chatbot.py

def chat_with_model():
    print("Welcome to the Question Answering Chatbot! (type 'exit' to quit)")
    while True:
        question = input("You: ")
        if question.lower() == 'exit':
            print("Goodbye!")
            break
        
        context = input("Context: ")
        if context.lower() == 'exit':
            print("Goodbye!")
            break
        
        response = question_answerer(question=question, context=context)
        answer = response['answer']
        score = response['score']
        
        print(f"Bot: {answer} (confidence: {score:.2f})")

# Save chat function code in a script
with open('chatbot.py', 'w') as f:
    f.write('''
from transformers import AutoTokenizer, TFAutoModelForQuestionAnswering, pipeline

# Load tokenizer and model
tokenizer = AutoTokenizer.from_pretrained("my_awesome_qa_model")
model = TFAutoModelForQuestionAnswering.from_pretrained("my_awesome_qa_model")

# Create a pipeline for question answering
question_answerer = pipeline("question-answering", model=model, tokenizer=tokenizer)

# Define the chat function
def chat_with_model():
    print("Welcome to the Question Answering Chatbot! (type 'exit' to quit)")
    while True:
        question = input("You: ")
        if question.lower() == 'exit':
            print("Goodbye!")
            break
        
        context = input("Context: ")
        if context.lower() == 'exit':
            print("Goodbye!")
            break
        
        response = question_answerer(question=question, context=context)
        answer = response['answer']
        score = response['score']
        
        print(f"Bot: {answer} (confidence: {score:.2f})")

# Run the chat function
if __name__ == "__main__":
    chat_with_model()
''')

print("Chatbot script 'chatbot.py' has been created.")

if __name__ == "__main__":
    chat_with_model()