README.md

---
license: apache-2.0
base_model: google/vit-base-patch16-224-in21k
tags:
- generated_from_trainer
widget:
- src: >-
    https://cdn.discordapp.com/attachments/1120417968032063538/1191101288428097727/1.jpg?ex=65a43684&is=6591c184&hm=aed9f3278325ea30e30557e201adcfc43ce2ce77f2218b5f8f232a26b4ac2985&
- src: >-
    https://cdn.discordapp.com/attachments/1120417968032063538/1191101301698867260/2.jpg?ex=65a43687&is=6591c187&hm=dee873150a2910177be30e5141f008b70ba7f55266e1e8725b422bfe0e6213f8&
metrics:
- accuracy
model-index:
- name: vogue-fashion-collection-15
  results: []
pipeline_tag: image-classification
---

# vogue-fashion-collection-15

## Model description
This model classifies an image into a fashion collection. It is trained on the [tonyassi/vogue-runway-top15-512px](https://huggingface.co/datasets/tonyassi/vogue-runway-top15-512px) dataset and fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k).

Try the [demo](https://huggingface.co/spaces/tonyassi/which-fashion-collection).

## Dataset description
[tonyassi/vogue-runway-top15-512px](https://huggingface.co/datasets/tonyassi/vogue-runway-top15-512px)
- 15 fashion houses
- 1679 collections
- 87,547 images

### How to use
```python
from transformers import pipeline

# Initialize image classification pipeline
pipe = pipeline("image-classification", model="tonyassi/vogue-fashion-collection-15")

# Perform classification
result = pipe('image.png')

# Print results
print(result)
```
## Examples
![image/jpeg](https://cdn-uploads.huggingface.co/production/uploads/648a824a8ca6cf9857d1349c/YWz7ZLk2Oa0xCvuUqVX3O.jpeg)
**fendi,spring 2023 couture**

![image/jpeg](https://cdn-uploads.huggingface.co/production/uploads/648a824a8ca6cf9857d1349c/qRBLjPrbCt0EX181pmu7K.jpeg)
**gucci,spring 2017 ready to wear**

![image/jpeg](https://cdn-uploads.huggingface.co/production/uploads/648a824a8ca6cf9857d1349c/Ghd9kUxoCOyOeyJNfUtnh.jpeg)
**prada,fall 2018 ready to wear**

## Training and evaluation data
It achieves the following results on the evaluation set:
- Loss: 0.1795
- Accuracy: 0.9454

### 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: 15

### Framework versions

- Transformers 4.35.2
- Pytorch 2.1.0+cu121
- Datasets 2.16.0
- Tokenizers 0.15.0