app.py

__all__ = ['block', 'make_clickable_model', 'make_clickable_user', 'get_submissions']

import gradio as gr
import pandas as pd
from huggingface_hub import HfApi, repocard


#def is_duplicated(space_id:str)->None:
#    card = repocard.RepoCard.load(space_id, repo_type="space")
#    return getattr(card.data, "duplicated_from", None) is not None



def make_clickable_model(model_name, repo_type, link=None):
    if link is None:
      if repo_type == "Dataset":
        link = "https://huggingface.co/" + "datasets/" + model_name
      else:
        link = "https://huggingface.co/" + model_name
    return f'<a target="_blank" href="{link}">{model_name.split("/")[-1]}</a>'

def get_repo_ids(repo_type):
    api = HfApi()
    if repo_type == "Model":
      notebooks = api.list_models(filter=["notebook-favorites"])
    elif repo_type == "Dataset":
      notebooks = api.list_datasets(filter=["notebook-favorites"])
    print(notebooks)
    notebook_ids = [x for x in notebooks]
    return notebook_ids


def make_clickable_user(user_id):
    link = "https://huggingface.co/" + user_id
    return f'<a  target="_blank" href="{link}">{user_id}</a>'

def get_submissions(repo_type):
    submissions = get_repo_ids(repo_type)
    leaderboard_models = []

    for submission in submissions:
        # user, model, likes
        #if not is_duplicated(submission.id):
        user_id = submission.id.split("/")[0]
        leaderboard_models.append(
            (
                make_clickable_user(user_id),
                make_clickable_model(submission.id, repo_type),
                submission.likes,
            )
        )

    df = pd.DataFrame(data=leaderboard_models, columns=["User", "Repository", "Likes"])
    df.sort_values(by=["Likes"], ascending=False, inplace=True)
    df.insert(0, "Rank", list(range(1, len(df) + 1)))
    return df



block = gr.Blocks()

with block:
    gr.Markdown(
        """# Notebooks Leaderboard

    This Space compiles coolest model and dataset repositories that contain notebooks!
    """
    )
    with gr.Tabs():
        with gr.TabItem("Notebooks in Model Repositories 🤖"):
            with gr.Row():
                model_data = gr.components.Dataframe(
                    type="pandas", datatype=["number", "markdown", "markdown", "number"]
                )
            with gr.Row():
                data_run = gr.Button("Refresh")
                data_run.click(
                    get_submissions, inputs=gr.Variable("Model"), outputs=model_data
                )
        with gr.TabItem("Notebooks in Dataset Repositories 📖"):
            with gr.Row():
                dataset_data = gr.components.Dataframe(
                    type="pandas", datatype=["number", "markdown", "markdown", "number"]
                )
            with gr.Row():
                data_run = gr.Button("Refresh")
                data_run.click(
                    get_submissions, inputs=gr.Variable("Dataset"), outputs=dataset_data
                )

    block.load(get_submissions, inputs=gr.Variable("Model"), outputs=model_data)
    block.load(get_submissions, inputs=gr.Variable("Dataset"), outputs=dataset_data)


block.launch(debug=True)