rank data added

environment.yml

@@ -17,4 +17,5 @@ dependencies:
       - dask>=2023.1.0
       - plotly
       - mkdocs-material
-      - yaml
+      - yaml
+      - scipy

filedownloadstat/parquet_analyzer.py

@@ -1,5 +1,6 @@
 import os
 import dask.dataframe as dd
+from scipy.stats import rankdata
 
 class ParquetAnalyzer:
     def __init__(self):
@@ -32,6 +33,12 @@ def persist_project_level_download_counts(self, ddf, project_level_download_coun
         # Compute the result
         result = project_level_counts.compute()
 
+        # Calculate percentile rank for the 'count' column
+        result["percentile"] = (rankdata(result["count"], method="average") / len(result) * 100).astype(int)
+
+        # Sort by count in descending order
+        result = result.sort_values(by="count", ascending=False)
+
         # Save to JSON
         result.to_json(project_level_download_counts, orient="records", lines=False)