---
license: apache-2.0
base_model: microsoft/beit-base-patch16-224
tags:
- generated_from_trainer
datasets:
- imagefolder
metrics:
- accuracy
model-index:
- name: beit-base-patch16-224-OT
  results:
  - task:
      name: Image Classification
      type: image-classification
    dataset:
      name: imagefolder
      type: imagefolder
      config: default
      split: validation
      args: default
    metrics:
    - name: Accuracy
      type: accuracy
      value: 0.8317757009345794
---

<!-- 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. -->

# beit-base-patch16-224-OT

This model is a fine-tuned version of [microsoft/beit-base-patch16-224](https://huggingface.co/microsoft/beit-base-patch16-224) on the imagefolder dataset.
It achieves the following results on the evaluation set:
- Loss: 0.5393
- Accuracy: 0.8318

## 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: 5e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 40

### Training results

| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
| No log        | 1.0   | 8    | 0.6887          | 0.5888   |
| 0.692         | 2.0   | 16   | 0.6782          | 0.5888   |
| 0.6801        | 3.0   | 24   | 0.6669          | 0.5888   |
| 0.6696        | 4.0   | 32   | 0.6644          | 0.5888   |
| 0.6607        | 5.0   | 40   | 0.6661          | 0.6636   |
| 0.6607        | 6.0   | 48   | 0.6241          | 0.6542   |
| 0.6341        | 7.0   | 56   | 0.6235          | 0.6542   |
| 0.6089        | 8.0   | 64   | 0.6088          | 0.6916   |
| 0.6095        | 9.0   | 72   | 0.5912          | 0.6916   |
| 0.5632        | 10.0  | 80   | 0.6607          | 0.6355   |
| 0.5632        | 11.0  | 88   | 0.5793          | 0.7009   |
| 0.5418        | 12.0  | 96   | 0.5953          | 0.6822   |
| 0.5336        | 13.0  | 104  | 0.5793          | 0.7103   |
| 0.5102        | 14.0  | 112  | 0.5292          | 0.7196   |
| 0.4762        | 15.0  | 120  | 0.6558          | 0.7009   |
| 0.4762        | 16.0  | 128  | 0.5371          | 0.7103   |
| 0.544         | 17.0  | 136  | 0.5401          | 0.7570   |
| 0.4256        | 18.0  | 144  | 0.4927          | 0.7944   |
| 0.4082        | 19.0  | 152  | 0.5801          | 0.7383   |
| 0.4014        | 20.0  | 160  | 0.5823          | 0.7383   |
| 0.4014        | 21.0  | 168  | 0.5393          | 0.7757   |
| 0.3483        | 22.0  | 176  | 0.5941          | 0.7103   |
| 0.3121        | 23.0  | 184  | 0.5569          | 0.7383   |
| 0.3484        | 24.0  | 192  | 0.5975          | 0.7664   |
| 0.263         | 25.0  | 200  | 0.6544          | 0.7570   |
| 0.263         | 26.0  | 208  | 0.5744          | 0.7757   |
| 0.2633        | 27.0  | 216  | 0.6095          | 0.7664   |
| 0.2935        | 28.0  | 224  | 0.5286          | 0.7664   |
| 0.2332        | 29.0  | 232  | 0.6028          | 0.7850   |
| 0.2314        | 30.0  | 240  | 0.5935          | 0.7944   |
| 0.2314        | 31.0  | 248  | 0.5393          | 0.8318   |
| 0.202         | 32.0  | 256  | 0.5556          | 0.8224   |
| 0.2127        | 33.0  | 264  | 0.5913          | 0.8037   |
| 0.2035        | 34.0  | 272  | 0.5337          | 0.8037   |
| 0.2618        | 35.0  | 280  | 0.6221          | 0.8037   |
| 0.2618        | 36.0  | 288  | 0.5090          | 0.8318   |
| 0.217         | 37.0  | 296  | 0.5649          | 0.8224   |
| 0.2111        | 38.0  | 304  | 0.5683          | 0.8131   |
| 0.2085        | 39.0  | 312  | 0.5398          | 0.8224   |
| 0.1912        | 40.0  | 320  | 0.5548          | 0.8224   |


### Framework versions

- Transformers 4.36.2
- Pytorch 2.1.2+cu118
- Datasets 2.16.1
- Tokenizers 0.15.0