File size: 1,827 Bytes
60b9491
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
import gradio as gr
from transformers import pipeline

auth_token = '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'
pipeline_en = pipeline(task="text-classification", model="Hello-SimpleAI/chatgpt-qa-detector-distil",use_auth_token=auth_token)
# pipeline_en = pipeline_zh 
# pipeline_zh = pipeline(task="text2text-generation", model="beyond/genius-base-chinese")


def predict_en(q,a):
    res = pipeline_en({"text":q, "text_pair":a})
    return res['label'],res['score']

def predict_zh(sketch):
#   generated_text = pipeline_zh(sketch, num_beams=3, do_sample=True, max_length=200)[0]['generated_text']
#   return generated_text.replace(' ','')
    return ''
  
 


with gr.Blocks() as demo:
    gr.Markdown("""
                ## ChatGPT detector
                """)
    with gr.Tab("English"):
        q1 = gr.Textbox(lines=2, label='Question',value="What stops a restaurant from noting down my credit card info and using it ? No offense to restaurants . Can be generalized to anyone who I give my credit card info to . Explain like I'm five.")
        a1 = gr.Textbox(lines=5, label='Answer',value="There are a few things that can help protect your credit card information from being misused when you give it to a restaurant or any other business")
        label = gr.Textbox(lines=1, label='Predicted Label 🎃')
        score = gr.Textbox(lines=1, label='Prob')
        button1 = gr.Button("🤖 Predict!")
    # with gr.Tab("Chinese"):
    #     input2 = gr.Textbox(lines=5, value="")
    #     output2 = gr.Textbox(lines=5)
    #     output2 = output2
    #     button2 = gr.Button("Generate")

    # with gr.Accordion("Open for More!"):
    #     gr.Markdown("Look at me...")

    button1.click(predict_en, inputs=[q1,a1], outputs=[label,score])
    # button2.click(predict_zh, inputs=input2, outputs=output2)

demo.launch()