app.py

import gradio as gr
from gradio_leaderboard import Leaderboard, ColumnFilter, SelectColumns
import pandas as pd
from apscheduler.schedulers.background import BackgroundScheduler
from huggingface_hub import snapshot_download
import os
import shutil
import time

from src.about import (
    CITATION_BUTTON_LABEL,
    CITATION_BUTTON_TEXT,
    EVALUATION_QUEUE_TEXT,
    INTRODUCTION_TEXT,
    LLM_BENCHMARKS_TEXT,
    TITLE,
)
from src.display.css_html_js import custom_css
from src.display.utils import (
    BENCHMARK_COLS,
    COLS,
    EVAL_COLS,
    EVAL_TYPES,
    AutoEvalColumn,
    ModelType,
    fields,
    WeightType,
    Precision
)

from src.envs import API, EVAL_REQUESTS_PATH, EVAL_RESULTS_PATH, QUEUE_REPO, REPO_ID, RESULTS_REPO, TOKEN
from src.populate import get_evaluation_queue_df, get_leaderboard_df
from src.submission.submit import add_new_eval

loggedin = False


def check_login(profile: gr.OAuthProfile | None) -> bool:
    if profile is None:
        return False
    return True

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

### Space initialisation

def cleanup_old_cache():
    """Remove old cache directories to free up space"""
    cache_dirs = [EVAL_REQUESTS_PATH, EVAL_RESULTS_PATH]
    for cache_dir in cache_dirs:
        if os.path.exists(cache_dir):
            # Check if cache is older than 6 hours
            cache_age = time.time() - os.path.getctime(cache_dir)
            if cache_age > 21600:  # 6 hours in seconds
                print(f"Removing old cache: {cache_dir}")
                shutil.rmtree(cache_dir, ignore_errors=True)

def safe_download_dataset(repo_id, local_dir, repo_type="dataset"):
    """Download dataset only if not already cached or cache is stale"""
    try:
        # Check if directory exists and has recent data
        if os.path.exists(local_dir) and os.listdir(local_dir):
            cache_age = time.time() - os.path.getctime(local_dir)
            if cache_age < 3600:  # Less than 1 hour old
                print(f"Using cached data: {local_dir}")
                return

        print(f"Downloading: {repo_id} to {local_dir}")
        snapshot_download(
            repo_id=repo_id,
            local_dir=local_dir,
            repo_type=repo_type,
            tqdm_class=None,
            etag_timeout=30,
            token=TOKEN,
            resume_download=True,
            force_download=False
        )
    except Exception as e:
        print(f"Download failed for {repo_id}: {e}")
        if not os.path.exists(local_dir) or not os.listdir(local_dir):
            restart_space()

# Clean up old cache to free space
cleanup_old_cache()

# Download datasets with caching
safe_download_dataset(QUEUE_REPO, EVAL_REQUESTS_PATH, "dataset")
safe_download_dataset(RESULTS_REPO, EVAL_RESULTS_PATH, "dataset")


LEADERBOARD_DF = get_leaderboard_df(EVAL_RESULTS_PATH, EVAL_REQUESTS_PATH, COLS, BENCHMARK_COLS)

(
    finished_eval_queue_df,
    running_eval_queue_df,
    pending_eval_queue_df,
) = get_evaluation_queue_df(EVAL_REQUESTS_PATH, EVAL_COLS)

def init_leaderboard(dataframe):
    # if dataframe is None or dataframe.empty:
    #     raise ValueError("Leaderboard DataFrame is empty or None.")
    # print(dataframe.columns)
    return Leaderboard(
        value=dataframe,
        datatype=[c.type for c in fields(AutoEvalColumn)],
        select_columns=SelectColumns(
            default_selection=[c.name for c in fields(AutoEvalColumn) if c.displayed_by_default],
            cant_deselect=[c.name for c in fields(AutoEvalColumn) if c.never_hidden],
            label="Select Columns to Display:",
        ),
        search_columns=[AutoEvalColumn.model.name],
        hide_columns=[c.name for c in fields(AutoEvalColumn) if c.hidden],
        filter_columns=[
            # ColumnFilter(AutoEvalColumn.model_type.name, type="checkboxgroup", label="Model types"),
            # ColumnFilter(AutoEvalColumn.precision.name, type="checkboxgroup", label="Precision"),
            # ColumnFilter(
            #     AutoEvalColumn.params.name,
            #     type="slider",
            #     min=0.01,
            #     max=150,
            #     label="Select the number of parameters (B)",
            # ),
            # ColumnFilter(
            #     AutoEvalColumn.still_on_hub.name, type="boolean", label="Deleted/incomplete", default=True
            # ),
        ],
        bool_checkboxgroup_label="Hide models",
        interactive=False,
    )


demo = gr.Blocks(css=custom_css)
with demo:
    gr.HTML(TITLE)
    gr.Markdown(INTRODUCTION_TEXT, elem_classes="markdown-text")


    with gr.Tabs(elem_classes="tab-buttons") as tabs:
        with gr.TabItem("🏅 LLM Benchmark", elem_id="llm-benchmark-tab-table", id=0):
            leaderboard = init_leaderboard(LEADERBOARD_DF)

        with gr.TabItem("📝 About", elem_id="llm-benchmark-tab-table", id=2):
            gr.Markdown(LLM_BENCHMARKS_TEXT, elem_classes="markdown-text")

        with gr.TabItem("🚀 Submit here! ", elem_id="llm-benchmark-tab-table", id=3):
            # with gr.Column():
            #     with gr.Row():
            #         gr.Markdown(EVALUATION_QUEUE_TEXT, elem_classes="markdown-text")

                # with gr.Column():
                    # with gr.Accordion(
                    #     f"✅ Finished Evaluations ({len(finished_eval_queue_df)})",
                    #     open=False,
                    # ):
                    #     with gr.Row():
                    #         finished_eval_table = gr.components.Dataframe(
                    #             value=finished_eval_queue_df,
                    #             headers=EVAL_COLS,
                    #             datatype=EVAL_TYPES,
                    #             row_count=5,
                    #         )
                    # with gr.Accordion(
                    #     f"🔄 Running Evaluation Queue ({len(running_eval_queue_df)})",
                    #     open=False,
                    # ):
                    #     with gr.Row():
                    #         running_eval_table = gr.components.Dataframe(
                    #             value=running_eval_queue_df,
                    #             headers=EVAL_COLS,
                    #             datatype=EVAL_TYPES,
                    #             row_count=5,
                    #         )

                    # with gr.Accordion(
                    #     f"⏳ Pending Evaluation Queue ({len(pending_eval_queue_df)})",
                    #     open=False,
                    # ):
                    #     with gr.Row():
                    #         pending_eval_table = gr.components.Dataframe(
                    #             value=pending_eval_queue_df,
                    #             headers=EVAL_COLS,
                    #             datatype=EVAL_TYPES,
                    #             row_count=5,

            with gr.Row():
                gr.Markdown("# ✉️✨ Submit your model here!", elem_classes="markdown-text")
                login_button = gr.LoginButton(elem_id="oauth-button")
                

            with gr.Row():
                with gr.Column():
                    model_name_textbox = gr.Textbox(label="Model name")
                    revision_name_textbox = gr.Textbox(label="Revision commit", placeholder="main")
                    model_type = gr.Dropdown(
                        choices=[t.to_str(" : ") for t in ModelType if t != ModelType.Unknown],
                        label="Model type",
                        multiselect=False,
                        value=None,
                        interactive=True,
                    )

                with gr.Column():
                    precision = gr.Dropdown(
                        choices=[i.value.name for i in Precision if i != Precision.Unknown],
                        label="Precision",
                        multiselect=False,
                        value="float16",
                        interactive=True,
                    )
                    weight_type = gr.Dropdown(
                        choices=[i.value.name for i in WeightType],
                        label="Weights type",
                        multiselect=False,
                        value="Original",
                        interactive=True,
                    )
                    base_model_name_textbox = gr.Textbox(label="Организация")
                    ans_file = gr.File(label="Arena Hard Answer File", file_types=[".json",".jsonl"])
            # loggedin = login_button.click(check_login)

            submit_button = gr.Button("Submit Eval")
            submission_result = gr.Markdown()
            submit_button.click(
                    add_new_eval,
                    [
                        model_name_textbox,
                        base_model_name_textbox,
                        revision_name_textbox,
                        precision,
                        weight_type,
                        model_type,
                        ans_file,
                    ],
                    submission_result,
                )

    with gr.Row():
        with gr.Accordion("📙 Citation", open=False):
            citation_button = gr.Textbox(
                value=CITATION_BUTTON_TEXT,
                label=CITATION_BUTTON_LABEL,
                lines=20,
                elem_id="citation-button",
                show_copy_button=True,
            )

scheduler = BackgroundScheduler()
scheduler.add_job(restart_space, "interval", seconds=1800)
scheduler.start()
demo.queue(default_concurrency_limit=40).launch()