core/org.jetbrains.kotlinx.dataframe.api/maxFor

maxFor

fun <T, C : Comparable<*>?> DataFrame<T>.maxFor(skipNaN: Boolean = skipNaNDefault, columns: ColumnsForAggregateSelector<T, C>): DataRow<T>

Returns the maximum of the values of each selected column of this DataFrame separately.

Only self-comparable values are supported: values of a type T : Comparable<T> that are mutually comparable (like strings, primitive numbers, or dates). This includes all primitive number types, but no mix of different number types, which would result in a ClassCastException.

null values in the input are always ignored.

If the input contains NaN values, the result will be NaN, unless skipNaN is set to true.

Result cells for which there is nothing left to compare (for instance, because the input was empty or contained only null values) simply become null.

For more information about the resulting types: See "min / max Type Conversion" on the documentation website.

The columns are selected with the ColumnsForAggregateSelectionDsl — an extension of the Columns Selection DSL which lets you rename the result of a column with into and supply a default value for columns without any values.

This can include column groups and nested columns.

See Selecting Columns.

See also:

max() — the same, but for all suitable columns at once.
max{ columns } — a single maximum of all values in the selected columns.
minFor — the mirror operation.
The Max Operation — an overview of all max modes.

For more information: See "min / max" on the documentation website.

Example

// A single row with the largest "age" and the largest "weight"
df.maxFor { age and weight }
// The same, ignoring `NaN` values, and naming the results explicitly
df.maxFor(skipNaN = true) { age into "maxAge" and (weight into "maxWeight") }

Return

A single DataRow with the maximum of each selected column.

Parameters

skipNaN

If true, NaN values are ignored, just like null values. If false (the default), a NaN in the input is propagated to the result. This only has an effect on Double and Float values.

columns

The ColumnsForAggregateSelector used to select the columns to compute the maximum of.

fun <T> DataFrame<T>.maxFor(vararg columns: String, skipNaN: Boolean = skipNaNDefault): DataRow<T>

Returns the maximum of the values of each selected column of this DataFrame separately.

null values in the input are always ignored.

If the input contains NaN values, the result will be NaN, unless skipNaN is set to true.

Result cells for which there is nothing left to compare (for instance, because the input was empty or contained only null values) simply become null.

For more information about the resulting types: See "min / max Type Conversion" on the documentation website.

This can include column groups and nested columns.

See Selecting Columns.

See also:

max() — the same, but for all suitable columns at once.
max{ columns } — a single maximum of all values in the selected columns.
minFor — the mirror operation.
The Max Operation — an overview of all max modes.

For more information: See "min / max" on the documentation website.

Example

// A single row with the largest "age" and the largest "weight"
df.maxFor("age", "weight")

Return

A single DataRow with the maximum of each selected column.

Parameters

columns

The names of the columns to compute the maximum of. These must be existing mutually comparable columns, else an IllegalStateException or IllegalArgumentException is thrown.

skipNaN

If true, NaN values are ignored, just like null values. If false (the default), a NaN in the input is propagated to the result. This only has an effect on Double and Float values.

fun <T, C : Comparable<*>?> Grouped<T>.maxFor(skipNaN: Boolean = skipNaNDefault, columns: ColumnsForAggregateSelector<T, C>): DataFrame<T>

Aggregates this GroupBy by computing the maximum of the values of each selected column separately, per group.

Returns a new DataFrame with one row per group, containing the group key columns and a column with the maximum for each selected column.

null values in the input are always ignored.

If the input contains NaN values, the result will be NaN, unless skipNaN is set to true.

Result cells for which there is nothing left to compare (for instance, because the input was empty or contained only null values) simply become null.

For more information about the resulting types: See "min / max Type Conversion" on the documentation website.

This can include column groups and nested columns.

See Selecting Columns.

See also:

max() — the same, but for all suitable columns at once.
max{ columns } — a single maximum of all values in the selected columns, per group.
minFor — the mirror operation.
aggregate — the general way to aggregate groups.
The Max Operation — an overview of all max modes.

For more information: See "groupBy statistics" on the documentation website., and See "GroupBy Aggregation Statistics" on the documentation website.

Example

// For each city, the largest "age" and the largest "weight"
df.groupBy { city }.maxFor { age and weight }

Return

A new DataFrame with the group keys and the maximum of each selected column per group.

Parameters

skipNaN

If true, NaN values are ignored, just like null values. If false (the default), a NaN in the input is propagated to the result. This only has an effect on Double and Float values.

columns

The ColumnsForAggregateSelector used to select the columns to compute the maximum of.

fun <T> Grouped<T>.maxFor(vararg columns: String, skipNaN: Boolean = skipNaNDefault): DataFrame<T>

Aggregates this GroupBy by computing the maximum of the values of each selected column separately, per group.

Returns a new DataFrame with one row per group, containing the group key columns and a column with the maximum for each selected column.

null values in the input are always ignored.

If the input contains NaN values, the result will be NaN, unless skipNaN is set to true.

Result cells for which there is nothing left to compare (for instance, because the input was empty or contained only null values) simply become null.

For more information about the resulting types: See "min / max Type Conversion" on the documentation website.

This can include column groups and nested columns.

See Selecting Columns.

See also:

max() — the same, but for all suitable columns at once.
max{ columns } — a single maximum of all values in the selected columns, per group.
minFor — the mirror operation.
aggregate — the general way to aggregate groups.
The Max Operation — an overview of all max modes.

For more information: See "groupBy statistics" on the documentation website., and See "GroupBy Aggregation Statistics" on the documentation website.

Example

// For each city, the largest "age" and the largest "weight"
df.groupBy { city }.maxFor("age", "weight")

Return

A new DataFrame with the group keys and the maximum of each selected column per group.

Parameters

columns

The names of the columns to compute the maximum of. These must be existing mutually comparable columns, else an IllegalStateException or IllegalArgumentException is thrown.

skipNaN

If true, NaN values are ignored, just like null values. If false (the default), a NaN in the input is propagated to the result. This only has an effect on Double and Float values.

fun <T, R : Comparable<*>?> Pivot<T>.maxFor(separate: Boolean = false, skipNaN: Boolean = skipNaNDefault, columns: ColumnsForAggregateSelector<T, R>): DataRow<T>

Aggregates this Pivot by computing the maximum of the values of each selected column separately, per group.

Returns a single DataRow with the pivot keys as (nested) columns, containing the maximum of each selected column of the corresponding group.

null values in the input are always ignored.

If the input contains NaN values, the result will be NaN, unless skipNaN is set to true.

Result cells for which there is nothing left to compare (for instance, because the input was empty or contained only null values) simply become null.

For more information about the resulting types: See "min / max Type Conversion" on the documentation website.

This can include column groups and nested columns.

See Selecting Columns, or check out the Pivot Grammar.

See also:

max() — the same, but for all suitable columns at once.
max{ columns } — a single maximum of all values in the selected columns, per group.
minFor — the mirror operation.
Pivot aggregation — all other ways to aggregate a Pivot.
The Max Operation — an overview of all max modes.

For more information: See "pivot statistics" on the documentation website., and See "Pivot Aggregation statistics" on the documentation website.

Example

// For each city, the largest "age" and the largest "weight"
df.pivot { city }.maxFor { age and weight }
// The same, but with the results grouped by aggregated column instead of by city
df.pivot { city }.maxFor(separate = true) { age and weight }

Return

A single DataRow with the maximum of each selected column per pivot group.

Parameters

separate

If false (the default), the resulting columns are indexed first by the pivot key(s) and then by the names of the aggregated columns. If true, this order is reversed: the results are grouped by aggregated column first.

skipNaN

If true, NaN values are ignored, just like null values. If false (the default), a NaN in the input is propagated to the result. This only has an effect on Double and Float values.

columns

The ColumnsForAggregateSelector used to select the columns to compute the maximum of.

fun <T> Pivot<T>.maxFor(vararg columns: String, separate: Boolean = false, skipNaN: Boolean = skipNaNDefault): DataRow<T>

Aggregates this Pivot by computing the maximum of the values of each selected column separately, per group.

Returns a single DataRow with the pivot keys as (nested) columns, containing the maximum of each selected column of the corresponding group.

null values in the input are always ignored.

If the input contains NaN values, the result will be NaN, unless skipNaN is set to true.

Result cells for which there is nothing left to compare (for instance, because the input was empty or contained only null values) simply become null.

For more information about the resulting types: See "min / max Type Conversion" on the documentation website.

This can include column groups and nested columns.

See Selecting Columns, or check out the Pivot Grammar.

See also:

max() — the same, but for all suitable columns at once.
max{ columns } — a single maximum of all values in the selected columns, per group.
minFor — the mirror operation.
Pivot aggregation — all other ways to aggregate a Pivot.
The Max Operation — an overview of all max modes.

For more information: See "pivot statistics" on the documentation website., and See "Pivot Aggregation statistics" on the documentation website.

Example

// For each city, the largest "age" and the largest "weight"
df.pivot { city }.maxFor("age", "weight")

Return

A single DataRow with the maximum of each selected column per pivot group.

Parameters

columns

The names of the columns to compute the maximum of. These must be existing mutually comparable columns, else an IllegalStateException or IllegalArgumentException is thrown.

separate

skipNaN

If true, NaN values are ignored, just like null values. If false (the default), a NaN in the input is propagated to the result. This only has an effect on Double and Float values.

fun <T, R : Comparable<*>?> PivotGroupBy<T>.maxFor(separate: Boolean = false, skipNaN: Boolean = skipNaNDefault, columns: ColumnsForAggregateSelector<T, R>): DataFrame<T>

Aggregates this PivotGroupBy by computing the maximum of the values of each selected column separately, per group.

Returns a DataFrame where each cell contains the maximum of each selected column of the group corresponding to that pivot key (column) and groupBy key (row).

null values in the input are always ignored.

If the input contains NaN values, the result will be NaN, unless skipNaN is set to true.

Result cells for which there exists a group, but there is nothing left to compare (for instance, because the group was empty or contained only null values) simply become null.

For more information about the resulting types: See "min / max Type Conversion" on the documentation website.

For empty pivot intersections, null or the set default is used.

This can include column groups and nested columns.

See Selecting Columns, or check out the PivotGroupBy Grammar.

See also:

max() — the same, but for all suitable columns at once.
max{ columns } — a single maximum of all values in the selected columns, per group.
minFor — the mirror operation.
PivotGroupBy aggregation — all other ways to aggregate a PivotGroupBy.
The Max Operation — an overview of all max modes.

For more information: See "pivot statistics" on the documentation website., and See "Pivot Aggregation statistics" on the documentation website.

Example

// Per city and last name, the largest "age" and the largest "weight"
df.pivot { city }.groupBy { name.lastName }.maxFor { age and weight }

Return

A DataFrame with the maximum of each selected column per group.

Parameters

separate

skipNaN

If true, NaN values are ignored, just like null values. If false (the default), a NaN in the input is propagated to the result. This only has an effect on Double and Float values.

columns

The ColumnsForAggregateSelector used to select the columns to compute the maximum of.

fun <T> PivotGroupBy<T>.maxFor(vararg columns: String, separate: Boolean = false, skipNaN: Boolean = skipNaNDefault): DataFrame<T>

Aggregates this PivotGroupBy by computing the maximum of the values of each selected column separately, per group.

Returns a DataFrame where each cell contains the maximum of each selected column of the group corresponding to that pivot key (column) and groupBy key (row).

null values in the input are always ignored.

If the input contains NaN values, the result will be NaN, unless skipNaN is set to true.

Result cells for which there exists a group, but there is nothing left to compare (for instance, because the group was empty or contained only null values) simply become null.

For more information about the resulting types: See "min / max Type Conversion" on the documentation website.

For empty pivot intersections, null or the set default is used.

This can include column groups and nested columns.

See Selecting Columns, or check out the PivotGroupBy Grammar.

See also:

max() — the same, but for all suitable columns at once.
max{ columns } — a single maximum of all values in the selected columns, per group.
minFor — the mirror operation.
PivotGroupBy aggregation — all other ways to aggregate a PivotGroupBy.
The Max Operation — an overview of all max modes.

For more information: See "pivot statistics" on the documentation website., and See "Pivot Aggregation statistics" on the documentation website.

Example

// Per city and last name, the largest "age" and the largest "weight"
df.pivot { city }.groupBy { name.lastName }.maxFor("age", "weight")

Return

A DataFrame with the maximum of each selected column per group.

Parameters

columns

The names of the columns to compute the maximum of. These must be existing mutually comparable columns, else an IllegalStateException or IllegalArgumentException is thrown.

separate

skipNaN

If true, NaN values are ignored, just like null values. If false (the default), a NaN in the input is propagated to the result. This only has an effect on Double and Float values.

fun <T, C : Comparable<*>?> DataFrame<T>.~~maxFor~~(vararg columns: ColumnReference<C>, skipNaN: Boolean = skipNaNDefault): DataRow<T>

fun <T, C : Comparable<*>?> DataFrame<T>.~~maxFor~~(vararg columns: KProperty<C>, skipNaN: Boolean = skipNaNDefault): DataRow<T>

fun <T, C : Comparable<*>?> Grouped<T>.~~maxFor~~(vararg columns: ColumnReference<C>, skipNaN: Boolean = skipNaNDefault): DataFrame<T>

fun <T, C : Comparable<*>?> Grouped<T>.~~maxFor~~(vararg columns: KProperty<C>, skipNaN: Boolean = skipNaNDefault): DataFrame<T>

fun <T, R : Comparable<*>?> Pivot<T>.~~maxFor~~(vararg columns: ColumnReference<R>, separate: Boolean = false, skipNaN: Boolean = skipNaNDefault): DataRow<T>

fun <T, R : Comparable<*>?> Pivot<T>.~~maxFor~~(vararg columns: KProperty<R>, separate: Boolean = false, skipNaN: Boolean = skipNaNDefault): DataRow<T>

fun <T, R : Comparable<*>?> PivotGroupBy<T>.~~maxFor~~(vararg columns: ColumnReference<R>, separate: Boolean = false, skipNaN: Boolean = skipNaNDefault): DataFrame<T>

fun <T, R : Comparable<*>?> PivotGroupBy<T>.~~maxFor~~(vararg columns: KProperty<R>, separate: Boolean = false, skipNaN: Boolean = skipNaNDefault): DataFrame<T>

Deprecated

It's recommended to migrate to the String or Extension Properties Access API https://kotlin.github.io/dataframe/apilevels.html