chinese-clip-vit-base-patch16/README.md

---
license: apache-2.0
---


# Chinese-CLIP-Base

## Introduction
This is the base-version of the Chinese CLIP. Chinese CLIP is a simple implementation of CLIP on a large-scale dataset of around 200 million Chinese image-text pairs. For more details, please refer to our technical report https://arxiv.org/abs/2211.01335 and our official github repo https://github.com/OFA-Sys/Chinese-CLIP

## Use with the official API
We provide a simple code snippet to show how to use the API for Chinese-CLIP. For starters, please install cn_clip:
```bash
# to install the latest stable release
pip install cn_clip

# or install from source code
cd Chinese-CLIP
pip install -e .
```
After installation, use Chinese CLIP as shown below:
```python
import torch
from PIL import Image

import cn_clip.clip as clip
from cn_clip.clip import load_from_name, available_models
print("Available models:", available_models())  
# Available models: ['ViT-B-16', 'ViT-L-14', 'ViT-L-14-336', 'ViT-H-14', 'RN50']

device = "cuda" if torch.cuda.is_available() else "cpu"
model, preprocess = load_from_name("ViT-B-16", device=device, download_root='./')
model.eval()
image = preprocess(Image.open("examples/pokemon.jpeg")).unsqueeze(0).to(device)
text = clip.tokenize(["杰尼龟", "妙蛙种子", "小火龙", "皮卡丘"]).to(device)

with torch.no_grad():
    image_features = model.encode_image(image)
    text_features = model.encode_text(text)
    # Normalize the features. Please use the normalized features for downstream tasks.
    image_features /= image_features.norm(dim=-1, keepdim=True) 
    text_features /= text_features.norm(dim=-1, keepdim=True)      

    logits_per_image, logits_per_text = model.get_similarity(image, text)
    probs = logits_per_image.softmax(dim=-1).cpu().numpy()

print("Label probs:", probs)  # [[1.268734e-03 5.436878e-02 6.795761e-04 9.436829e-01]]
```

However, if you are not satisfied with only using the API, feel free to check our github repo https://github.com/OFA-Sys/Chinese-CLIP for more details about training and inference. 
<br><br>

## Results
### MUGE Text-to-Image Retrieval
<table border="1" width="100%">
    <tr align="center">
        <th>Setup</th><th colspan="4">Zero-shot</th><th colspan="4">Finetune</th>
    </tr>
    <tr align="center">
        <td>Metric</td><td>R@1</td><td>R@5</td><td>R@10</td><td>MR</td><td>R@1</td><td>R@5</td><td>R@10</td><td>MR</td>
    </tr>
    <tr align="center">
        <td>Wukong<sub>ViT-B</sub></td><td>33.4</td><td>59.3</td><td>69.7</td><td>54.1</td><td>39.2</td><td>66.9</td><td>77.4</td><td>61.2</td>
	</tr>
	<tr align="center">
        <td>R2D2<sub>ViT-B</sub></td><td>-</td><td>-</td><td>-</td><td>-</td><td>47.4</td><td>75.1</td><td>83.5</td><td>68.7</td>
	</tr>
	<tr align="center">
        <td>CN-CLIP<sub>ViT-B</sub></td><td><b>52.1</b></td><td><b>76.7</b></td><td><b>84.4</b></td><td><b>71.1</b></td><td><b>58.4</b></td><td><b>83.6</b></td><td><b>90.0</b></td><td><b>77.4</b></td>
	</tr>
</table>


### Flickr30K-CN Retrieval
<table border="1" width="100%">
	<tr align="center">
        <th>Task</th><th colspan="6">Text-to-Image</th><th colspan="6">Image-to-Text</th>
    </tr>
    <tr align="center">
        <th>Setup</th><th colspan="3">Zero-shot</th><th colspan="3">Finetune</th><th colspan="3">Zero-shot</th><th colspan="3">Finetune</th>
    </tr>
    <tr align="center">
        <td>Metric</td><td>R@1</td><td>R@5</td><td>R@10</td><td>R@1</td><td>R@5</td><td>R@10</td><td>R@1</td><td>R@5</td><td>R@10</td><td>R@1</td><td>R@5</td><td>R@10</td>
    </tr>
    <tr align="center">
        <td>Wukong<sub>ViT-B</sub></td><td>45.7</td><td>73.8</td><td>82.2</td><td>67.6</td><td>89.6</td><td>94.2</td><td>66.2</td><td>88.7</td><td>94.3</td><td>83.9</td><td>97.6</td><td>99.0</td>
	</tr>
	<tr align="center">
        <td>R2D2<sub>ViT-B</sub></td><td>-</td><td>-</td><td>-</td><td>78.3</td><td>94.6</td><td>97.0</td><td>-</td><td>-</td><td>-</td><td>92.6</td><td><b>99.1</b></td><td><b>99.8</b></td>
	</tr>
	<tr align="center">
        <td>CN-CLIP<sub>ViT-B</sub></td><td><b>62.7</b></td><td><b>86.9</b></td><td><b>92.8</b></td><td><b>79.1</b></td><td><b>94.8</b></td><td><b>97.4</b></td><td><b>74.6</b></td><td><b>93.5</b></td><td><b>97.1</b></td><td><b>93.5</b></td><td>99.0</td><td>99.5</td>
	</tr>
</table>


### COCO-CN Retrieval
<table border="1" width="100%">
	<tr align="center">
        <th>Task</th><th colspan="6">Text-to-Image</th><th colspan="6">Image-to-Text</th>
    </tr>
    <tr align="center">
        <th>Setup</th><th colspan="3">Zero-shot</th><th colspan="3">Finetune</th><th colspan="3">Zero-shot</th><th colspan="3">Finetune</th>
    </tr>
    <tr align="center">
        <td>Metric</td><td>R@1</td><td>R@5</td><td>R@10</td><td>R@1</td><td>R@5</td><td>R@10</td><td>R@1</td><td>R@5</td><td>R@10</td><td>R@1</td><td>R@5</td><td>R@10</td>
    </tr>
    <tr align="center">
        <td>Wukong<sub>ViT-B</sub></td><td>49.2</td><td>79.4</td><td>87.9</td><td>67.0</td><td>91.4</td><td>96.7</td><td>48.3</td><td>77.8</td><td>88.8</td><td>65.8</td><td>90.3</td><td>96.6</td>
	</tr>
	<tr align="center">
        <td>R2D2<sub>ViT-B</sub></td><td>-</td><td>-</td><td>-</td><td>75.1</td><td>94.2</td><td>98.1</td><td>-</td><td>-</td><td>-</td><td>76.1</td><td>95.3</td><td>98.5</td>
	</tr>
	<tr align="center">
        <td>CN-CLIP<sub>ViT-B</sub></td><td><b>62.2</b></td><td><b>86.6</b></td><td><b>94.9</b></td><td><b>77.0</b></td><td><b>97.1</b></td><td><b>99.0</b></td><td><b>57.0</b></td><td><b>84.1</b></td><td><b>93.6</b></td><td><b>77.4</b></td><td><b>96.2</b></td><td><b>98.9</b></td>
	</tr>
</table>
<br>

## Citation
If you find Chinese CLIP helpful, feel free to cite our paper. Thanks for your support!

```
@article{chinese-clip,
  title={Chinese CLIP: Contrastive Vision-Language Pretraining in Chinese},
  author={Yang, An and Pan, Junshu and Lin, Junyang and Men, Rui and Zhang, Yichang and Zhou, Jingren and Zhou, Chang},
  journal={arXiv preprint arXiv:2211.01335},
  year={2022}
}
```
<br>