DataFrameStatFunctions¶

DataFrameStatFunctions is used to work with <> in a structured query (a <>).

[[methods]] .DataFrameStatFunctions API [cols="1,2",options="header",width="100%"] |=== | Method | Description

| <> a|

[source, scala]¶

approxQuantile( cols: Array[String], probabilities: Array[Double], relativeError: Double): Array[Array[Double]] approxQuantile( col: String, probabilities: Array[Double], relativeError: Double): Array[Double]

| <> a|

[source, scala]¶

bloomFilter(col: Column, expectedNumItems: Long, fpp: Double): BloomFilter bloomFilter(col: Column, expectedNumItems: Long, numBits: Long): BloomFilter bloomFilter(colName: String, expectedNumItems: Long, fpp: Double): BloomFilter bloomFilter(colName: String, expectedNumItems: Long, numBits: Long): BloomFilter

| <> a|

[source, scala]¶

corr(col1: String, col2: String): Double corr(col1: String, col2: String, method: String): Double

| <> a|

[source, scala]¶

countMinSketch(col: Column, eps: Double, confidence: Double, seed: Int): CountMinSketch countMinSketch(col: Column, depth: Int, width: Int, seed: Int): CountMinSketch countMinSketch(colName: String, eps: Double, confidence: Double, seed: Int): CountMinSketch countMinSketch(colName: String, depth: Int, width: Int, seed: Int): CountMinSketch

| <> a|

[source, scala]¶

cov(col1: String, col2: String): Double¶

| <> a|

[source, scala]¶

crosstab(col1: String, col2: String): DataFrame¶

| <> a|

[source, scala]¶

freqItems(cols: Array[String]): DataFrame freqItems(cols: Array[String], support: Double): DataFrame freqItems(cols: Seq[String]): DataFrame freqItems(cols: Seq[String], support: Double): DataFrame

| <> a|

[source, scala]¶

sampleByT: DataFrame¶

|===

[[creating-instance]] DataFrameStatFunctions is available using stat untyped transformation.

[source, scala]¶

val q: DataFrame = ... q.stat

=== [[approxQuantile]] approxQuantile Method

[source, scala]¶

approxQuantile( cols: Array[String], probabilities: Array[Double], relativeError: Double): Array[Array[Double]] approxQuantile( col: String, probabilities: Array[Double], relativeError: Double): Array[Double]

approxQuantile...FIXME

=== [[bloomFilter]] bloomFilter Method

[source, scala]¶

bloomFilter(col: Column, expectedNumItems: Long, fpp: Double): BloomFilter bloomFilter(col: Column, expectedNumItems: Long, numBits: Long): BloomFilter bloomFilter(colName: String, expectedNumItems: Long, fpp: Double): BloomFilter bloomFilter(colName: String, expectedNumItems: Long, numBits: Long): BloomFilter

bloomFilter...FIXME

=== [[buildBloomFilter]] buildBloomFilter Internal Method

[source, scala]¶

buildBloomFilter(col: Column, zero: BloomFilter): BloomFilter¶

buildBloomFilter...FIXME

NOTE: convertToDouble is used when...FIXME

=== [[corr]] corr Method

[source, scala]¶

corr(col1: String, col2: String): Double corr(col1: String, col2: String, method: String): Double

corr...FIXME

=== [[countMinSketch]] countMinSketch Method

[source, scala]¶

countMinSketch(col: Column, eps: Double, confidence: Double, seed: Int): CountMinSketch countMinSketch(col: Column, depth: Int, width: Int, seed: Int): CountMinSketch countMinSketch(colName: String, eps: Double, confidence: Double, seed: Int): CountMinSketch countMinSketch(colName: String, depth: Int, width: Int, seed: Int): CountMinSketch // PRIVATE API countMinSketch(col: Column, zero: CountMinSketch): CountMinSketch

countMinSketch...FIXME

=== [[cov]] cov Method

[source, scala]¶

cov(col1: String, col2: String): Double¶

cov...FIXME

=== [[crosstab]] crosstab Method

[source, scala]¶

crosstab(col1: String, col2: String): DataFrame¶

crosstab...FIXME

=== [[freqItems]] freqItems Method

[source, scala]¶

freqItems(cols: Array[String]): DataFrame freqItems(cols: Array[String], support: Double): DataFrame freqItems(cols: Seq[String]): DataFrame freqItems(cols: Seq[String], support: Double): DataFrame

freqItems...FIXME

=== [[sampleBy]] sampleBy Method

[source, scala]¶

sampleByT: DataFrame¶

sampleBy...FIXME