app.py

import os
import json
import gradio as gr
import pandas as pd
import numpy as np

from pathlib import Path
from apscheduler.schedulers.background import BackgroundScheduler
from huggingface_hub import snapshot_download


from src.about import (
    CITATION_BUTTON_LABEL,
    CITATION_BUTTON_TEXT,
    EVALUATION_QUEUE_TEXT,
    INTRODUCTION_TEXT,
    LLM_BENCHMARKS_TEXT,
    TITLE,
    ABOUT_TEXT
)
from src.display.css_html_js import custom_css
# from src.display.utils import (
#     BENCHMARK_COLS,
#     COLS,
#     EVAL_COLS,
#     EVAL_TYPES,
#     NUMERIC_INTERVALS,
#     TYPES,
#     AutoEvalColumn,
#     ModelType,
#     fields,
#     WeightType,
#     Precision
# )
from src.envs import API, EVAL_REQUESTS_PATH, EVAL_RESULTS_PATH, QUEUE_REPO, REPO_ID, RESULTS_REPO, TOKEN

try:
    print(EVAL_RESULTS_PATH)
    snapshot_download(
        repo_id=RESULTS_REPO, local_dir=EVAL_RESULTS_PATH, repo_type="dataset", tqdm_class=None, etag_timeout=30, token=TOKEN
    )
except Exception:
    pass
    # restart_space()

SUBSET_COUNTS = {
    "Alignment-Object": 250,
    "Alignment-Attribute": 229,
    "Alignment-Action": 115,
    "Alignment-Count": 55,
    "Alignment-Location": 75,
    "Safety-Toxicity-Crime": 29,
    "Safety-Toxicity-Shocking": 31,
    "Safety-Toxicity-Disgust": 42,
    "Safety-Nsfw-Evident": 197,
    "Safety-Nsfw-Evasive": 177,
    "Safety-Nsfw-Subtle": 98,
    "Quality-Distortion-Human_face": 169,
    "Quality-Distortion-Human_limb": 152,
    "Quality-Distortion-Object": 100,
    "Quality-Blurry-Defocused": 350,
    "Quality-Blurry-Motion": 350,
    "Bias-Age": 80,
    "Bias-Gender": 140,
    "Bias-Race": 140,
    "Bias-Nationality": 120,
    "Bias-Religion": 60,
}

PERSPECTIVE_COUNTS= {
    "Alignment": 724,
    "Safety": 574,
    "Quality": 1121,
    "Bias": 540
}



META_DATA = ['Model']



def restart_space():
    API.restart_space(repo_id=REPO_ID)


# color_map = {
#     "Score Model": "#7497db",
#     "Opensource VLM": "#E8ECF2",
#     "Closesource VLM": "#ffcd75",
#     "Others": "#75809c",

#     # #7497db #E8ECF2 #ffcd75 #75809c
# }
# def color_model_type_column(df, color_map):
#     """
#     Apply color to the 'Model Type' column of the DataFrame based on a given color mapping.

#     Parameters:
#     df (pd.DataFrame): The DataFrame containing the 'Model Type' column.
#     color_map (dict): A dictionary mapping model types to colors.

#     Returns:
#     pd.Styler: The styled DataFrame.
#     """
#     # Function to apply color based on the model type
#     def apply_color(val):
#         color = color_map.get(val, "default")  # Default color if not specified in color_map
#         return f'background-color: {color}'
    
#     # Format for different columns
#     format_dict = {col: "{:.1f}" for col in df.columns if col not in META_DATA}
#     format_dict['Overall Score'] = "{:.2f}"
#     format_dict[''] = "{:d}"

#     return df.style.applymap(apply_color, subset=['Model Type']).format(format_dict, na_rep='')

def regex_table(dataframe, regex, filter_button, style=True):
    """
    Takes a model name as a regex, then returns only the rows that has that in it.
    """
    # Split regex statement by comma and trim whitespace around regexes
    regex_list = [x.strip() for x in regex.split(",")]
    # Join the list into a single regex pattern with '|' acting as OR
    combined_regex = '|'.join(regex_list)
    
    # if filter_button, remove all rows with "ai2" in the model name
    update_scores = False
    if isinstance(filter_button, list) or isinstance(filter_button, str):
        if "Integrated LVLM" not in filter_button:
            dataframe = dataframe[~dataframe["Model Type"].str.contains("Integrated LVLM", case=False, na=False)]
        if "Interleaved LVLM" not in filter_button:
            dataframe = dataframe[~dataframe["Model Type"].str.contains("Interleaved LVLM", case=False, na=False)]
    # Filter the dataframe such that 'model' contains any of the regex patterns
    data = dataframe[dataframe["Model"].str.contains(combined_regex, case=False, na=False)]

    data.reset_index(drop=True, inplace=True)

    # replace column '' with count/rank
    data.insert(0, '', range(1, 1 + len(data)))
    
    # if style:
    #     # apply color
    #     data = color_model_type_column(data, color_map)

    return data

def get_leaderboard_results(results_path):
    data_dir = Path(results_path)
    files = [d for d in os.listdir(data_dir)] # TODO check if "Path(data_dir) / d" is a dir

    df = pd.DataFrame()
    for file in files:
        if not file.endswith(".json"):
            continue
        with open(results_path / file) as rf:
            result = json.load(rf)
            result = pd.DataFrame(result)
            df = pd.concat([result, df])
    df.reset_index(drop=True, inplace=True)
    return df


def avg_all_perspective(orig_df: pd.DataFrame, columns_name: list, meta_data=META_DATA, perspective_counts=PERSPECTIVE_COUNTS):
    new_df = orig_df[meta_data + columns_name]
    new_perspective_counts = {col: perspective_counts[col] for col in columns_name}
    total_count = sum(perspective_counts.values())
    weights = {perspective: count / total_count for perspective, count in perspective_counts.items()}
    def calculate_weighted_avg(row):
        weighted_sum = sum(row[col] * weights[col] for col in columns_name)
        return weighted_sum
    new_df["Overall Score"] = new_df.apply(calculate_weighted_avg, axis=1)

    cols = meta_data + ["Overall Score"]  + columns_name
    new_df = new_df[cols].sort_values(by="Overall Score", ascending=False).reset_index(drop=True)
    return new_df
    
data = {
    "Model": [
        "MiniGPT-5", "EMU-2", "GILL", "Anole",
        "GPT-4o | Openjourney", "GPT-4o | SD-3", "GPT-4o | SD-XL", "GPT-4o | Flux",
        "Gemini-1.5 | Openjourney", "Gemini-1.5 | SD-3", "Gemini-1.5 | SD-XL", "Gemini-1.5 | Flux",
        "LLAVA-34b | Openjourney", "LLAVA-34b | SD-3", "LLAVA-34b | SD-XL", "LLAVA-34b | Flux",
        "Qwen-VL-70b | Openjourney", "Qwen-VL-70b | SD-3", "Qwen-VL-70b | SD-XL", "Qwen-VL-70b | Flux"
    ],
    "Model Type":[
        "Interleaved LVLM", "Interleaved LVLM", "Interleaved LVLM", "Interleaved LVLM",
        "Integrated LVLM", "Integrated LVLM", "Integrated LVLM", "Integrated LVLM",
        "Integrated LVLM", "Integrated LVLM", "Integrated LVLM", "Integrated LVLM",
        "Integrated LVLM", "Integrated LVLM", "Integrated LVLM", "Integrated LVLM",
        "Integrated LVLM", "Integrated LVLM", "Integrated LVLM", "Integrated LVLM",
    ],
    "Situational analysis": [
        47.63, 39.65, 46.72, 48.95,
        53.05, 53.00, 56.12, 54.97,
        48.08, 47.48, 49.43, 47.07,
        54.12, 54.72, 55.97, 54.23,
        52.73, 54.98, 52.58, 54.23
    ],
    "Project-based learning": [
        55.12, 46.12, 57.57, 59.05,
        71.40, 71.20, 73.25, 68.80,
        67.93, 68.70, 71.85, 68.33,
        73.47, 72.55, 74.60, 71.32,
        71.63, 71.87, 73.57, 69.47
    ],
    "Multi-step reasoning": [
        42.17, 50.75, 39.33, 51.72,
        53.67, 53.67, 53.67, 53.67,
        60.05, 60.05, 60.05, 60.05,
        47.28, 47.28, 47.28, 47.28,
        55.63, 55.63, 55.63, 55.63
    ],
    "AVG": [
        50.92, 45.33, 51.58, 55.22,
        63.65, 63.52, 65.47, 62.63,
        61.57, 61.87, 64.15, 61.55,
        63.93, 63.57, 65.05, 62.73,
        64.05, 64.75, 65.12, 63.18
    ]
}
df = pd.DataFrame(data)
total_models = len(df)

with gr.Blocks(css=custom_css) as app:
    with gr.Row():
        with gr.Column(scale=6):
            gr.Markdown(INTRODUCTION_TEXT.format(str(total_models)))
        with gr.Column(scale=4): 
            gr.Markdown("![](https://huggingface.co/spaces/MMIE/Leaderboard/resolve/main/src/overview.jpeg)")
            # gr.HTML(BGB_LOGO, elem_classes="logo")
    
    with gr.Tabs(elem_classes="tab-buttons") as tabs:
        with gr.TabItem("🏆 MMIE Leaderboard"):
            with gr.Row():
                search_overall = gr.Textbox(
                    label="Model Search (delimit with , )", 
                    placeholder="🔍 Search model (separate multiple queries with ``) and press ENTER...",
                    show_label=False
                )
                model_type_overall = gr.CheckboxGroup(
                    choices=["Interleaved LVLM", "Integrated LVLM"], 
                    value=["Interleaved LVLM", "Integrated LVLM"], 
                    label="Model Type", 
                    show_label=False, 
                    interactive=True,
                )
            with gr.Row():
                mmie_table_overall_hidden = gr.Dataframe(
                    df,
                    headers=df.columns.tolist(),
                    elem_id="mmie_leadboard_overall_hidden",
                    wrap=True,
                    visible=False,
                )
                mmie_table_overall = gr.Dataframe(
                    regex_table(
                        df.copy(), 
                        "", 
                        ["Interleaved LVLM", "Integrated LVLM"]
                     ),
                    headers=df.columns.tolist(),
                    elem_id="mmie_leadboard_overall",
                    wrap=True,
                    height=1000,
                )
        with gr.TabItem("About"):
            with gr.Row():
                gr.Markdown(ABOUT_TEXT)
    
    with gr.Accordion("📚 Citation", open=False):
            citation_button = gr.Textbox(
                value=r"""""",
                lines=7,
                label="Copy the following to cite these results.",
                elem_id="citation-button",
                show_copy_button=True,
            )
        
    search_overall.change(regex_table, inputs=[mmie_table_overall_hidden, search_overall, model_type_overall], outputs=mmie_table_overall)
    model_type_overall.change(regex_table, inputs=[mmie_table_overall_hidden, search_overall, model_type_overall], outputs=mmie_table_overall)
    
scheduler = BackgroundScheduler()
scheduler.add_job(restart_space, "interval", seconds=18000) # restarted every 3h
scheduler.start()
# app.queue(default_concurrency_limit=40).launch()
app.launch(allowed_paths=['./', "./src", "./evals"])