File size: 2,441 Bytes
37b0c2c
 
 
b55eeb4
37b0c2c
 
 
 
 
 
 
 
 
 
 
 
 
c7ce0d9
 
9010034
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
37b0c2c
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
---
tags:
- image-classification
- vision
- pytorch
license: apache-2.0
datasets:
- cifar10
metrics:
- accuracy
thumbnail: https://avatars3.githubusercontent.com/u/32437151?s=460&u=4ec59abc8d21d5feea3dab323d23a5860e6996a4&v=4
---

# Vision Transformer Fine Tuned on CIFAR10

Vision Transformer (ViT) model pre-trained on ImageNet-21k (14 million images, 21,843 classes) and **fine-tuned on CIFAR10** at resolution 224x224.

Check out the code at my [my Github repo](https://github.com/nateraw/huggingface-vit-finetune).

## Usage

```python
from transformers import ViTFeatureExtractor, ViTForImageClassification
from PIL import Image
import requests

url = 'https://www.cs.toronto.edu/~kriz/cifar-10-sample/dog10.png'
image = Image.open(requests.get(url, stream=True).raw)
feature_extractor = ViTFeatureExtractor.from_pretrained('nateraw/vit-base-patch16-224-cifar10')
model = ViTForImageClassification.from_pretrained('nateraw/vit-base-patch16-224-cifar10')
inputs = feature_extractor(images=image, return_tensors="pt")
outputs = model(**inputs)
preds = outputs.logits.argmax(dim=1)

classes = [
    'airplane', 'automobile', 'bird', 'cat', 'deer', 'dog', 'frog', 'horse', 'ship', 'truck'
]
classes[preds[0]]
```

## Model description

The Vision Transformer (ViT) is a transformer encoder model (BERT-like) pretrained on a large collection of images in a supervised fashion, namely ImageNet-21k, at a resolution of 224x224 pixels. 
Images are presented to the model as a sequence of fixed-size patches (resolution 16x16), which are linearly embedded. One also adds a [CLS] token to the beginning of a sequence to use it for classification tasks. One also adds absolute position embeddings before feeding the sequence to the layers of the Transformer encoder.
Note that this model does not provide any fine-tuned heads, as these were zero'd by Google researchers. However, the model does include the pre-trained pooler, which can be used for downstream tasks (such as image classification).

By pre-training the model, it learns an inner representation of images that can then be used to extract features useful for downstream tasks: if you have a dataset of labeled images for instance, you can train a standard classifier by placing a linear layer on top of the pre-trained encoder. One typically places a linear layer on top of the [CLS] token, as the last hidden state of this token can be seen as a representation of an entire image.