GroupBy API

GroupBy operations for aggregating data across groups. Available for both Series and DataFrame.

Usage

Series GroupBy

const sales = new Series([100, 200, 150, 300, 250], { name: "sales" });
const regions = ["North", "North", "South", "South", "South"];
const grouped = sales.groupBy(regions); // Group by region labels
// This creates groups: "North" contains [100, 200], "South" contains [150, 300, 250]

How Series GroupBy works:

• You provide an array of group labels that matches the Series length
• Each position in the Series gets assigned to a group based on the label at that position
• Values with the same label are grouped together

DataFrame GroupBy

const df = new DataFrame({
    name: ["Alice", "Bob", "Charlie", "Alice", "Bob"],
    age: [25, 30, 35, 28, 32],
    salary: [50000, 60000, 70000, 55000, 65000]
});
const grouped = df.groupBy("name"); // Group by column

Methods

Data Access

• groups() - Get the groups as a Map
• getGroup(name: string) - Get a specific group (Series or DataFrame)
• size() - Get the size of each group

Aggregation Methods

• agg(aggregation: AggSpec) - Apply aggregation function(s)
• sum() - Calculate sum for each group
• mean() - Calculate mean for each group
• count() - Calculate count for each group
• min() - Calculate minimum for each group
• max() - Calculate maximum for each group
• std() - Calculate standard deviation for each group
• var() - Calculate variance for each group
• first() - Get first value for each group
• last() - Get last value for each group

Examples

Series GroupBy

const sales = new Series([100, 200, 150, 300, 250], { name: "sales" });
const regions = ["North", "North", "South", "South", "South"];
const grouped = sales.groupBy(regions);
const result = grouped.sum(); // Returns Series with total sales by region

DataFrame GroupBy

const df = new DataFrame({
    name: ["Alice", "Bob", "Charlie", "Alice", "Bob"],
    age: [25, 30, 35, 28, 32],
    salary: [50000, 60000, 70000, 55000, 65000]
});

// Single aggregation
const byName = df.groupBy("name");
const avgAge = byName.mean().get("age");

// Multiple aggregations
const result = df.groupBy("name").agg({
    age: "mean",
    salary: ["sum", "max"]
});

Return Types

• Series GroupBy → Returns Series for single aggregations
• DataFrame GroupBy → Returns DataFrame for all aggregations