app.py

import gradio as gr
import openai

# OpenAI API Key (यहाँ अपनी API Key डालें)
openai.api_key = "YOUR_API_KEY"

# Backend Function: यूजर का मैसेज लेकर OpenAI से रिस्पॉन्स लाता है
def respond_to_message(message, chat_history):
    response = openai.ChatCompletion.create(
        model="gpt-3.5-turbo",
        messages=[{"role": "user", "content": message}]
    )
    bot_message = response.choices[0].message['content']
    chat_history.append((message, bot_message))
    return "", chat_history

# Frontend: Gradio UI
with gr.Blocks() as demo:
    chatbot = gr.Chatbot(label="AI चैट बोर्ड")
    msg = gr.Textbox(label="आपका मैसेज")
    clear = gr.ClearButton([msg, chatbot])

    msg.submit(respond_to_message, [msg, chatbot], [msg, chatbot])

demo.launch()

import gradio as gr
from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

# Model और Tokenizer लोड करें (आप चाहें तो कोई और चैट मॉडल भी ले सकते हैं)
model_name = "microsoft/DialoGPT-medium"   # या "mistralai/Mistral-7B-Instruct-v0.2" (अगर Spaces पर चलता है)
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(model_name)

def respond_to_message(message, chat_history):
    # Chat history को एक स्ट्रिंग में जोड़ें
    chat_input = ""
    for user, bot in chat_history:
        chat_input += f"User: {user}\nBot: {bot}\n"
    chat_input += f"User: {message}\nBot:"

    # Encode input
    input_ids = tokenizer.encode(chat_input, return_tensors="pt")
    # Generate response
    output = model.generate(
        input_ids,
        max_length=input_ids.shape[1] + 64,
        pad_token_id=tokenizer.eos_token_id,
        do_sample=True,
        top_k=50,
        top_p=0.95
    )
    response = tokenizer.decode(output[0][input_ids.shape[1]:], skip_special_tokens=True)
    chat_history.append((message, response.strip()))
    return "", chat_history

with gr.Blocks() as demo:
    chatbot = gr.Chatbot(label="AI चैट बोर्ड")
    msg = gr.Textbox(label="आपका मैसेज")
    clear = gr.ClearButton([msg, chatbot])

    msg.submit(respond_to_message, [msg, chatbot], [msg, chatbot])

demo.launch()

from transformers import GPT2Tokenizer, GPT2LMHeadModel

tokenizer = GPT2Tokenizer.from_pretrained("gpt2")
model = GPT2LMHeadModel.from_pretrained("gpt2")



from datasets import load_dataset

# Login using e.g. `huggingface-cli login` to access this dataset
ds = load_dataset("KadamParth/NCERT_Chemistry_11th")