---
library_name: transformers
license: apache-2.0
base_model: google/vit-base-patch16-224-in21k
tags:
- image-classification
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: vit-cropped-faces
  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-cropped-faces

This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the emigomez/vit-cropped-faces dataset.
It achieves the following results on the evaluation set:
- Loss: 0.0109
- Accuracy: 1.0

## 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: 0.0002
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 20
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch  | Step | Validation Loss | Accuracy |
|:-------------:|:------:|:----:|:---------------:|:--------:|
| 0.0254        | 3.125  | 100  | 0.0136          | 1.0      |
| 0.0053        | 6.25   | 200  | 0.0109          | 1.0      |
| 0.0033        | 9.375  | 300  | 0.0139          | 1.0      |
| 0.0025        | 12.5   | 400  | 0.0128          | 1.0      |
| 0.0021        | 15.625 | 500  | 0.0122          | 1.0      |
| 0.0019        | 18.75  | 600  | 0.0120          | 1.0      |


### Framework versions

- Transformers 4.48.3
- Pytorch 2.5.1+cu124
- Datasets 3.3.2
- Tokenizers 0.21.0