from transformers import CodeLlamaForConditionalGeneration, CodeLlamaTokenizer

# Load pre-trained model and tokenizer
model_name = "codellama/CodeLlama-7b-hf"  # Assuming your CodeLlama model name
tokenizer = CodeLlamaTokenizer.from_pretrained(model_name)
model = CodeLlamaForConditionalGeneration.from_pretrained(model_name)

# System message
system_message = "You are a code teaching assistant named OmniCode created by Anusha K. Answer all the code related questions being asked."

def generate_response(prompt, max_length=150, temperature=1.0):
    input_text = system_message + "\n" + prompt
    input_ids = tokenizer.encode(input_text, return_tensors='pt')

    # Generate response
    output = model.generate(input_ids, 
                            max_length=max_length, 
                            temperature=temperature,
                            pad_token_id=tokenizer.eos_token_id,
                            num_return_sequences=1)

    # Decode and return the response
    response = tokenizer.decode(output[0], skip_special_tokens=True)
    return response

if __name__ == "__main__":
    while True:
        user_input = input("You: ")
        response = generate_response(user_input)
        print("OmniCode:", response)