app.py

import gradio as gr
from transformers import AutoTokenizer, AutoModelForSequenceClassification
import torch
from sklearn.metrics.pairwise import cosine_similarity
import numpy as np

# Load a pre-trained model and tokenizer from Hugging Face
model_name = "sentence-transformers/all-MiniLM-L6-v2"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForSequenceClassification.from_pretrained(model_name)

# Sample dataset of questions and answers
dataset = [
    ("What is the capital of France?", "Paris is the capital of France."),
    ("Who is the creator of Python?", "Guido van Rossum created Python."),
    ("What is the tallest mountain in the world?", "Mount Everest is the tallest mountain in the world."),
]

# Function to find the most relevant answer
def find_most_relevant_answer(question):
    question_embedding = model(**tokenizer(question, return_tensors="pt", padding=True, truncation=True))[0].mean(dim=1).detach().numpy()
    
    highest_similarity = -1
    most_relevant_answer = ""
    
    for q, a in dataset:
        answer_embedding = model(**tokenizer(q, return_tensors="pt", padding=True, truncation=True))[0].mean(dim=1).detach().numpy()
        similarity = cosine_similarity([question_embedding], [answer_embedding])[0][0]
        
        if similarity > highest_similarity:
            highest_similarity = similarity
            most_relevant_answer = a
            
    return most_relevant_answer

# Set up Gradio interface
def chat_with_bot(question):
    return find_most_relevant_answer(question)

iface = gr.Interface(fn=chat_with_bot, inputs="text", outputs="text", title="Simple QA Chatbot")
iface.launch()