df.sort_values("amount", ascending=False).head(10)
df.sort_values(["region", "amount"], ascending=[True, False])
df.sort_index() # by row indexdf["region"].value_counts() # counts per region
df["region"].value_counts(normalize=True) # as percentages
df["region"].nunique() # distinct countdf["rank_by_amount"] = df["amount"].rank(ascending=False)
df["pct_rank"] = df["amount"].rank(pct=True)df.nlargest(10, "amount")
df.nsmallest(5, "amount")sort_values(["a","b"], ascending=[True,False]) handles multi-key sort.value_counts() is the one-liner for a quick frequency table.nlargest() / nsmallest() are faster and clearer than sort+head.From an orders DataFrame, print the top 5 products by total revenue.