spark linearRegression 源码

spark linearRegression 代码

文件路径：/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/aggregate/linearRegression.scala

/*
 * Licensed to the Apache Software Foundation (ASF) under one or more
 * contributor license agreements.  See the NOTICE file distributed with
 * this work for additional information regarding copyright ownership.
 * The ASF licenses this file to You under the Apache License, Version 2.0
 * (the "License"); you may not use this file except in compliance with
 * the License.  You may obtain a copy of the License at
 *
 *    http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */

package org.apache.spark.sql.catalyst.expressions.aggregate

import org.apache.spark.sql.catalyst.dsl.expressions._
import org.apache.spark.sql.catalyst.expressions.{And, AttributeReference, Expression, ExpressionDescription, If, ImplicitCastInputTypes, IsNotNull, IsNull, Literal, Or, RuntimeReplaceableAggregate}
import org.apache.spark.sql.catalyst.trees.BinaryLike
import org.apache.spark.sql.types.{AbstractDataType, DataType, DoubleType, NumericType}

// scalastyle:off line.size.limit
@ExpressionDescription(
  usage = "_FUNC_(y, x) - Returns the number of non-null number pairs in a group, where `y` is the dependent variable and `x` is the independent variable.",
  examples = """
    Examples:
      > SELECT _FUNC_(y, x) FROM VALUES (1, 2), (2, 2), (2, 3), (2, 4) AS tab(y, x);
       4
      > SELECT _FUNC_(y, x) FROM VALUES (1, 2), (2, null), (2, 3), (2, 4) AS tab(y, x);
       3
      > SELECT _FUNC_(y, x) FROM VALUES (1, 2), (2, null), (null, 3), (2, 4) AS tab(y, x);
       2
  """,
  group = "agg_funcs",
  since = "3.3.0")
// scalastyle:on line.size.limit
case class RegrCount(left: Expression, right: Expression)
  extends AggregateFunction
  with RuntimeReplaceableAggregate
  with ImplicitCastInputTypes
  with BinaryLike[Expression] {
  override lazy val replacement: Expression = Count(Seq(left, right))
  override def nodeName: String = "regr_count"
  override def inputTypes: Seq[AbstractDataType] = Seq(NumericType, NumericType)
  override protected def withNewChildrenInternal(
      newLeft: Expression, newRight: Expression): RegrCount =
    this.copy(left = newLeft, right = newRight)
}

// scalastyle:off line.size.limit
@ExpressionDescription(
  usage = "_FUNC_(y, x) - Returns the average of the independent variable for non-null pairs in a group, where `y` is the dependent variable and `x` is the independent variable.",
  examples = """
    Examples:
      > SELECT _FUNC_(y, x) FROM VALUES (1, 2), (2, 2), (2, 3), (2, 4) AS tab(y, x);
       2.75
      > SELECT _FUNC_(y, x) FROM VALUES (1, null) AS tab(y, x);
       NULL
      > SELECT _FUNC_(y, x) FROM VALUES (null, 1) AS tab(y, x);
       NULL
      > SELECT _FUNC_(y, x) FROM VALUES (1, 2), (2, null), (2, 3), (2, 4) AS tab(y, x);
       3.0
      > SELECT _FUNC_(y, x) FROM VALUES (1, 2), (2, null), (null, 3), (2, 4) AS tab(y, x);
       3.0
  """,
  group = "agg_funcs",
  since = "3.3.0")
// scalastyle:on line.size.limit
case class RegrAvgX(
    left: Expression,
    right: Expression)
  extends AggregateFunction
  with RuntimeReplaceableAggregate
  with ImplicitCastInputTypes
  with BinaryLike[Expression] {
  override lazy val replacement: Expression =
    Average(If(And(IsNotNull(left), IsNotNull(right)), right, Literal.create(null, right.dataType)))
  override def nodeName: String = "regr_avgx"
  override def inputTypes: Seq[AbstractDataType] = Seq(NumericType, NumericType)
  override protected def withNewChildrenInternal(
      newLeft: Expression, newRight: Expression): RegrAvgX =
    this.copy(left = newLeft, right = newRight)
}

// scalastyle:off line.size.limit
@ExpressionDescription(
  usage = "_FUNC_(y, x) - Returns the average of the dependent variable for non-null pairs in a group, where `y` is the dependent variable and `x` is the independent variable.",
  examples = """
    Examples:
      > SELECT _FUNC_(y, x) FROM VALUES (1, 2), (2, 2), (2, 3), (2, 4) AS tab(y, x);
       1.75
      > SELECT _FUNC_(y, x) FROM VALUES (1, null) AS tab(y, x);
       NULL
      > SELECT _FUNC_(y, x) FROM VALUES (null, 1) AS tab(y, x);
       NULL
      > SELECT _FUNC_(y, x) FROM VALUES (1, 2), (2, null), (2, 3), (2, 4) AS tab(y, x);
       1.6666666666666667
      > SELECT _FUNC_(y, x) FROM VALUES (1, 2), (2, null), (null, 3), (2, 4) AS tab(y, x);
       1.5
  """,
  group = "agg_funcs",
  since = "3.3.0")
// scalastyle:on line.size.limit
case class RegrAvgY(
    left: Expression,
    right: Expression)
  extends AggregateFunction
  with RuntimeReplaceableAggregate
  with ImplicitCastInputTypes
  with BinaryLike[Expression] {
  override lazy val replacement: Expression =
    Average(If(And(IsNotNull(left), IsNotNull(right)), left, Literal.create(null, left.dataType)))
  override def nodeName: String = "regr_avgy"
  override def inputTypes: Seq[AbstractDataType] = Seq(NumericType, NumericType)
  override protected def withNewChildrenInternal(
      newLeft: Expression, newRight: Expression): RegrAvgY =
    this.copy(left = newLeft, right = newRight)
}

// scalastyle:off line.size.limit
@ExpressionDescription(
  usage = "_FUNC_(y, x) - Returns the coefficient of determination for non-null pairs in a group, where `y` is the dependent variable and `x` is the independent variable.",
  examples = """
    Examples:
      > SELECT _FUNC_(y, x) FROM VALUES (1, 2), (2, 2), (2, 3), (2, 4) AS tab(y, x);
       0.2727272727272727
      > SELECT _FUNC_(y, x) FROM VALUES (1, null) AS tab(y, x);
       NULL
      > SELECT _FUNC_(y, x) FROM VALUES (null, 1) AS tab(y, x);
       NULL
      > SELECT _FUNC_(y, x) FROM VALUES (1, 2), (2, null), (2, 3), (2, 4) AS tab(y, x);
       0.7500000000000001
      > SELECT _FUNC_(y, x) FROM VALUES (1, 2), (2, null), (null, 3), (2, 4) AS tab(y, x);
       1.0
  """,
  group = "agg_funcs",
  since = "3.3.0")
// scalastyle:on line.size.limit
case class RegrR2(y: Expression, x: Expression) extends PearsonCorrelation(y, x, true) {
  override def prettyName: String = "regr_r2"
  override val evaluateExpression: Expression = {
    val corr = ck / sqrt(xMk * yMk)
    If(xMk === 0.0, Literal.create(null, DoubleType),
      If(yMk === 0.0, Literal.create(1.0, DoubleType), corr * corr))
  }
  override protected def withNewChildrenInternal(
      newLeft: Expression, newRight: Expression): RegrR2 =
    this.copy(y = newLeft, x = newRight)
}

// scalastyle:off line.size.limit
@ExpressionDescription(
  usage = "_FUNC_(y, x) - Returns REGR_COUNT(y, x) * VAR_POP(x) for non-null pairs in a group, where `y` is the dependent variable and `x` is the independent variable.",
  examples = """
    Examples:
      > SELECT _FUNC_(y, x) FROM VALUES (1, 2), (2, 2), (2, 3), (2, 4) AS tab(y, x);
       2.75
      > SELECT _FUNC_(y, x) FROM VALUES (1, 2), (2, null), (2, 3), (2, 4) AS tab(y, x);
       2.0
      > SELECT _FUNC_(y, x) FROM VALUES (1, 2), (2, null), (null, 3), (2, 4) AS tab(y, x);
       2.0
  """,
  group = "agg_funcs",
  since = "3.4.0")
// scalastyle:on line.size.limit
case class RegrSXX(
    left: Expression,
    right: Expression)
  extends AggregateFunction
    with RuntimeReplaceableAggregate
    with ImplicitCastInputTypes
    with BinaryLike[Expression] {
  override lazy val replacement: Expression =
    RegrReplacement(If(Or(IsNull(left), IsNull(right)), Literal.create(null, DoubleType), right))
  override def nodeName: String = "regr_sxx"
  override def inputTypes: Seq[DoubleType] = Seq(DoubleType, DoubleType)
  override protected def withNewChildrenInternal(
      newLeft: Expression, newRight: Expression): RegrSXX =
    this.copy(left = newLeft, right = newRight)
}

// scalastyle:off line.size.limit
@ExpressionDescription(
  usage = "_FUNC_(y, x) - Returns REGR_COUNT(y, x) * COVAR_POP(y, x) for non-null pairs in a group, where `y` is the dependent variable and `x` is the independent variable.",
  examples = """
    Examples:
      > SELECT _FUNC_(y, x) FROM VALUES (1, 2), (2, 2), (2, 3), (2, 4) AS tab(y, x);
       0.75
      > SELECT _FUNC_(y, x) FROM VALUES (1, 2), (2, null), (2, 3), (2, 4) AS tab(y, x);
       1.0
      > SELECT _FUNC_(y, x) FROM VALUES (1, 2), (2, null), (null, 3), (2, 4) AS tab(y, x);
       1.0
  """,
  group = "agg_funcs",
  since = "3.4.0")
// scalastyle:on line.size.limit
case class RegrSXY(y: Expression, x: Expression) extends Covariance(y, x, true) {
  override def prettyName: String = "regr_sxy"
  override val evaluateExpression: Expression = {
    If(n === 0.0, Literal.create(null, DoubleType), ck)
  }
  override protected def withNewChildrenInternal(
      newLeft: Expression, newRight: Expression): RegrSXY =
    this.copy(y = newLeft, x = newRight)
}

// scalastyle:off line.size.limit
@ExpressionDescription(
  usage = "_FUNC_(y, x) - Returns REGR_COUNT(y, x) * VAR_POP(y) for non-null pairs in a group, where `y` is the dependent variable and `x` is the independent variable.",
  examples = """
    Examples:
      > SELECT _FUNC_(y, x) FROM VALUES (1, 2), (2, 2), (2, 3), (2, 4) AS tab(y, x);
       0.75
      > SELECT _FUNC_(y, x) FROM VALUES (1, 2), (2, null), (2, 3), (2, 4) AS tab(y, x);
       0.6666666666666666
      > SELECT _FUNC_(y, x) FROM VALUES (1, 2), (2, null), (null, 3), (2, 4) AS tab(y, x);
       0.5
  """,
  group = "agg_funcs",
  since = "3.4.0")
// scalastyle:on line.size.limit
case class RegrSYY(
    left: Expression,
    right: Expression)
  extends AggregateFunction
    with RuntimeReplaceableAggregate
    with ImplicitCastInputTypes
    with BinaryLike[Expression] {
  override lazy val replacement: Expression =
    RegrReplacement(If(Or(IsNull(left), IsNull(right)), Literal.create(null, DoubleType), left))
  override def nodeName: String = "regr_syy"
  override def inputTypes: Seq[DoubleType] = Seq(DoubleType, DoubleType)
  override protected def withNewChildrenInternal(
      newLeft: Expression, newRight: Expression): RegrSYY =
    this.copy(left = newLeft, right = newRight)
}

// scalastyle:off line.size.limit
@ExpressionDescription(
  usage = "_FUNC_(y, x) - Returns the slope of the linear regression line for non-null pairs in a group, where `y` is the dependent variable and `x` is the independent variable.",
  examples = """
    Examples:
      > SELECT _FUNC_(y, x) FROM VALUES (1,1), (2,2), (3,3) AS tab(y, x);
       1.0
      > SELECT _FUNC_(y, x) FROM VALUES (1, null) AS tab(y, x);
       NULL
      > SELECT _FUNC_(y, x) FROM VALUES (null, 1) AS tab(y, x);
       NULL
  """,
  group = "agg_funcs",
  since = "3.4.0")
// scalastyle:on line.size.limit
case class RegrSlope(left: Expression, right: Expression) extends DeclarativeAggregate
  with ImplicitCastInputTypes with BinaryLike[Expression] {

  private val covarPop = new CovPopulation(right, left)

  private val varPop = new VariancePop(right)

  override def nullable: Boolean = true

  override def dataType: DataType = DoubleType

  override def inputTypes: Seq[DoubleType] = Seq(DoubleType, DoubleType)

  override lazy val aggBufferAttributes: Seq[AttributeReference] =
    covarPop.aggBufferAttributes ++ varPop.aggBufferAttributes

  override lazy val initialValues: Seq[Expression] = covarPop.initialValues ++ varPop.initialValues

  override lazy val updateExpressions: Seq[Expression] =
    covarPop.updateExpressions ++ varPop.updateExpressions

  override lazy val mergeExpressions: Seq[Expression] =
    covarPop.mergeExpressions ++ varPop.mergeExpressions

  override lazy val evaluateExpression: Expression = {
    If(varPop.m2 === 0.0, Literal.create(null, DoubleType), covarPop.ck / varPop.m2)
  }

  override lazy val inputAggBufferAttributes: Seq[AttributeReference] =
    covarPop.inputAggBufferAttributes ++ varPop.inputAggBufferAttributes

  override def prettyName: String = "regr_slope"

  override protected def withNewChildrenInternal(
      newLeft: Expression, newRight: Expression): RegrSlope =
    copy(left = newLeft, right = newRight)
}

// scalastyle:off line.size.limit
@ExpressionDescription(
  usage = "_FUNC_(y, x) - Returns the intercept of the univariate linear regression line for non-null pairs in a group, where `y` is the dependent variable and `x` is the independent variable.",
  examples = """
    Examples:
      > SELECT _FUNC_(y, x) FROM VALUES (1,1), (2,2), (3,3) AS tab(y, x);
       0.0
      > SELECT _FUNC_(y, x) FROM VALUES (1, null) AS tab(y, x);
       NULL
      > SELECT _FUNC_(y, x) FROM VALUES (null, 1) AS tab(y, x);
       NULL
  """,
  group = "agg_funcs",
  since = "3.4.0")
// scalastyle:on line.size.limit
case class RegrIntercept(left: Expression, right: Expression) extends DeclarativeAggregate
  with ImplicitCastInputTypes with BinaryLike[Expression] {

  private val covarPop = new CovPopulation(right, left)

  private val varPop = new VariancePop(right)

  override def nullable: Boolean = true

  override def dataType: DataType = DoubleType

  override def inputTypes: Seq[DoubleType] = Seq(DoubleType, DoubleType)

  override lazy val aggBufferAttributes: Seq[AttributeReference] =
    covarPop.aggBufferAttributes ++ varPop.aggBufferAttributes

  override lazy val initialValues: Seq[Expression] = covarPop.initialValues ++ varPop.initialValues

  override lazy val updateExpressions: Seq[Expression] =
    covarPop.updateExpressions ++ varPop.updateExpressions

  override lazy val mergeExpressions: Seq[Expression] =
    covarPop.mergeExpressions ++ varPop.mergeExpressions

  override lazy val evaluateExpression: Expression = {
    If(varPop.m2 === 0.0, Literal.create(null, DoubleType),
      covarPop.yAvg - covarPop.ck / varPop.m2 * covarPop.xAvg)
  }

  override lazy val inputAggBufferAttributes: Seq[AttributeReference] =
    covarPop.inputAggBufferAttributes ++ varPop.inputAggBufferAttributes

  override def prettyName: String = "regr_intercept"

  override protected def withNewChildrenInternal(
      newLeft: Expression, newRight: Expression): RegrIntercept =
    copy(left = newLeft, right = newRight)
}

相关信息

spark 源码目录

相关文章

spark AnyValue 源码

spark ApproxCountDistinctForIntervals 源码

spark ApproximatePercentile 源码

spark Average 源码

spark BloomFilterAggregate 源码

spark CentralMomentAgg 源码

spark Corr 源码

spark Count 源码

spark CountIf 源码

spark CountMinSketchAgg 源码