import gradio as gr
from transformers import GPT2LMHeadModel, GPT2Tokenizer

# Load pre-trained model and tokenizer from Hugging Face
model_name = "distilgpt2"  # Use a smaller model like distilgpt2 for testing
tokenizer = GPT2Tokenizer.from_pretrained(model_name)
model = GPT2LMHeadModel.from_pretrained(model_name)

# Ensure the model is in evaluation mode
model.eval()

# Check if model is loaded correctly
print("Model and tokenizer loaded successfully!")

def chat_with_bot(input_text):
    # Tokenize the input and encode it as tensor
    inputs = tokenizer.encode(input_text + tokenizer.eos_token, return_tensors="pt")
    
    # Generate a response from the model
    with torch.no_grad():
        outputs = model.generate(inputs, max_length=150, num_return_sequences=1, no_repeat_ngram_size=2)
    
    # Decode the generated tokens back to text
    response = tokenizer.decode(outputs[0], skip_special_tokens=True)
    
    return response.strip()  # Remove unnecessary tokens

# Define the Gradio interface
iface = gr.Interface(fn=chat_with_bot, 
                     inputs="text", 
                     outputs="text", 
                     title="Chatbot", 
                     description="A simple chatbot using distilgpt2")

iface.launch()