app.py

import streamlit as st
from transformers import GPT2LMHeadModel, AutoTokenizer, AutoModelForCausalLM
import torch

# Initialize the tokenizer and model
tokenizer = AutoTokenizer.from_pretrained('gpt2-large')
model = AutoModelForCausalLM.from_pretrained('gpt2-large')

def generate_blog(topic, max_length=500, num_return_sequences=1):
    # Encode the topic as input IDs
    input_ids = tokenizer.encode(topic, return_tensors='pt')
    
    # Generate the blog text
    outputs = model.generate(
        input_ids,
        max_length=max_length,
        num_return_sequences=num_return_sequences,
        no_repeat_ngram_size=2,
        early_stopping=True
    )

    # Decode the generated IDs to text
    generated_texts = [tokenizer.decode(output, skip_special_tokens=True) for output in outputs]
    return generated_texts

# Streamlit UI
st.title("Blog Generator")
topic = st.text_input("Enter the topic's name:")

if st.button("Generate Blog"):
    if topic:
        generated_blogs = generate_blog(topic)
        for i, blog in enumerate(generated_blogs):
            st.subheader(f"Blog {i+1}")
            st.write(blog)
    else:
        st.write("Please enter a topic to generate a blog.")