README.md

---
license: apache-2.0
tags:
- generated_from_trainer
datasets:
- imagefolder
metrics:
- accuracy
- f1
- recall
- precision
model-index:
- name: vit-base-patch16-224-in21k-Intel_Images
  results:
  - task:
      name: Image Classification
      type: image-classification
    dataset:
      name: imagefolder
      type: imagefolder
      config: default
      split: train
      args: default
    metrics:
    - name: Accuracy
      type: accuracy
      value: 0.9486666666666667
language:
- en
pipeline_tag: image-classification
---

# vit-base-patch16-224-in21k-Intel_Images

This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k).

It achieves the following results on the evaluation set:
- Loss: 0.1822
- Accuracy: 0.9487
- F1
  - Weighted: 0.9485
  - Micro: 0.9487
  - Macro: 0.9497
- Recall
  - Weighted: 0.9487
  - Micro: 0.9487
  - Macro: 0.9500
- Precision
  - Weighted: 0.9485
  - Micro: 0.9487
  - Macro: 0.9496

## Model description

This is a multiclass image classification model of different scenery types.

For more information on how it was created, check out the following link: https://github.com/DunnBC22/Vision_Audio_and_Multimodal_Projects/blob/main/Computer%20Vision/Image%20Classification/Multiclass%20Classification/Intel%20Image%20Classification/Intel_ViT.ipynb

## Intended uses & limitations

This model is intended to demonstrate my ability to solve a complex problem using technology.

## Training and evaluation data

Dataset Source: https://www.kaggle.com/datasets/puneet6060/intel-image-classification

_Sample Images From Dataset:_

![Sample Images](https://github.com/DunnBC22/Vision_Audio_and_Multimodal_Projects/raw/main/Computer%20Vision/Image%20Classification/Multiclass%20Classification/Intel%20Image%20Classification/Images/Sample%20Images.png)

## 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: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3

### Training results

| Training Loss | Epoch | Step | Validation Loss | Accuracy | Weighted F1 | Micro F1 | Macro F1 | Weighted Recall | Micro Recall | Macro Recall | Weighted Precision | Micro Precision | Macro Precision |
|:-------------:|:-----:|:----:|:---------------:|:--------:|:-----------:|:--------:|:--------:|:---------------:|:------------:|:------------:|:------------------:|:---------------:|:---------------:|
| 0.2305        | 1.0   | 878  | 0.2362          | 0.9153   | 0.9144      | 0.9153   | 0.9152   | 0.9153          | 0.9153       | 0.9148       | 0.9208             | 0.9153          | 0.9231          |
| 0.1136        | 2.0   | 1756 | 0.1785          | 0.9393   | 0.9391      | 0.9393   | 0.9405   | 0.9393          | 0.9393       | 0.9405       | 0.9391             | 0.9393          | 0.9407          |
| 0.0435        | 3.0   | 2634 | 0.1822          | 0.9487   | 0.9485      | 0.9487   | 0.9497   | 0.9487          | 0.9487       | 0.9500       | 0.9485             | 0.9487          | 0.9496          |

### Framework versions

- Transformers 4.27.4
- Pytorch 2.0.0
- Datasets 2.11.0
- Tokenizers 0.13.3


## License Notice
This model is a fine-tuned derivative of a pretrained model.
Users must comply with the original model license.


## Dataset Notice
This model was fine-tuned on third-party datasets which may have separate licenses or usage restrictions.