File size: 2,188 Bytes
8fcc055
5c8be37
8fcc055
f27161f
100aaa7
 
f27161f
1b691d8
86bb08d
1b691d8
911ff06
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
5c8be37
 
8fcc055
86bb08d
911ff06
5c8be37
8fcc055
 
 
 
 
5bdcf36
8fcc055
86bb08d
 
 
 
 
 
8fcc055
 
 
 
 
 
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
52
53
54
55
56
import gradio as gr
from transformers import AutoModelForSeq2SeqLM, AutoTokenizer

# Load the model and tokenizer
model = AutoModelForSeq2SeqLM.from_pretrained("vennify/t5-base-grammar-correction")
tokenizer = AutoTokenizer.from_pretrained("vennify/t5-base-grammar-correction")

def correct_text(text, max_length, max_new_tokens, min_length, num_beams, temperature, top_p):
    inputs = tokenizer.encode("grammar: " + text, return_tensors="pt")
    
    if max_new_tokens > 0:
        outputs = model.generate(
            inputs,
            max_length=max_length,
            max_new_tokens=max_new_tokens,
            min_length=min_length,
            num_beams=num_beams,
            temperature=temperature,
            top_p=top_p,
            early_stopping=True
        )
    else:
        outputs = model.generate(
            inputs,
            max_length=max_length,
            min_length=min_length,
            num_beams=num_beams,
            temperature=temperature,
            top_p=top_p,
            early_stopping=True
        )
    corrected_text = tokenizer.decode(outputs[0], skip_special_tokens=True)
    return corrected_text

def respond(message, history, max_length, min_length, max_new_tokens, num_beams, temperature, top_p):
    response = correct_text(message, max_length, max_new_tokens, min_length, num_beams, temperature, top_p)
    yield response

"""
For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
"""
demo = gr.ChatInterface(
    respond,
    additional_inputs=[
        gr.Slider(minimum=1,   maximum=256, value=100,  step=1,    label="Max Length"),
        gr.Slider(minimum=1,   maximum=256, value=0,    step=1,    label="Min Length"),
        gr.Slider(minimum=0,   maximum=256, value=0,    step=1,    label="Max New Tokens (optional)"),
        gr.Slider(minimum=1,   maximum=10,  value=5,    step=1,    label="Num Beams"),
        gr.Slider(minimum=0.1, maximum=4.0, value=0.7,  step=0.1,  label="Temperature"),
        gr.Slider(minimum=0.1, maximum=1.0, value=0.95, step=0.05, label="Top-p (nucleus sampling)"),
    ],
)


if __name__ == "__main__":
    demo.launch()