 0.000018 - momentum: 0.000000
2023-10-25 11:02:02,324 epoch 5 - iter 1078/1546 - loss 0.02424334 - time (sec): 59.87 - samples/sec: 1473.27 - lr: 0.000018 - momentum: 0.000000
2023-10-25 11:02:11,012 epoch 5 - iter 1232/1546 - loss 0.02389485 - time (sec): 68.56 - samples/sec: 1454.57 - lr: 0.000017 - momentum: 0.000000
2023-10-25 11:02:19,339 epoch 5 - iter 1386/1546 - loss 0.02362872 - time (sec): 76.89 - samples/sec: 1460.40 - lr: 0.000017 - momentum: 0.000000
2023-10-25 11:02:27,570 epoch 5 - iter 1540/1546 - loss 0.02429264 - time (sec): 85.12 - samples/sec: 1454.78 - lr: 0.000017 - momentum: 0.000000
2023-10-25 11:02:27,893 ----------------------------------------------------------------------------------------------------
2023-10-25 11:02:27,894 EPOCH 5 done: loss 0.0243 - lr: 0.000017
2023-10-25 11:02:30,998 DEV : loss 0.10089725255966187 - f1-score (micro avg) 0.7832
2023-10-25 11:02:31,025 saving best model
2023-10-25 11:02:31,719 ----------------------------------------------------------------------------------------------------
2023-10-25 11:02:40,219 epoch 6 - iter 154/1546 - loss 0.02003037 - time (sec): 8.50 - samples/sec: 1487.59 - lr: 0.000016 - momentum: 0.000000
2023-10-25 11:02:48,637 epoch 6 - iter 308/1546 - loss 0.02358693 - time (sec): 16.92 - samples/sec: 1486.16 - lr: 0.000016 - momentum: 0.000000
2023-10-25 11:02:57,000 epoch 6 - iter 462/1546 - loss 0.02117204 - time (sec): 25.28 - samples/sec: 1460.82 - lr: 0.000016 - momentum: 0.000000
2023-10-25 11:03:05,526 epoch 6 - iter 616/1546 - loss 0.02069151 - time (sec): 33.80 - samples/sec: 1472.70 - lr: 0.000015 - momentum: 0.000000
2023-10-25 11:03:13,317 epoch 6 - iter 770/1546 - loss 0.02110916 - time (sec): 41.60 - samples/sec: 1516.05 - lr: 0.000015 - momentum: 0.000000
2023-10-25 11:03:20,828 epoch 6 - iter 924/1546 - loss 0.01896075 - time (sec): 49.11 - samples/sec: 1543.87 - lr: 0.000015 - momentum: 0.000000
2023-10-25 11:03:28,603 epoch 6 - iter 1078/1546 - loss 0.01988098 - time (sec): 56.88 - samples/sec: 1544.66 - lr: 0.000014 - momentum: 0.000000
2023-10-25 11:03:36,263 epoch 6 - iter 1232/1546 - loss 0.01957224 - time (sec): 64.54 - samples/sec: 1543.12 - lr: 0.000014 - momentum: 0.000000
2023-10-25 11:03:43,921 epoch 6 - iter 1386/1546 - loss 0.01892299 - time (sec): 72.20 - samples/sec: 1547.68 - lr: 0.000014 - momentum: 0.000000
2023-10-25 11:03:51,944 epoch 6 - iter 1540/1546 - loss 0.01883330 - time (sec): 80.22 - samples/sec: 1544.46 - lr: 0.000013 - momentum: 0.000000
2023-10-25 11:03:52,223 ----------------------------------------------------------------------------------------------------
2023-10-25 11:03:52,223 EPOCH 6 done: loss 0.0188 - lr: 0.000013
2023-10-25 11:03:54,800 DEV : loss 0.11670850217342377 - f1-score (micro avg) 0.7555
2023-10-25 11:03:54,821 ----------------------------------------------------------------------------------------------------
2023-10-25 11:04:02,733 epoch 7 - iter 154/1546 - loss 0.01309731 - time (sec): 7.91 - samples/sec: 1610.31 - lr: 0.000013 - momentum: 0.000000
2023-10-25 11:04:10,327 epoch 7 - iter 308/1546 - loss 0.01392533 - time (sec): 15.50 - samples/sec: 1602.90 - lr: 0.000013 - momentum: 0.000000
2023-10-25 11:04:17,838 epoch 7 - iter 462/1546 - loss 0.01201188 - time (sec): 23.02 - samples/sec: 1683.89 - lr: 0.000012 - momentum: 0.000000
2023-10-25 11:04:25,232 epoch 7 - iter 616/1546 - loss 0.01294140 - time (sec): 30.41 - samples/sec: 1631.46 - lr: 0.000012 - momentum: 0.000000
2023-10-25 11:04:32,693 epoch 7 - iter 770/1546 - loss 0.01266465 - time (sec): 37.87 - samples/sec: 1633.04 - lr: 0.000012 - momentum: 0.000000
2023-10-25 11:04:40,354 epoch 7 - iter 924/1546 - loss 0.01178168 - time (sec): 45.53 - samples/sec: 1636.18 - lr: 0.000011 - momentum: 0.000000
2023-10-25 11:04:48,093 epoch 7 - iter 1078/1546 - loss 0.01250303 - time (sec): 53.27 - samples/sec: 1612.54 - lr: 0.000011 - momentum: 0.000000
2023-10-25 11:04:55,739 epoch 7 - iter 1232/1546 - loss 0.01293497 - time (sec): 60.92 - samples/sec: 1606.69 - lr: 0.000011 - momentum: 0.000000
2023-10-25 11:05:03,548 epoch 7 - iter 1386/1546 - loss 0.01307263 - time (sec): 68.72 - samples/sec: 1614.44 - lr: 0.000010 - momentum: 0.000000
2023-10-25 11:05:11,371 epoch 7 - iter 1540/1546 - loss 0.01279875 - time (sec): 76.55 - samples/sec: 1616.14 - lr: 0.000010 - momentum: 0.000000
2023-10-25 11:05:11,669 ----------------------------------------------------------------------------------------------------
2023-10-25 11:05:11,669 EPOCH 7 done: loss 0.0128 - lr: 0.000010
2023-10-25 11:05:15,197 DEV : loss 0.12019308656454086 - f1-score (micro avg) 0.7705
2023-10-25 11:05:15,216 ----------------------------------------------------------------------------------------------------
2023-10-25 11:05:23,412 epoch 8 - iter 154/1546 - loss 0.00807042 - time (sec): 8.19 - samples/sec: 1504.53 - lr: 0.000010 - momentum: 0.000000
2023-10-25 11:05:31,450 epoch 8 - iter 308/1546 - loss 0.00746783 - time (sec): 16.23 - samples/sec: 1529.78 - lr: 0.000009 - momentum: 0.000000
2023-10-25 11:05:39,597 epoch 8 - iter 462/1546 - loss 0.00860174 - time (sec): 24.38 - samples/sec: 1492.71 - lr: 0.000009 - momentum: 0.000000
2023-10-25 11:05:47,486 epoch 8 - iter 616/1546 - loss 0.00814822 - time (sec): 32.27 - samples/sec: 1505.31 - lr: 0.000009 - momentum: 0.000000
2023-10-25 11:05:55,271 epoch 8 - iter 770/1546 - loss 0.00790389 - time (sec): 40.05 - samples/sec: 1526.31 - lr: 0.000008 - momentum: 0.000000
2023-10-25 11:06:03,085 epoch 8 - iter 924/1546 - loss 0.00841588 - time (sec): 47.87 - samples/sec: 1555.20 - lr: 0.000008 - momentum: 0.000000
2023-10-25 11:06:10,637 epoch 8 - iter 1078/1546 - loss 0.00891299 - time (sec): 55.42 - samples/sec: 1569.17 - lr: 0.000008 - momentum: 0.000000
2023-10-25 11:06:18,161 epoch 8 - iter 1232/1546 - loss 0.00897866 - time (sec): 62.94 - samples/sec: 1575.65 - lr: 0.000007 - momentum: 0.000000
2023-10-25 11:06:26,204 epoch 8 - iter 1386/1546 - loss 0.00922351 - time (sec): 70.99 - samples/sec: 1566.90 - lr: 0.000007 - momentum: 0.000000
2023-10-25 11:06:34,371 epoch 8 - iter 1540/1546 - loss 0.00871974 - time (sec): 79.15 - samples/sec: 1563.32 - lr: 0.000007 - momentum: 0.000000
2023-10-25 11:06:34,699 ----------------------------------------------------------------------------------------------------
2023-10-25 11:06:34,700 EPOCH 8 done: loss 0.0087 - lr: 0.000007
2023-10-25 11:06:37,436 DEV : loss 0.11731629073619843 - f1-score (micro avg) 0.7885
2023-10-25 11:06:37,454 saving best model
2023-10-25 11:06:38,144 ----------------------------------------------------------------------------------------------------
2023-10-25 11:06:45,643 epoch 9 - iter 154/1546 - loss 0.00438247 - time (sec): 7.50 - samples/sec: 1680.48 - lr: 0.000006 - momentum: 0.000000
2023-10-25 11:06:52,902 epoch 9 - iter 308/1546 - loss 0.00372726 - time (sec): 14.76 - samples/sec: 1696.09 - lr: 0.000006 - momentum: 0.000000
2023-10-25 11:07:00,237 epoch 9 - iter 462/1546 - loss 0.00385813 - time (sec): 22.09 - samples/sec: 1693.20 - lr: 0.000006 - momentum: 0.000000
2023-10-25 11:07:07,380 epoch 9 - iter 616/1546 - loss 0.00377794 - time (sec): 29.23 - samples/sec: 1723.44 - lr: 0.000005 - momentum: 0.000000
2023-10-25 11:07:14,702 epoch 9 - iter 770/1546 - loss 0.00460608 - time (sec): 36.55 - samples/sec: 1714.33 - lr: 0.000005 - momentum: 0.000000
2023-10-25 11:07:22,280 epoch 9 - iter 924/1546 - loss 0.00436436 - time (sec): 44.13 - samples/sec: 1698.21 - lr: 0.000005 - momentum: 0.000000
2023-10-25 11:07:29,621 epoch 9 - iter 1078/1546 - loss 0.00482249 - time (sec): 51.47 - samples/sec: 1689.03 - lr: 0.000004 - momentum: 0.000000
2023-10-25 11:07:36,843 epoch 9 - iter 1232/1546 - loss 0.00450838 - time (sec): 58.70 - samples/sec: 1687.18 - lr: 0.000004 - momentum: 0.000000
2023-10-25 11:07:44,631 epoch 9 - iter 1386/1546 - loss 0.00440336 - time (sec): 66.48 - samples/sec: 1687.63 - lr: 0.000004 - momentum: 0.000000
2023-10-25 11:07:52,067 epoch 9 - iter 1540/1546 - loss 0.00411316 - time (sec): 73.92 - samples/sec: 1677.25 - lr: 0.000003 - momentum: 0.000000
2023-10-25 11:07:52,352 ----------------------------------------------------------------------------------------------------
2023-10-25 11:07:52,352 EPOCH 9 done: loss 0.0041 - lr: 0.000003
2023-10-25 11:07:54,893 DEV : loss 0.1307971030473709 - f1-score (micro avg) 0.7692
2023-10-25 11:07:54,916 ----------------------------------------------------------------------------------------------------
2023-10-25 11:08:02,148 epoch 10 - iter 154/1546 - loss 0.00204746 - time (sec): 7.23 - samples/sec: 1637.33 - lr: 0.000003 - momentum: 0.000000
2023-10-25 11:08:09,285 epoch 10 - iter 308/1546 - loss 0.00355914 - time (sec): 14.37 - samples/sec: 1636.15 - lr: 0.000003 - momentum: 0.000000
2023-10-25 11:08:16,545 epoch 10 - iter 462/1546 - loss 0.00381047 - time (sec): 21.63 - samples/sec: 1639.14 - lr: 0.000002 - momentum: 0.000000
2023-10-25 11:08:23,743 epoch 10 - iter 616/1546 - loss 0.00367323 - time (sec): 28.83 - samples/sec: 1655.86 - lr: 0.000002 - momentum: 0.000000
2023-10-25 11:08:31,133 epoch 10 - iter 770/1546 - loss 0.00339860 - time (sec): 36.22 - samples/sec: 1636.15 - lr: 0.000002 - momentum: 0.000000
2023-10-25 11:08:38,585 epoch 10 - iter 924/1546 - loss 0.00310024 - time (sec): 43.67 - samples/sec: 1652.22 - lr: 0.000001 - momentum: 0.000000
2023-10-25 11:08:46,049 epoch 10 - iter 1078/1546 - loss 0.00291881 - time (sec): 51.13 - samples/sec: 1664.18 - lr: 0.000001 - momentum: 0.000000
2023-10-25 11:08:53,576 epoch 10 - iter 1232/1546 - loss 0.00282386 - time (sec): 58.66 - samples/sec: 1675.15 - lr: 0.000001 - momentum: 0.000000
2023-10-25 11:09:01,133 epoch 10 - iter 1386/1546 - loss 0.00272207 - time (sec): 66.22 - samples/sec: 1680.98 - lr: 0.000000 - momentum: 0.000000
2023-10-25 11:09:08,863 epoch 10 - iter 1540/1546 - loss 0.00284420 - time (sec): 73.95 - samples/sec: 1671.06 - lr: 0.000000 - momentum: 0.000000
2023-10-25 11:09:09,151 ----------------------------------------------------------------------------------------------------
2023-10-25 11:09:09,151 EPOCH 10 done: loss 0.0029 - lr: 0.000000
2023-10-25 11:09:12,044 DEV : loss 0.13086602091789246 - f1-score (micro avg) 0.7695
2023-10-25 11:09:12,513 ----------------------------------------------------------------------------------------------------
2023-10-25 11:09:12,514 Loading model from best epoch ...
2023-10-25 11:09:14,434 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
2023-10-25 11:09:23,205
Results:
- F-score (micro) 0.8064
- F-score (macro) 0.7186
- Accuracy 0.6987
By class:
precision recall f1-score support
LOC 0.8493 0.8520 0.8507 946
BUILDING 0.6020 0.6378 0.6194 185
STREET 0.7347 0.6429 0.6857 56
micro avg 0.8040 0.8088 0.8064 1187
macro avg 0.7287 0.7109 0.7186 1187
weighted avg 0.8054 0.8088 0.8068 1187
2023-10-25 11:09:23,205 ----------------------------------------------------------------------------------------------------
|