app.py

import gradio as gr
from huggingface_hub import list_models, list_datasets, list_spaces
import pandas as pd
from datetime import datetime


def get_user_stats():
    users = {}
    for k, fn in zip(['model', 'dataset', 'space'], [list_models, list_datasets, list_spaces]):
        for repo in fn(full=True):
            if repo.author is None:
                continue

            if repo.author not in users:
                users[repo.author] = {
                    x: 0 for x in [
                        'model_likes',
                        'num_models',
                        'dataset_likes',
                        'num_datasets',
                        'space_likes',
                        'num_spaces',
                        'total_likes',
                        'total_repos'
                    ]
                }
            users[repo.author][f"{k}_likes"] += repo.likes
            users[repo.author][f"num_{k}s"] += 1
    
    for username, user_stats in users.items():
        users[username]['total_likes'] += sum([v for k, v in user_stats.items() if "likes" in k])
        users[username]['total_repos'] += sum([v for k, v in user_stats.items() if "num_" in k])

    for k, v in users.items():
        users[k] = dict(users[k])

    return users

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


def get_user_stats_df(limit=1000):
    users = get_user_stats()
    df = pd.DataFrame([{'username': make_clickable_user(k), **v} for k, v in users.items()])
    df.sort_values(by=["total_likes"], ascending=False, inplace=True)
    df.insert(0, "rank", list(range(1, len(df) + 1)))
    df = df[
        [
            'rank',
            'username',
            'total_likes',
            'model_likes',
            'num_models',
            'dataset_likes',
            'num_datasets',
            'space_likes',
            'num_spaces',
            'total_repos'
        ]
    ]
    if limit:
        df = df.head(limit)
    return df

df = get_user_stats_df()
desc = f"""
# 🤗 Hugging Face User Stats

Here are some stats on the top 1000 users/organizations on the Hugging Face Hub. **Updated {datetime.now().strftime("%d/%m/%y")}**
"""
with gr.Blocks() as demo:
    gr.Markdown(desc)
    data = gr.components.Dataframe(
        df,
        type="pandas",
        datatype=["number", "markdown", "number", "number", "number", "number", "number", "number", "number", "number"],
    )
demo.launch()