data leaderboard 2

app.py

@@ -4,6 +4,7 @@ import pandas as pd
 import time
 import threading
 from huggingface_hub import HfApi
+from humanize import naturalsize
 
 api = HfApi()
 
@@ -38,13 +39,15 @@ df = pd.read_csv("author_data_hf_merged.csv")
 df_author_copy = df.copy()
 
 df["author"] = df["author"].apply(lambda x: clickable(x))
-df = df.sort_values(by='models', ascending=False)
 df['Serial Number'] = [i for i in range(1, len(df)+1)]
 df['Total Usage'] = df[['models', 'datasets', 'spaces']].sum(axis=1)
 df = df[['Serial Number', "author", "Total Usage", "models", "datasets", "spaces"]]
+df = df.sort_values(by='Total Usage', ascending=False)
 
-df = apply_headers(df, ["🔢 Serial Number", "👤 Author", "⚡️ Total Usage", "🏛️ Models", "📊 Datasets", "🚀 Spaces"])
+naturalsize_columns = ['Total Usage', 'models', 'datasets', 'spaces']
+df[cols_to_naturalsize] = df[naturalsize_columns].applymap(humanize.naturalsize)
 
+df = apply_headers(df, ["🔢 Serial Number", "👤 Author", "⚡️ Total Usage", "🏛️ Models", "📊 Datasets", "🚀 Spaces"])
 
 desc = f"""
 🎯 The Leaderboard aims to track authors data usage in 🤗 Huggingface.