semantic-entropy-probes

Sleeping

File size: 12,421 Bytes

import os
import pickle as pkl
from pathlib import Path
from threading import Thread
from typing import List, Tuple, Iterator
from queue import Queue

import spaces
import gradio as gr
import torch
from transformers import AutoModelForCausalLM, AutoTokenizer, TextIteratorStreamer


MAX_MAX_NEW_TOKENS = 2048
DEFAULT_MAX_NEW_TOKENS = 1024
MAX_INPUT_TOKEN_LENGTH = int(os.getenv("MAX_INPUT_TOKEN_LENGTH", "4096"))

DESCRIPTION = """
<h1>Llama-2 7B Chat with Uncertainty Probes</h1>
<p>This Space demonstrates the Llama-2-7b-chat model with a semantic uncertainty probe.</p>
<p>The highlighted text shows the model's uncertainty in real-time:</p>
<ul>
    <li><span style="background-color: #00FF00; color: black">Green</span> indicates more certain generations</li>
    <li><span style="background-color: #FF0000; color: black">Red</span> indicates more uncertain generations</li>
</ul>
"""

EXAMPLES = [
    ["What is the capital of France?", "You are a helpful assistant.", []],
    ["Explain the theory of relativity in simple terms.", "You are an expert physicist explaining concepts to a layman.", []],
    ["Write a short poem about artificial intelligence.", "You are a creative poet with a interest in technology.", []]
]

if torch.cuda.is_available():
    model_id = "meta-llama/Llama-2-7b-chat-hf"
    # TODO load the full model not the 8bit one?
    model = AutoModelForCausalLM.from_pretrained(model_id, device_map="auto", load_in_8bit=True)
    tokenizer = AutoTokenizer.from_pretrained(model_id)
    tokenizer.use_default_system_prompt = False

    # load the probe data
    # TODO compare accuracy and SE probe in different tabs/sections
    with open("./model/20240625-131035_demo.pkl", "rb") as f:
        probe_data = pkl.load(f)
    # take the NQ open one
    probe_data = probe_data[-2]
    se_probe = probe_data['t_bmodel']
    se_layer_range = probe_data['sep_layer_range']
    acc_probe = probe_data['t_amodel']
    acc_layer_range = probe_data['ap_layer_range']
else:
    DESCRIPTION += "\n<p>Running on CPU 🥶 This demo does not work on CPU.</p>"


class CustomStreamer(TextIteratorStreamer):
    """
    Streamer to also store hidden states in a queue.
    TODO check this works
    """
    def __init__(self, tokenizer, skip_prompt: bool = False, skip_special_tokens: bool = False, **decode_kwargs):
        super().__init__(tokenizer, skip_prompt, skip_special_tokens, **decode_kwargs)
        self.hidden_states_queue = Queue()

    def put(self, value):
        if isinstance(value, dict) and 'hidden_states' in value:
            self.hidden_states_queue.put(value['hidden_states'])
        super().put(value)

# Streamer claude
# def generate(
#     message: str,
#     system_prompt: str,
#     chat_history: List[Tuple[str, str]],
#     max_new_tokens: int = DEFAULT_MAX_NEW_TOKENS,
#     temperature: float = 0.6,
#     top_p: float = 0.9,
#     top_k: int = 50,
#     repetition_penalty: float = 1.2,
# ) -> Iterator[Tuple[str, str]]:
#     conversation = []
#     if system_prompt:
#         conversation.append({"role": "system", "content": system_prompt})
#     for user, assistant in chat_history:
#         conversation.extend([{"role": "user", "content": user}, {"role": "assistant", "content": assistant}])
#     conversation.append({"role": "user", "content": message})

#     input_ids = tokenizer.apply_chat_template(conversation, return_tensors="pt")
#     if input_ids.shape[1] > MAX_INPUT_TOKEN_LENGTH:
#         input_ids = input_ids[:, -MAX_INPUT_TOKEN_LENGTH:]
#         gr.Warning(f"Trimmed input from conversation as it was longer than {MAX_INPUT_TOKEN_LENGTH} tokens.")
#     input_ids = input_ids.to(model.device)

#     streamer = CustomStreamer(tokenizer, skip_prompt=True, skip_special_tokens=True)
#     generation_kwargs = dict(
#         input_ids=input_ids,
#         max_new_tokens=max_new_tokens,
#         do_sample=True,
#         top_p=top_p,
#         top_k=top_k,
#         temperature=temperature,
#         repetition_penalty=repetition_penalty,
#         streamer=streamer,
#         output_hidden_states=True,
#         return_dict_in_generate=True,
#     )

#     thread = Thread(target=model.generate, kwargs=generation_kwargs)
#     thread.start()

#     se_highlighted_text = ""
#     acc_highlighted_text = ""
#     for new_text in streamer:
#         hidden_states = streamer.hidden_states_queue.get()
        
#         # Semantic Uncertainty Probe
#         se_token_embeddings = torch.stack([layer[0, -1, :].cpu() for layer in hidden_states])
#         se_concat_layers = se_token_embeddings.numpy()[se_layer_range[0]:se_layer_range[1]].reshape(-1)
#         se_probe_pred = se_probe.predict_proba(se_concat_layers.reshape(1, -1))[0][1] * 2 - 1
        
#         # Accuracy Probe
#         acc_token_embeddings = torch.stack([layer[0, -1, :].cpu() for layer in hidden_states])
#         acc_concat_layers = acc_token_embeddings.numpy()[acc_layer_range[0]:acc_layer_range[1]].reshape(-1)
#         acc_probe_pred = acc_probe.predict_proba(acc_concat_layers.reshape(1, -1))[0][1] * 2 - 1
        
#         se_new_highlighted_text = highlight_text(new_text, se_probe_pred)
#         acc_new_highlighted_text = highlight_text(new_text, acc_probe_pred)
        
#         se_highlighted_text += se_new_highlighted_text
#         acc_highlighted_text += acc_new_highlighted_text
        
#         yield se_highlighted_text, acc_highlighted_text

@spaces.GPU
def generate(
    message: str,
    chat_history: List[Tuple[str, str]],
    system_prompt: str,
    max_new_tokens: int = DEFAULT_MAX_NEW_TOKENS,
    temperature: float = 0.6,
    top_p: float = 0.9,
    top_k: int = 50,
    repetition_penalty: float = 1.2,
) -> Iterator[str]:
    conversation = []
    if system_prompt:
        conversation.append({"role": "system", "content": system_prompt})
    for user, assistant in chat_history:
        conversation.extend([{"role": "user", "content": user}, {"role": "assistant", "content": assistant}])
    conversation.append({"role": "user", "content": message})

    input_ids = tokenizer.apply_chat_template(conversation, return_tensors="pt")
    if input_ids.shape[1] > MAX_INPUT_TOKEN_LENGTH:
        input_ids = input_ids[:, -MAX_INPUT_TOKEN_LENGTH:]
        gr.Warning(f"Trimmed input from conversation as it was longer than {MAX_INPUT_TOKEN_LENGTH} tokens.")
    input_ids = input_ids.to(model.device)

    streamer = TextIteratorStreamer(tokenizer, timeout=10.0, skip_prompt=True, skip_special_tokens=True)
    generation_kwargs = dict(
        input_ids=input_ids,
        max_new_tokens=max_new_tokens,
        do_sample=True,
        top_p=top_p,
        top_k=top_k,
        temperature=temperature,
        repetition_penalty=repetition_penalty,
        streamer=streamer,
        output_hidden_states=True,
        return_dict_in_generate=True,
    )

    # Generate without threading
    with torch.no_grad():
        outputs = model.generate(**generation_kwargs)
    generated_tokens = outputs.sequences[0, input_ids.shape[1]:]
    generated_text = tokenizer.decode(generated_tokens, skip_special_tokens=True)
    # hidden states
    hidden = outputs.hidden_states  # list of tensors, one for each token, then (batch size, sequence length, hidden size)

    # TODO do this loop on the fly instead of waiting for the whole generation
    se_highlighted_text = ""
    acc_highlighted_text = ""
    for i in range(1, len(hidden)):

        # Semantic Uncertainty Probe
        token_embeddings = torch.stack([generated_token[0, 0, :].cpu() for generated_token in hidden[i]]).numpy()   # (num_layers, hidden_size)
        se_concat_layers = token_embeddings[se_layer_range[0]:se_layer_range[1]].reshape(-1)
        se_probe_pred = se_probe.predict_proba(se_concat_layers.reshape(1, -1))[0][1] * 2 - 1
        
        # Accuracy Probe
        # acc_token_embeddings = torch.stack([layer[0, -1, :].cpu() for layer in hidden_states])
        acc_concat_layers = token_embeddings[acc_layer_range[0]:acc_layer_range[1]].reshape(-1)
        acc_probe_pred = -1 * acc_probe.predict_proba(acc_concat_layers.reshape(1, -1))[0][1] * 2 - 1
        
        output_id = outputs.sequences[0, input_ids.shape[1]+i]
        output_word = tokenizer.decode(output_id)
        print(output_id, output_word, se_probe_pred, acc_probe_pred)  

        se_new_highlighted_text = highlight_text(output_word, se_probe_pred)
        acc_new_highlighted_text = highlight_text(output_word, acc_probe_pred)
        se_highlighted_text += f" {se_new_highlighted_text}"
        acc_highlighted_text += f" {acc_new_highlighted_text}"
        
        yield se_highlighted_text, acc_highlighted_text


def highlight_text(text: str, uncertainty_score: float) -> str:
    if uncertainty_score > 0:
        html_color = "#%02X%02X%02X" % (
            255,
            int(255 * (1 - uncertainty_score)),
            int(255 * (1 - uncertainty_score)),
        )
    else:
        html_color = "#%02X%02X%02X" % (
            int(255 * (1 + uncertainty_score)),
            255,
            int(255 * (1 + uncertainty_score)),
        )
    return '<span style="background-color: {}; color: black">{}</span>'.format(
        html_color, text
    )

with gr.Blocks(title="Llama-2 7B Chat with Dual Probes", css="footer {visibility: hidden}") as demo:
    gr.HTML(DESCRIPTION)
    
    with gr.Row():
        with gr.Column():
            message = gr.Textbox(label="Message")
            system_prompt = gr.Textbox(label="System prompt", lines=2)
        
        with gr.Column():
            max_new_tokens = gr.Slider(label="Max new tokens", minimum=1, maximum=MAX_MAX_NEW_TOKENS, step=1, value=DEFAULT_MAX_NEW_TOKENS)
            temperature = gr.Slider(label="Temperature", minimum=0.1, maximum=4.0, step=0.1, value=0.6)
            top_p = gr.Slider(label="Top-p (nucleus sampling)", minimum=0.05, maximum=1.0, step=0.05, value=0.9)
            top_k = gr.Slider(label="Top-k", minimum=1, maximum=1000, step=1, value=50)
            repetition_penalty = gr.Slider(label="Repetition penalty", minimum=1.0, maximum=2.0, step=0.05, value=1.2)
    
    with gr.Row():
        generate_btn = gr.Button("Generate")
    # add spacing between probes and titles for each output
    with gr.Row():
        with gr.Column():
            title = gr.HTML("<h2>Semantic Uncertainty Probe</h2>")
            se_output = gr.HTML(label="Semantic Uncertainty Probe")
        with gr.Column():
            title = gr.HTML("<h2>Accuracy Probe</h2>")
            acc_output = gr.HTML(label="Accuracy Probe")
    
    chat_history = gr.State([])
    
    # gr.Examples(
    #     examples=EXAMPLES,
    #     inputs=[message, system_prompt, chat_history],
    #     outputs=[se_output, acc_output],
    #     fn=generate,
    # )
    
    generate_btn.click(
        generate,
        inputs=[message, system_prompt, chat_history, max_new_tokens, temperature, top_p, top_k, repetition_penalty],
        outputs=[se_output, acc_output]
    )

# chat_interface = gr.ChatInterface(
#     fn=generate,
#     additional_inputs=[
#         gr.Textbox(label="System prompt", lines=6),
#         gr.Slider(
#             label="Max new tokens",
#             minimum=1,
#             maximum=MAX_MAX_NEW_TOKENS,
#             step=1,
#             value=DEFAULT_MAX_NEW_TOKENS,
#         ),
#         gr.Slider(
#             label="Temperature",
#             minimum=0.1,
#             maximum=4.0,
#             step=0.1,
#             value=0.6,
#         ),
#         gr.Slider(
#             label="Top-p (nucleus sampling)",
#             minimum=0.05,
#             maximum=1.0,
#             step=0.05,
#             value=0.9,
#         ),
#         gr.Slider(
#             label="Top-k",
#             minimum=1,
#             maximum=1000,
#             step=1,
#             value=50,
#         ),
#         gr.Slider(
#             label="Repetition penalty",
#             minimum=1.0,
#             maximum=2.0,
#             step=0.05,
#             value=1.2,
#         ),
#     ],
#     stop_btn=None,
#     examples=[
#         ["What is the capital of France?"],
#         ["Who landed on the moon?"],
#         ["Who is Yarin Gal?"]
#     ],
#     title="Llama-2 7B Chat with Streamable Semantic Uncertainty Probe",
#     description=DESCRIPTION,
# )

# if __name__ == "__main__":
#     chat_interface.launch()


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