import gradio as gr
from huggingface_hub import InferenceClient

"""
For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
"""
client = InferenceClient("fadodr/finetuned_mental_health_therapy_original")


def respond(
    message,
    history: list[tuple[str, str]],
    system_message,
    max_tokens,
    temperature,
    top_p,
):
    print("Message:", message)
    print("History:", history)
    print("System Message:", system_message)
    print("Max Tokens:", max_tokens)
    print("Temperature:", temperature)
    print("Top-p:", top_p)
    print(dir(client))
    try:
        messages = [{"role": "instruction", "content": system_message}]

        for val in history:
            if val[0]:
                messages.append({"role": "input", "content": val[0]})
            if val[1]:
                messages.append({"role": "response", "content": val[1]})

        messages.append({"role": "input", "content": message})

        response = ""
        print("sending message")
        print(messages)

        for message in client.text_generation(
            messages,
            stream=True,
            temperature=temperature,
            top_p=top_p,
        ):
            print(message)
            token = message.choices[0].delta.content[len(messages):]

            response += token
            yield response
    except Exception as e:
        print(e)

# from transformers import pipeline, BitsAndBytesConfig

# config = BitsAndBytesConfig(load_in_4bit=True)

# # Load the pipeline with your custom model
# generator = pipeline('text-generation', model='fadodr/finetuned_mental_health_therapy_original', quantization_config=config)

# # Generate text based on the input message
# def respond(message, history, system_message, max_tokens, temperature, top_p):
#     prompt = f"""
#     ### Input:
#     {message}

#     ### Instruction:
#     {system_message}

#     ### Response:
#     """

#     response = generator(prompt, max_length=max_tokens, temperature=temperature, top_p=top_p)
#     print(response)
#     yield response[0]['generated_text']

"""
For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
"""
demo = gr.ChatInterface(
    respond,
    additional_inputs=[
        gr.Textbox(value="I need your help as a mental health therapist", label="System message"),
        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
        gr.Slider(
            minimum=0.1,
            maximum=1.0,
            value=0.95,
            step=0.05,
            label="Top-p (nucleus sampling)",
        ),
    ],
)


if __name__ == "__main__":
    demo.launch()