<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
rhenus_v3.5
This model is a fine-tuned version of microsoft/layoutlmv3-base on the sroie dataset. It achieves the following results on the evaluation set:
- Loss: 0.1811
- Precision: 0.8134
- Recall: 0.8412
- F1: 0.8271
- Accuracy: 0.9642
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 1e-05
- train_batch_size: 2
- eval_batch_size: 2
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- training_steps: 5500
Training results
Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
---|---|---|---|---|---|---|---|
No log | 0.42 | 100 | 0.8800 | 0.0256 | 0.0010 | 0.0019 | 0.8497 |
No log | 0.83 | 200 | 0.7576 | 0.0256 | 0.0010 | 0.0019 | 0.8497 |
No log | 1.25 | 300 | 0.6665 | 0.1045 | 0.0141 | 0.0248 | 0.8539 |
No log | 1.67 | 400 | 0.5845 | 0.2211 | 0.0673 | 0.1032 | 0.8669 |
0.7141 | 2.08 | 500 | 0.5312 | 0.2549 | 0.1166 | 0.16 | 0.8778 |
0.7141 | 2.5 | 600 | 0.4970 | 0.3097 | 0.1759 | 0.2244 | 0.8772 |
0.7141 | 2.92 | 700 | 0.4455 | 0.3603 | 0.2281 | 0.2794 | 0.8909 |
0.7141 | 3.33 | 800 | 0.3984 | 0.4536 | 0.3146 | 0.3715 | 0.9069 |
0.7141 | 3.75 | 900 | 0.3447 | 0.4848 | 0.3839 | 0.4285 | 0.9149 |
0.361 | 4.17 | 1000 | 0.3283 | 0.5557 | 0.5618 | 0.5587 | 0.9238 |
0.361 | 4.58 | 1100 | 0.3005 | 0.6024 | 0.5528 | 0.5765 | 0.9280 |
0.361 | 5.0 | 1200 | 0.2841 | 0.6866 | 0.6010 | 0.6409 | 0.9372 |
0.361 | 5.42 | 1300 | 0.2407 | 0.6731 | 0.6995 | 0.6861 | 0.9448 |
0.361 | 5.83 | 1400 | 0.2486 | 0.7316 | 0.6794 | 0.7045 | 0.9477 |
0.1955 | 6.25 | 1500 | 0.2110 | 0.7038 | 0.7045 | 0.7042 | 0.9490 |
0.1955 | 6.67 | 1600 | 0.2023 | 0.7075 | 0.7367 | 0.7218 | 0.9504 |
0.1955 | 7.08 | 1700 | 0.2137 | 0.6864 | 0.7457 | 0.7148 | 0.9454 |
0.1955 | 7.5 | 1800 | 0.2022 | 0.6903 | 0.7548 | 0.7211 | 0.9496 |
0.1955 | 7.92 | 1900 | 0.1818 | 0.7160 | 0.7678 | 0.7410 | 0.9562 |
0.1152 | 8.33 | 2000 | 0.1980 | 0.7132 | 0.7548 | 0.7334 | 0.9492 |
0.1152 | 8.75 | 2100 | 0.1930 | 0.7386 | 0.7638 | 0.7510 | 0.9519 |
0.1152 | 9.17 | 2200 | 0.1955 | 0.7333 | 0.7930 | 0.7620 | 0.9539 |
0.1152 | 9.58 | 2300 | 0.1815 | 0.7363 | 0.7719 | 0.7537 | 0.9572 |
0.1152 | 10.0 | 2400 | 0.1760 | 0.7905 | 0.8 | 0.7952 | 0.9627 |
0.0777 | 10.42 | 2500 | 0.1756 | 0.7532 | 0.7638 | 0.7585 | 0.9583 |
0.0777 | 10.83 | 2600 | 0.1901 | 0.7779 | 0.7920 | 0.7849 | 0.9595 |
0.0777 | 11.25 | 2700 | 0.1918 | 0.7371 | 0.7920 | 0.7636 | 0.9533 |
0.0777 | 11.67 | 2800 | 0.1744 | 0.8018 | 0.8090 | 0.8054 | 0.9644 |
0.0777 | 12.08 | 2900 | 0.1762 | 0.7805 | 0.8040 | 0.7921 | 0.9612 |
0.0562 | 12.5 | 3000 | 0.1774 | 0.7892 | 0.8090 | 0.7990 | 0.9595 |
0.0562 | 12.92 | 3100 | 0.1770 | 0.7722 | 0.8141 | 0.7926 | 0.9591 |
0.0562 | 13.33 | 3200 | 0.1712 | 0.8249 | 0.8191 | 0.8220 | 0.9661 |
0.0562 | 13.75 | 3300 | 0.1773 | 0.7909 | 0.8211 | 0.8057 | 0.9609 |
0.0562 | 14.17 | 3400 | 0.1964 | 0.7640 | 0.8201 | 0.7911 | 0.9594 |
0.0411 | 14.58 | 3500 | 0.1867 | 0.7797 | 0.8111 | 0.7951 | 0.9618 |
0.0411 | 15.0 | 3600 | 0.1850 | 0.7983 | 0.8271 | 0.8124 | 0.9610 |
0.0411 | 15.42 | 3700 | 0.1747 | 0.7948 | 0.8251 | 0.8097 | 0.9621 |
0.0411 | 15.83 | 3800 | 0.1861 | 0.7813 | 0.8332 | 0.8064 | 0.9588 |
0.0411 | 16.25 | 3900 | 0.1805 | 0.7935 | 0.8342 | 0.8133 | 0.9635 |
0.0313 | 16.67 | 4000 | 0.1910 | 0.7874 | 0.8261 | 0.8063 | 0.9609 |
0.0313 | 17.08 | 4100 | 0.1739 | 0.7856 | 0.8251 | 0.8049 | 0.9620 |
0.0313 | 17.5 | 4200 | 0.1828 | 0.7916 | 0.8362 | 0.8133 | 0.9621 |
0.0313 | 17.92 | 4300 | 0.1850 | 0.8239 | 0.8322 | 0.8280 | 0.9644 |
0.0313 | 18.33 | 4400 | 0.1771 | 0.8252 | 0.8352 | 0.8302 | 0.9644 |
0.0254 | 18.75 | 4500 | 0.1974 | 0.7971 | 0.8291 | 0.8128 | 0.9604 |
0.0254 | 19.17 | 4600 | 0.1865 | 0.8057 | 0.8291 | 0.8172 | 0.9627 |
0.0254 | 19.58 | 4700 | 0.1895 | 0.8197 | 0.8362 | 0.8279 | 0.9638 |
0.0254 | 20.0 | 4800 | 0.1781 | 0.8107 | 0.8352 | 0.8228 | 0.9642 |
0.0254 | 20.42 | 4900 | 0.1751 | 0.8303 | 0.8362 | 0.8332 | 0.9656 |
0.0212 | 20.83 | 5000 | 0.1823 | 0.8209 | 0.8382 | 0.8294 | 0.9650 |
0.0212 | 21.25 | 5100 | 0.1727 | 0.8246 | 0.8412 | 0.8328 | 0.9653 |
0.0212 | 21.67 | 5200 | 0.1758 | 0.7907 | 0.8392 | 0.8142 | 0.9627 |
0.0212 | 22.08 | 5300 | 0.1787 | 0.8104 | 0.8422 | 0.8260 | 0.9642 |
0.0212 | 22.5 | 5400 | 0.1816 | 0.8134 | 0.8412 | 0.8271 | 0.9639 |
0.0184 | 22.92 | 5500 | 0.1811 | 0.8134 | 0.8412 | 0.8271 | 0.9642 |
Framework versions
- Transformers 4.28.0
- Pytorch 2.0.1+cu118
- Datasets 2.2.2
- Tokenizers 0.13.3