import streamlit as st
from transformers import GPT2Tokenizer, GPT2LMHeadModel

# Define model and tokenizer
model_name = 'gpt2-large'
st.write("Loading model and tokenizer...")
model = GPT2LMHeadModel.from_pretrained(model_name)
tokenizer = GPT2Tokenizer.from_pretrained(model_name)
st.write("Model and tokenizer loaded.")

def generate_blogpost(topic):
    try:
        inputs = tokenizer.encode(topic, return_tensors='pt')
        attention_mask = tokenizer.encode_plus(topic, return_tensors='pt')['attention_mask']
        outputs = model.generate(
            inputs,
            attention_mask=attention_mask,
            max_length=500,
            num_return_sequences=1,
            pad_token_id=tokenizer.eos_token_id
        )
        text = tokenizer.decode(outputs[0], skip_special_tokens=True)
        return text
    except Exception as e:
        return f"Error: {e}"

# Streamlit app
st.title('Blog Post Generator')
topic = st.text_input('Enter a topic:')

if topic:
    st.write("Generating blog post...")
    blogpost = generate_blogpost(topic)
    st.write(blogpost)