File size: 1,647 Bytes
3aa3e20
 
 
 
 
232a012
3aa3e20
232a012
13b79a4
 
 
 
232a012
3aa3e20
 
232a012
3aa3e20
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
13b79a4
3aa3e20
 
 
 
 
 
13b79a4
 
3aa3e20
dfb5811
 
 
 
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
57
58
59
from fastapi import FastAPI, HTTPException
from pydantic import BaseModel
from transformers import AutoModelForSeq2SeqLM, AutoTokenizer
import torch
import gradio as gr

app = FastAPI()

# Initialize model and tokenizer
model_name = "google/flan-t5-large"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForSeq2SeqLM.from_pretrained(model_name)

class Query(BaseModel):
    inputs: str

@app.post("/")
async def generate(query: Query):
    try:
        # Tokenize input
        inputs = tokenizer(query.inputs, return_tensors="pt", max_length=512, truncation=True)
        
        # Generate response
        outputs = model.generate(
            inputs.input_ids,
            max_length=512,
            num_beams=4,
            temperature=0.7,
            top_p=0.9,
            repetition_penalty=1.2,
            early_stopping=True
        )
        
        # Decode response
        response = tokenizer.decode(outputs[0], skip_special_tokens=True)
        return {"generated_text": response}
    
    except Exception as e:
        raise HTTPException(status_code=500, detail=str(e))

# Gradio interface
def generate_text(prompt):
    query = Query(inputs=prompt)
    response = generate(query)
    return response["generated_text"]

iface = gr.Interface(
    fn=generate_text,
    inputs=gr.Textbox(lines=2, placeholder="Enter your text here..."),
    outputs="text",
    title="Medical Assistant",
    description="Ask me anything about medical topics!"
)

# Mount the Gradio app
app = gr.mount_gradio_app(app, iface, path="/")

if __name__ == "__main__":
    import train  # This will start the training process