|
--- |
|
license: apache-2.0 |
|
base_model: google/vit-base-patch16-224 |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- precision |
|
- recall |
|
- f1 |
|
- accuracy |
|
model-index: |
|
- name: vit-cxr4 |
|
results: [] |
|
--- |
|
|
|
<!-- 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-cxr4 |
|
|
|
This model is a fine-tuned version of [google/vit-base-patch16-224](https://huggingface.co./google/vit-base-patch16-224) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.3774 |
|
- Precision: 0.8587 |
|
- Recall: 0.9317 |
|
- F1: 0.8937 |
|
- Accuracy: 0.8924 |
|
|
|
## 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: 3e-05 |
|
- train_batch_size: 96 |
|
- eval_batch_size: 64 |
|
- seed: 17 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 6 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy | |
|
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:| |
|
| 0.3151 | 0.31 | 100 | 0.3317 | 0.8152 | 0.9143 | 0.8619 | 0.8552 | |
|
| 0.319 | 0.63 | 200 | 0.3048 | 0.8670 | 0.8514 | 0.8591 | 0.8620 | |
|
| 0.2926 | 0.94 | 300 | 0.2867 | 0.8580 | 0.8662 | 0.8621 | 0.8631 | |
|
| 0.1884 | 1.25 | 400 | 0.2635 | 0.8468 | 0.9381 | 0.8901 | 0.8856 | |
|
| 0.234 | 1.57 | 500 | 0.2639 | 0.8232 | 0.9677 | 0.8896 | 0.8814 | |
|
| 0.2349 | 1.88 | 600 | 0.2478 | 0.8530 | 0.9328 | 0.8911 | 0.8874 | |
|
| 0.1476 | 2.19 | 700 | 0.2560 | 0.8584 | 0.9297 | 0.8926 | 0.8895 | |
|
| 0.1289 | 2.51 | 800 | 0.2698 | 0.8809 | 0.8916 | 0.8862 | 0.8869 | |
|
| 0.1579 | 2.82 | 900 | 0.2614 | 0.8879 | 0.8715 | 0.8796 | 0.8822 | |
|
| 0.0745 | 3.13 | 1000 | 0.2783 | 0.8854 | 0.8905 | 0.8880 | 0.8889 | |
|
| 0.0697 | 3.45 | 1100 | 0.2844 | 0.8893 | 0.8879 | 0.8886 | 0.8900 | |
|
| 0.0602 | 3.76 | 1200 | 0.3213 | 0.8797 | 0.8932 | 0.8864 | 0.8869 | |
|
| 0.0246 | 4.08 | 1300 | 0.3393 | 0.8753 | 0.9096 | 0.8921 | 0.8913 | |
|
| 0.0301 | 4.39 | 1400 | 0.3593 | 0.8644 | 0.9307 | 0.8964 | 0.8937 | |
|
| 0.0348 | 4.7 | 1500 | 0.3804 | 0.8653 | 0.9344 | 0.8986 | 0.8957 | |
|
| 0.011 | 5.02 | 1600 | 0.3897 | 0.8622 | 0.9365 | 0.8978 | 0.8947 | |
|
| 0.0077 | 5.33 | 1700 | 0.4088 | 0.8754 | 0.9180 | 0.8962 | 0.8950 | |
|
| 0.0064 | 5.64 | 1800 | 0.4281 | 0.8780 | 0.9170 | 0.8971 | 0.8960 | |
|
| 0.0031 | 5.96 | 1900 | 0.4289 | 0.8736 | 0.9207 | 0.8965 | 0.8950 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.35.2 |
|
- Pytorch 2.1.0+cu121 |
|
- Datasets 2.16.1 |
|
- Tokenizers 0.15.0 |
|
|
