<!-- 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. -->
perioli_vgm_v3.0
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: 0.0975
- Precision: 0.6585
- Recall: 0.5912
- F1: 0.6231
- Accuracy: 0.9881
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: 2500
Training results
Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
---|---|---|---|---|---|---|---|
No log | 0.94 | 100 | 0.1390 | 0.1538 | 0.0292 | 0.0491 | 0.9768 |
No log | 1.89 | 200 | 0.1011 | 0.4674 | 0.3139 | 0.3755 | 0.9820 |
No log | 2.83 | 300 | 0.0857 | 0.4896 | 0.3431 | 0.4034 | 0.9836 |
No log | 3.77 | 400 | 0.0836 | 0.6053 | 0.5036 | 0.5498 | 0.9872 |
0.0974 | 4.72 | 500 | 0.0751 | 0.5369 | 0.5839 | 0.5594 | 0.9858 |
0.0974 | 5.66 | 600 | 0.0808 | 0.6493 | 0.6350 | 0.6421 | 0.9872 |
0.0974 | 6.6 | 700 | 0.0760 | 0.6231 | 0.5912 | 0.6067 | 0.9876 |
0.0974 | 7.55 | 800 | 0.0906 | 0.5942 | 0.5985 | 0.5964 | 0.9843 |
0.0974 | 8.49 | 900 | 0.0831 | 0.6693 | 0.6204 | 0.6439 | 0.9872 |
0.0169 | 9.43 | 1000 | 0.0907 | 0.6538 | 0.6204 | 0.6367 | 0.9858 |
0.0169 | 10.38 | 1100 | 0.0834 | 0.6535 | 0.6058 | 0.6288 | 0.9879 |
0.0169 | 11.32 | 1200 | 0.0861 | 0.6721 | 0.5985 | 0.6332 | 0.9883 |
0.0169 | 12.26 | 1300 | 0.0930 | 0.6 | 0.5693 | 0.5843 | 0.9849 |
0.0169 | 13.21 | 1400 | 0.0976 | 0.625 | 0.5839 | 0.6038 | 0.9861 |
0.0058 | 14.15 | 1500 | 0.0933 | 0.5827 | 0.5912 | 0.5870 | 0.9863 |
0.0058 | 15.09 | 1600 | 0.0892 | 0.6429 | 0.5912 | 0.6160 | 0.9885 |
0.0058 | 16.04 | 1700 | 0.0889 | 0.6613 | 0.5985 | 0.6284 | 0.9888 |
0.0058 | 16.98 | 1800 | 0.0918 | 0.6723 | 0.5839 | 0.625 | 0.9881 |
0.0058 | 17.92 | 1900 | 0.0924 | 0.6512 | 0.6131 | 0.6316 | 0.9879 |
0.003 | 18.87 | 2000 | 0.0958 | 0.6803 | 0.6058 | 0.6409 | 0.9881 |
0.003 | 19.81 | 2100 | 0.0941 | 0.6923 | 0.5912 | 0.6378 | 0.9888 |
0.003 | 20.75 | 2200 | 0.0966 | 0.6349 | 0.5839 | 0.6084 | 0.9876 |
0.003 | 21.7 | 2300 | 0.0973 | 0.6423 | 0.5766 | 0.6077 | 0.9879 |
0.003 | 22.64 | 2400 | 0.0965 | 0.6585 | 0.5912 | 0.6231 | 0.9881 |
0.0015 | 23.58 | 2500 | 0.0975 | 0.6585 | 0.5912 | 0.6231 | 0.9881 |
Framework versions
- Transformers 4.27.0.dev0
- Pytorch 1.13.1+cu116
- Datasets 2.2.2
- Tokenizers 0.13.2