---
license: apache-2.0
tags:
- generated_from_trainer
datasets:
- imagefolder
metrics:
- accuracy
model-index:
- name: vc-bantai-vit-withoutAMBI-adunest-v1
  results:
  - task:
      name: Image Classification
      type: image-classification
    dataset:
      name: imagefolder
      type: imagefolder
      args: Violation-Classification---Raw-9
    metrics:
    - name: Accuracy
      type: accuracy
      value: 0.8039014373716632
---

<!-- 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. -->

# vc-bantai-vit-withoutAMBI-adunest-v1

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 imagefolder dataset.
It achieves the following results on the evaluation set:
- Loss: 0.7736
- Accuracy: 0.8039

## 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: 0.0005
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 200
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
| No log        | 0.4   | 100  | 1.2037          | 0.4081   |
| No log        | 0.8   | 200  | 0.9935          | 0.4410   |
| No log        | 1.2   | 300  | 0.6461          | 0.6915   |
| No log        | 1.61  | 400  | 0.4938          | 0.7705   |
| 0.357         | 2.01  | 500  | 0.4602          | 0.7844   |
| 0.357         | 2.41  | 600  | 0.5220          | 0.7295   |
| 0.357         | 2.81  | 700  | 0.4665          | 0.7782   |
| 0.357         | 3.21  | 800  | 0.4440          | 0.8301   |
| 0.357         | 3.61  | 900  | 0.5122          | 0.7177   |
| 0.2437        | 4.02  | 1000 | 0.6155          | 0.7320   |
| 0.2437        | 4.42  | 1100 | 0.5802          | 0.7685   |
| 0.2437        | 4.82  | 1200 | 0.4709          | 0.8029   |
| 0.2437        | 5.22  | 1300 | 0.4694          | 0.8352   |
| 0.2437        | 5.62  | 1400 | 0.4652          | 0.8203   |
| 0.1841        | 6.02  | 1500 | 0.5424          | 0.7649   |
| 0.1841        | 6.43  | 1600 | 0.4616          | 0.8060   |
| 0.1841        | 6.83  | 1700 | 0.3569          | 0.8547   |
| 0.1841        | 7.23  | 1800 | 0.3652          | 0.8737   |
| 0.1841        | 7.63  | 1900 | 0.7778          | 0.7438   |
| 0.1328        | 8.03  | 2000 | 0.5460          | 0.8162   |
| 0.1328        | 8.43  | 2100 | 0.8070          | 0.7767   |
| 0.1328        | 8.84  | 2200 | 0.6873          | 0.7798   |
| 0.1328        | 9.24  | 2300 | 0.8943          | 0.7782   |
| 0.1328        | 9.64  | 2400 | 0.5378          | 0.8552   |
| 0.1059        | 10.04 | 2500 | 0.7081          | 0.8070   |
| 0.1059        | 10.44 | 2600 | 0.9941          | 0.7012   |
| 0.1059        | 10.84 | 2700 | 0.9152          | 0.7900   |
| 0.1059        | 11.24 | 2800 | 0.7494          | 0.7736   |
| 0.1059        | 11.65 | 2900 | 0.7681          | 0.7870   |
| 0.081         | 12.05 | 3000 | 0.7736          | 0.8039   |


### Framework versions

- Transformers 4.20.1
- Pytorch 1.12.0+cu113
- Datasets 2.3.2
- Tokenizers 0.12.1