---
library_name: transformers
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: vit-weldclassifyv4
  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: 0.9244604316546763
---

<!-- 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-weldclassifyv4

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: 0.3162
- Accuracy: 0.9245

## 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: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 15
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch   | Step | Validation Loss | Accuracy |
|:-------------:|:-------:|:----:|:---------------:|:--------:|
| 0.8146        | 0.6410  | 100  | 0.7349          | 0.6835   |
| 0.6048        | 1.2821  | 200  | 0.6821          | 0.6978   |
| 0.4796        | 1.9231  | 300  | 0.4833          | 0.8129   |
| 0.4532        | 2.5641  | 400  | 0.5380          | 0.8022   |
| 0.1242        | 3.2051  | 500  | 0.3899          | 0.8741   |
| 0.124         | 3.8462  | 600  | 0.5237          | 0.8273   |
| 0.1239        | 4.4872  | 700  | 0.4221          | 0.8849   |
| 0.0785        | 5.1282  | 800  | 0.3683          | 0.9137   |
| 0.093         | 5.7692  | 900  | 0.6376          | 0.8597   |
| 0.0056        | 6.4103  | 1000 | 0.3162          | 0.9245   |
| 0.0472        | 7.0513  | 1100 | 0.5225          | 0.8885   |
| 0.0234        | 7.6923  | 1200 | 0.6096          | 0.8597   |
| 0.0354        | 8.3333  | 1300 | 0.5520          | 0.8777   |
| 0.026         | 8.9744  | 1400 | 0.4938          | 0.8993   |
| 0.002         | 9.6154  | 1500 | 0.4350          | 0.9173   |
| 0.0021        | 10.2564 | 1600 | 0.4224          | 0.9173   |
| 0.0016        | 10.8974 | 1700 | 0.3838          | 0.9281   |
| 0.0014        | 11.5385 | 1800 | 0.3943          | 0.9281   |
| 0.0013        | 12.1795 | 1900 | 0.4012          | 0.9281   |
| 0.0012        | 12.8205 | 2000 | 0.4067          | 0.9281   |
| 0.0011        | 13.4615 | 2100 | 0.4101          | 0.9281   |
| 0.0011        | 14.1026 | 2200 | 0.4124          | 0.9281   |
| 0.0012        | 14.7436 | 2300 | 0.4136          | 0.9281   |


### Framework versions

- Transformers 4.44.2
- Pytorch 2.5.0+cu121
- Datasets 3.1.0
- Tokenizers 0.19.1