File size: 3,636 Bytes
23cf698
 
 
 
 
 
 
29a378b
e4fced7
 
 
 
29a378b
ceb927d
33640e7
 
 
 
ceb927d
e4fced7
 
 
 
33640e7
 
 
 
ceb927d
23cf698
 
33640e7
494fb65
ceb927d
5119d09
33640e7
23cf698
ceb927d
23cf698
520d4cb
 
33640e7
 
 
520d4cb
 
b725b48
33640e7
83dac04
520d4cb
23cf698
 
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
49
50
51
from turtle import title
import gradio as gr
from transformers import pipeline
import numpy as np
from PIL import Image


pipes = {
    "ViT/B-16": pipeline("zero-shot-image-classification", model="OFA-Sys/chinese-clip-vit-base-patch16"),
    "ViT/L-14": pipeline("zero-shot-image-classification", model="OFA-Sys/chinese-clip-vit-large-patch14"),
    "ViT/L-14@336px": pipeline("zero-shot-image-classification", model="OFA-Sys/chinese-clip-vit-large-patch14-336px"),
    "ViT/H-14": pipeline("zero-shot-image-classification", model="OFA-Sys/chinese-clip-vit-huge-patch14"),
}
inputs = [
    gr.inputs.Image(type='pil', 
                    label="Image 输入图片"),
    gr.inputs.Textbox(lines=1, 
                      label="Candidate Labels 候选分类标签"),
    gr.inputs.Radio(choices=[
                                "ViT/B-16",
                                "ViT/L-14", 
                                "ViT/L-14@336px", 
                                "ViT/H-14",
                            ], type="value", default="ViT/B-16", label="Model 模型规模"), 
    gr.inputs.Textbox(lines=1, 
                      label="Prompt Template Prompt模板 ({}指代候选标签)", 
                      default="一张{}的图片。"),
]
images="festival.jpg"

def shot(image, labels_text, model_name, hypothesis_template):
    labels = [label.strip(" ") for label in labels_text.strip(" ").split(",")]
    res = pipes[model_name](images=image, 
           candidate_labels=labels,
           hypothesis_template=hypothesis_template)
    return {dic["label"]: dic["score"] for dic in res}

iface = gr.Interface(shot, 
            inputs, 
            "label", 
            examples=[["festival.jpg", "灯笼, 鞭炮, 对联", "ViT/B-16", "一张{}的图片。"], 
                      ["cat-dog-music.png", "音乐表演, 体育运动", "ViT/B-16", "一张{}的图片。"],
                      ["football-match.jpg", "梅西, C罗, 马奎尔", "ViT/B-16", "一张{}的图片。"]],
            description="""<p>Chinese CLIP is a contrastive-learning-based vision-language foundation model pretrained on large-scale Chinese data. For more information, please refer to the paper and official github. Also, Chinese CLIP has already been merged into Huggingface Transformers! <br><br>
            Paper: <a href='https://arxiv.org/abs/2211.01335'>https://arxiv.org/abs/2211.01335</a> <br>
            Github: <a href='https://github.com/OFA-Sys/Chinese-CLIP'>https://github.com/OFA-Sys/Chinese-CLIP</a> (Welcome to star! 🔥🔥) <br><br>
            To play with this demo, add a picture and a list of labels in Chinese separated by commas. 上传图片,并输入多个分类标签,用英文逗号分隔。可点击页面最下方示例参考。<br>
            You can duplicate this space and run it privately: <a href='https://huggingface.co/spaces/OFA-Sys/chinese-clip-zero-shot-image-classification?duplicate=true'><img src='https://img.shields.io/badge/-Duplicate%20Space-blue?labelColor=white&style=flat&logo=data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAABAAAAAQCAYAAAAf8/9hAAAAAXNSR0IArs4c6QAAAP5JREFUOE+lk7FqAkEURY+ltunEgFXS2sZGIbXfEPdLlnxJyDdYB62sbbUKpLbVNhyYFzbrrA74YJlh9r079973psed0cvUD4A+4HoCjsA85X0Dfn/RBLBgBDxnQPfAEJgBY+A9gALA4tcbamSzS4xq4FOQAJgCDwV2CPKV8tZAJcAjMMkUe1vX+U+SMhfAJEHasQIWmXNN3abzDwHUrgcRGmYcgKe0bxrblHEB4E/pndMazNpSZGcsZdBlYJcEL9Afo75molJyM2FxmPgmgPqlWNLGfwZGG6UiyEvLzHYDmoPkDDiNm9JR9uboiONcBXrpY1qmgs21x1QwyZcpvxt9NS09PlsPAAAAAElFTkSuQmCC&logoWidth=14' alt='Duplicate Space'></a></p>""",
            title="Zero-shot Image Classification (中文零样本图像分类)")

iface.launch()