app.py

import pandas as pd

import gradio as gr
from gradio_leaderboard import Leaderboard

from utils import fetch_hf_results, show_output_box
from constants import (
    ASSAY_LIST,
    ASSAY_RENAME,
    ASSAY_EMOJIS,
    ASSAY_DESCRIPTION,
    EXAMPLE_FILE_DICT,
    LEADERBOARD_DISPLAY_COLUMNS,
)
from about import ABOUT_INTRO, ABOUT_TEXT, FAQS
from submit import make_submission

def format_leaderboard_table(df_results: pd.DataFrame, assay: str | None = None):
    df = df_results.query("assay.isin(@ASSAY_RENAME.keys())").copy()
    if assay is not None:
        df = df[df["assay"] == assay]
    df = df[LEADERBOARD_DISPLAY_COLUMNS]
    df = df.sort_values(by="spearman", ascending=False)
    # After sorting, just add the reason for excluding heldout test set
    # Note: We can also just say the following as a text box at the bottom of the leaderboard: "Note: Results for the Heldout Test Set are only evaluated at competition close"
    # Convert spearman column to string to avoid dtype incompatibility when assigning text
    df["spearman"] = df["spearman"].astype(str)
    df.loc[(df["dataset"] == "Heldout Test Set") & (df["spearman"] == "nan"), "spearman"] = "N/A, evaluated at competition close"
    return df


def get_leaderboard_object(assay: str | None = None):
    filter_columns = ["dataset"]
    if assay is None:
        filter_columns.append("property")
    # TODO how to sort filter columns alphabetically?
    # Bug: Can't leave search_columns empty because then it says "Column None not found in headers"
    # Note(Lood): Would be nice to make it clear that the Search Column is searching on model name
    lb = Leaderboard(
        value=format_leaderboard_table(df_results=current_dataframe, assay=assay),
        datatype=["str", "str", "str", "number"],
        select_columns=["model", "property", "spearman", "dataset"],
        search_columns=["model"],  
        filter_columns=filter_columns,
        every=15,
        render=True,
    )
    return lb


# Initialize global dataframe
current_dataframe = fetch_hf_results()

with gr.Blocks() as demo:
    timer = gr.Timer(3)  # Run every 3 seconds when page is focused
    data_version = gr.State(value=0)  # Track data changes
    
    def update_current_dataframe():
        global current_dataframe
        new_dataframe = fetch_hf_results()
        
        # Check if data has actually changed
        if not current_dataframe.equals(new_dataframe):
            current_dataframe = new_dataframe
            return data_version.value + 1  # Increment version to trigger updates
        return data_version.value
    
    timer.tick(fn=update_current_dataframe, outputs=data_version)
    
    # TODO: Add Ginkgo logo here on the top right
    gr.Markdown("""
        ## Welcome to the Ginkgo Antibody Developability Benchmark!

        **Beta version, not publicly launched yet**

        Participants can submit their model to the leaderboard by uploading a CSV file (see the "✉️ Submit" tab).
        See more details in the "❔About" tab.
        """)
    with gr.Tabs(elem_classes="tab-buttons"):
        with gr.TabItem("❔About", elem_id="abdev-benchmark-tab-table"):
            gr.Image(
                value="./assets/competition_logo.jpg",
                show_label=False,
                # elem_classes=["resized-image"],
                show_download_button=False,
                width="50vw",  # 50% of the "viewport width"
            )
            gr.Markdown(ABOUT_INTRO)
            gr.Image(value="./assets/prediction_explainer.png", show_label=False, show_download_button=False, width="50vw")
            gr.Markdown(ABOUT_TEXT)
            gr.Markdown("### FAQs")
            for i, (question, answer) in enumerate(FAQS.items()):
                # Would love to make questions bold but accordion doesn't support it
                question = f"{i+1}. {question}"
                with gr.Accordion(question, open=False):
                    gr.Markdown(f"*{answer}*")  # Italics for answers

        # Procedurally make these 5 tabs
        for i, assay in enumerate(ASSAY_LIST):
            with gr.TabItem(
                f"{ASSAY_EMOJIS[assay]} {ASSAY_RENAME[assay]}",
                elem_id="abdev-benchmark-tab-table",
            ) as tab_item:
                gr.Markdown(f"# {ASSAY_DESCRIPTION[assay]}")
                lb = get_leaderboard_object(assay=assay)
                
                def refresh_leaderboard(assay=assay):
                    return format_leaderboard_table(df_results=current_dataframe, assay=assay)
                
                # Refresh when data version changes
                data_version.change(fn=refresh_leaderboard, outputs=lb)

        with gr.TabItem("🚀 Overall", elem_id="abdev-benchmark-tab-table") as overall_tab:
            gr.Markdown(
                "# Antibody Developability Benchmark Leaderboard over all properties"
            )
            lb = get_leaderboard_object()
            
            def refresh_overall_leaderboard():
                return format_leaderboard_table(df_results=current_dataframe)
            
            # Refresh when data version changes
            data_version.change(fn=refresh_overall_leaderboard, outputs=lb)

        with gr.TabItem("✉️ Submit", elem_id="boundary-benchmark-tab-table"):
            gr.Markdown(
            f"""
            # Antibody Developability Submission
            Upload a CSV to get a score!
            List of valid property names: `{', '.join(ASSAY_LIST)}`.

            Please use your Hugging Face account name to submit your model - we use this to track separate submissions, and if you would like to remain anonymous please set up an anonymous huggingface account.
            Your submission will be evaluated and added to the leaderboard.
            """
            )
            submission_type_state = gr.State(value="GDPa1")
            download_file_state = gr.State(value=EXAMPLE_FILE_DICT["GDPa1"])

            with gr.Row():
                with gr.Column():
                    username_input = gr.Textbox(
                        label="Username",
                        placeholder="Enter your Hugging Face username",
                        info="This will be used to track your submissions, and to update your results if you submit again.",
                    )
                    model_name_input = gr.Textbox(
                        label="Model Name",
                        placeholder="Enter your model name (e.g., 'MyProteinLM-v1')",
                        info="This will be displayed on the leaderboard.",
                    )
                    model_description_input = gr.Textbox(
                        label="Model Description (optional)",
                        placeholder="Brief description of your model and approach",
                        info="Describe your model, training data, or methodology.",
                        lines=3,
                    )
                    anonymous_checkbox = gr.Checkbox(
                        label="Anonymous",
                        value=False,
                        info="If checked, your username will be replaced with an anonymous username on the leaderboard.",
                    )
                    registration_code = gr.Textbox(
                        label="Registration Code",
                        placeholder="Enter your registration code",
                        info="If you did not receive a registration code, please sign up on the <a href='https://datapoints.ginkgo.bio/ai-competitions/2025-abdev-competition'>Competition Registration page</a> or email <a href='mailto:[email protected]'>[email protected]</a>.",
                    )
                with gr.Column():
                    submission_type_dropdown = gr.Dropdown(
                        choices=["GDPa1", "GDPa1_cross_validation", "Heldout Test Set"],
                        value="GDPa1",
                        label="Submission Type",
                    )
                    download_button = gr.DownloadButton(
                        label="📥 Download example submission CSV for GDPa1",
                        value=EXAMPLE_FILE_DICT["GDPa1"],
                        variant="secondary",
                    )
                    submission_file = gr.File(label="Submission CSV")

            def update_submission_type_and_file(submission_type):
                """
                Based on the submission type selected in the dropdown,
                    Update the submission type state
                    Dynamically update example file for download
                """
                download_file = EXAMPLE_FILE_DICT.get(
                    submission_type, EXAMPLE_FILE_DICT[submission_type]
                )
                download_label = (
                    f"📥 Download example submission CSV for {submission_type}"
                )
                return (
                    submission_type,
                    download_file,
                    gr.DownloadButton(
                        label=download_label,
                        value=download_file,
                        variant="secondary",
                    ),
                )

            # Update submission type state and download button when dropdown changes
            submission_type_dropdown.change(
                fn=update_submission_type_and_file,
                inputs=submission_type_dropdown,
                outputs=[submission_type_state, download_file_state, download_button],
            )

            submit_btn = gr.Button("Evaluate")
            message = gr.Textbox(label="Status", lines=1, visible=False)
            # help message
            gr.Markdown(
                "If you have issues with submission or using the leaderboard, please start a discussion in the Community tab of this Space."
            )

            submit_btn.click(
                make_submission,
                inputs=[
                    submission_file,
                    username_input,
                    submission_type_state,
                    model_name_input,
                    model_description_input,
                    anonymous_checkbox,
                    registration_code,
                ],
                outputs=[message],
            ).then(
                fn=show_output_box,
                inputs=[message],
                outputs=[message],
            )
    # Footnote
    gr.Markdown(
        """
        <div style="text-align: center; font-size: 14px; color: gray; margin-top: 2em;">
        📬 For questions or feedback, contact <a href="mailto:[email protected]">[email protected]</a> or visit the Community tab at the top of this page.<br>
        Visit the <a href="https://datapoints.ginkgo.bio/ai-competitions/2025-abdev-competition">Competition Registration page</a> to sign up for updates and to register a team.
        </div>
        """,
        elem_id="contact-footer",
    )

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