inference.py

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer

MODEL_NAME = "microsoft/DialoGPT-small"
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")

tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME)
model = AutoModelForCausalLM.from_pretrained(MODEL_NAME).to(device)

# Example conversation prompts
user_inputs = [
    "Hello, how are you?",
    "What is your favorite movie?",
    "Tell me a joke."
]

chat_history_ids = None

for user in user_inputs:
    new_user_input_ids = tokenizer.encode(user + tokenizer.eos_token, return_tensors="pt").to(device)
    bot_input_ids = torch.cat([chat_history_ids, new_user_input_ids], dim=-1) if chat_history_ids is not None else new_user_input_ids
    chat_history_ids = model.generate(bot_input_ids, max_length=1000, pad_token_id=tokenizer.eos_token_id)
    response = tokenizer.decode(chat_history_ids[:, bot_input_ids.shape[-1]:][0], skip_special_tokens=True)
    print(f"You: {user}")
    print(f"Bot: {response}\n")