<!-- 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. -->
vit-base_rvl_cdip_entropy2_softmax
This model is a fine-tuned version of jordyvl/vit-base_rvl-cdip on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.8809
- Accuracy: 0.8968
- Brier Loss: 0.1890
- Nll: 1.1526
- F1 Micro: 0.8968
- F1 Macro: 0.8969
- Ece: 0.0923
- Aurc: 0.0205
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: 2e-05
- train_batch_size: 64
- eval_batch_size: 64
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Brier Loss | Nll | F1 Micro | F1 Macro | Ece | Aurc |
|---|---|---|---|---|---|---|---|---|---|---|
| 0.3547 | 1.0 | 2500 | 0.7036 | 0.8958 | 0.1806 | 0.9568 | 0.8958 | 0.8955 | 0.0815 | 0.0174 |
| 0.3049 | 2.0 | 5000 | 0.7030 | 0.8972 | 0.1784 | 1.0077 | 0.8972 | 0.8975 | 0.0825 | 0.0168 |
| 0.2103 | 3.0 | 7500 | 0.7465 | 0.8946 | 0.1857 | 1.0229 | 0.8946 | 0.8954 | 0.0883 | 0.0178 |
| 0.1548 | 4.0 | 10000 | 0.7640 | 0.8957 | 0.1860 | 1.0530 | 0.8957 | 0.8960 | 0.0893 | 0.0182 |
| 0.1077 | 5.0 | 12500 | 0.7964 | 0.8955 | 0.1877 | 1.0743 | 0.8955 | 0.8955 | 0.0903 | 0.0182 |
| 0.0742 | 6.0 | 15000 | 0.8253 | 0.8959 | 0.1887 | 1.0996 | 0.8959 | 0.8967 | 0.0919 | 0.0202 |
| 0.0495 | 7.0 | 17500 | 0.8505 | 0.8964 | 0.1884 | 1.1281 | 0.8964 | 0.8963 | 0.0920 | 0.0201 |
| 0.0352 | 8.0 | 20000 | 0.8645 | 0.8964 | 0.1895 | 1.1397 | 0.8964 | 0.8964 | 0.0931 | 0.0207 |
| 0.0235 | 9.0 | 22500 | 0.8733 | 0.8984 | 0.1876 | 1.1365 | 0.8984 | 0.8986 | 0.0914 | 0.0204 |
| 0.0176 | 10.0 | 25000 | 0.8809 | 0.8968 | 0.1890 | 1.1526 | 0.8968 | 0.8969 | 0.0923 | 0.0205 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.13.1.post200
- Datasets 2.9.0
- Tokenizers 0.13.2
