---
license: apache-2.0
base_model: google/vit-base-patch16-224-in21k
tags:
- image-classification
- generated_from_trainer
datasets:
- imagefolder
metrics:
- accuracy
model-index:
- name: Action_all_10_class
  results:
  - task:
      name: Image Classification
      type: image-classification
    dataset:
      name: Action_small_dataset
      type: imagefolder
      config: default
      split: train
      args: default
    metrics:
    - name: Accuracy
      type: accuracy
      value: 0.8517382413087935
---

<!-- 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. -->

# Action_all_10_class

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 Action_small_dataset dataset.
It achieves the following results on the evaluation set:
- Loss: 0.4725
- Accuracy: 0.8517

## 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: 0.0001
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
| 1.2411        | 0.36  | 100  | 1.1517          | 0.7546   |
| 0.8932        | 0.72  | 200  | 0.7856          | 0.7975   |
| 0.6907        | 1.08  | 300  | 0.6636          | 0.8221   |
| 0.5841        | 1.43  | 400  | 0.6388          | 0.8160   |
| 0.5425        | 1.79  | 500  | 0.5871          | 0.8436   |
| 0.5929        | 2.15  | 600  | 0.5646          | 0.8211   |
| 0.4406        | 2.51  | 700  | 0.5439          | 0.8405   |
| 0.4541        | 2.87  | 800  | 0.5318          | 0.8415   |
| 0.3835        | 3.23  | 900  | 0.5225          | 0.8344   |
| 0.3924        | 3.58  | 1000 | 0.5515          | 0.8303   |
| 0.5741        | 3.94  | 1100 | 0.5519          | 0.8252   |
| 0.3991        | 4.3   | 1200 | 0.4990          | 0.8446   |
| 0.4732        | 4.66  | 1300 | 0.5336          | 0.8303   |
| 0.3324        | 5.02  | 1400 | 0.5351          | 0.8282   |
| 0.3433        | 5.38  | 1500 | 0.4725          | 0.8517   |
| 0.2187        | 5.73  | 1600 | 0.5042          | 0.8466   |
| 0.2952        | 6.09  | 1700 | 0.5240          | 0.8548   |
| 0.2687        | 6.45  | 1800 | 0.5523          | 0.8364   |
| 0.3111        | 6.81  | 1900 | 0.5304          | 0.8497   |
| 0.2431        | 7.17  | 2000 | 0.5104          | 0.8569   |
| 0.3265        | 7.53  | 2100 | 0.5085          | 0.8691   |
| 0.2595        | 7.89  | 2200 | 0.5015          | 0.8569   |
| 0.1825        | 8.24  | 2300 | 0.4920          | 0.8620   |
| 0.2602        | 8.6   | 2400 | 0.5016          | 0.8620   |
| 0.2628        | 8.96  | 2500 | 0.4746          | 0.8681   |
| 0.1024        | 9.32  | 2600 | 0.4818          | 0.8691   |
| 0.1468        | 9.68  | 2700 | 0.4765          | 0.8681   |


### Framework versions

- Transformers 4.39.3
- Pytorch 2.1.2
- Datasets 2.18.0
- Tokenizers 0.15.2