[SPARK-32310] Add *args to different Params constructors (#515)

* Add *args to different Params constructors

Resolves #441

third_party/3/pyspark/ml/classification.pyi

@@ -254,6 +254,7 @@ class _LogisticRegressionParams(
  upperBoundsOnCoefficients: Param[Matrix]
  lowerBoundsOnIntercepts: Param[Vector]
  upperBoundsOnIntercepts: Param[Vector]
+ def __init__(self, *args: Any): ...
  def setThreshold(self: P, value: float) -> P: ...
  def getThreshold(self) -> float: ...
  def setThresholds(self: P, value: List[float]) -> P: ...
@@ -371,7 +372,9 @@ class BinaryLogisticRegressionSummary(
 class BinaryLogisticRegressionTrainingSummary(
  BinaryLogisticRegressionSummary, LogisticRegressionTrainingSummary
 ): ...
-class _DecisionTreeClassifierParams(_DecisionTreeParams, _TreeClassifierParams): ...
+
+class _DecisionTreeClassifierParams(_DecisionTreeParams, _TreeClassifierParams):
+ def __init__(self, *args: Any): ...
 
 class DecisionTreeClassifier(
  _JavaProbabilisticClassifier[DecisionTreeClassificationModel],
@@ -443,7 +446,8 @@ class DecisionTreeClassificationModel(
  @property
  def featureImportances(self) -> Vector: ...
 
-class _RandomForestClassifierParams(_RandomForestParams, _TreeClassifierParams): ...
+class _RandomForestClassifierParams(_RandomForestParams, _TreeClassifierParams):
+ def __init__(self, *args: Any): ...
 
 class RandomForestClassifier(
  _JavaProbabilisticClassifier[RandomForestClassificationModel],
@@ -544,6 +548,7 @@ class BinaryRandomForestClassificationTrainingSummary(
 class _GBTClassifierParams(_GBTParams, _HasVarianceImpurity):
  supportedLossTypes: List[str]
  lossType: Param[str]
+ def __init__(self, *args: Any): ...
  def getLossType(self) -> str: ...
 
 class GBTClassifier(
@@ -636,6 +641,7 @@ class GBTClassificationModel(
 class _NaiveBayesParams(_PredictorParams, HasWeightCol):
  smoothing: Param[float]
  modelType: Param[str]
+ def __init__(self, *args: Any): ...
  def getSmoothing(self) -> float: ...
  def getModelType(self) -> str: ...
 
@@ -702,6 +708,7 @@ class _MultilayerPerceptronParams(
  layers: Param[List[int]]
  solver: Param[str]
  initialWeights: Param[Vector]
+ def __init__(self, *args: Any): ...
  def getLayers(self) -> List[int]: ...
  def getInitialWeights(self) -> Vector: ...
 

third_party/3/pyspark/ml/clustering.pyi

@@ -58,6 +58,7 @@ class _GaussianMixtureParams(
  HasBlockSize,
 ):
  k: Param[int]
+ def __init__(self, *args: Any): ...
  def getK(self) -> int: ...
 
 class GaussianMixtureModel(
@@ -149,6 +150,7 @@ class _KMeansParams(
  k: Param[int]
  initMode: Param[str]
  initSteps: Param[int]
+ def __init__(self, *args: Any): ...
  def getK(self) -> int: ...
  def getInitMode(self) -> str: ...
  def getInitSteps(self) -> int: ...
@@ -219,6 +221,7 @@ class _BisectingKMeansParams(
 ):
  k: Param[int]
  minDivisibleClusterSize: Param[float]
+ def __init__(self, *args: Any): ...
  def getK(self) -> int: ...
  def getMinDivisibleClusterSize(self) -> float: ...
 
@@ -291,6 +294,7 @@ class _LDAParams(HasMaxIter, HasFeaturesCol, HasSeed, HasCheckpointInterval):
  topicConcentration: Param[float]
  topicDistributionCol: Param[str]
  keepLastCheckpoint: Param[bool]
+ def __init__(self, *args: Any): ...
  def setK(self, value: int) -> LDA: ...
  def getOptimizer(self) -> str: ...
  def getLearningOffset(self) -> float: ...
@@ -381,6 +385,7 @@ class _PowerIterationClusteringParams(HasMaxIter, HasWeightCol):
  initMode: Param[str]
  srcCol: Param[str]
  dstCol: Param[str]
+ def __init__(self, *args: Any): ...
  def getK(self) -> int: ...
  def getInitMode(self) -> str: ...
  def getSrcCol(self) -> str: ...

third_party/3/pyspark/ml/feature.pyi

@@ -80,6 +80,7 @@ class Binarizer(
 
 class _LSHParams(HasInputCol, HasOutputCol):
  numHashTables: Param[int]
+ def __init__(self, *args: Any): ...
  def getNumHashTables(self) -> int: ...
 
 class _LSH(Generic[JM], JavaEstimator[JM], _LSHParams, JavaMLReadable, JavaMLWritable):
@@ -386,6 +387,7 @@ class HashingTF(
 
 class _IDFParams(HasInputCol, HasOutputCol):
  minDocFreq: Param[int]
+ def __init__(self, *args: Any): ...
  def getMinDocFreq(self) -> int: ...
 
 class IDF(JavaEstimator[IDFModel], _IDFParams, JavaMLReadable[IDF], JavaMLWritable):
@@ -558,6 +560,7 @@ class MinHashLSHModel(_LSHModel, JavaMLReadable[MinHashLSHModel], JavaMLWritable
 class _MinMaxScalerParams(HasInputCol, HasOutputCol):
  min: Param[float]
  max: Param[float]
+ def __init__(self, *args: Any): ...
  def getMin(self) -> float: ...
  def getMax(self) -> float: ...
 
@@ -653,6 +656,7 @@ class Normalizer(
 class _OneHotEncoderParams(HasInputCols, HasOutputCols, HasHandleInvalid):
  handleInvalid: Param[str]
  dropLast: Param[bool]
+ def __init__(self, *args: Any): ...
  def getDropLast(self) -> bool: ...
 
 class OneHotEncoder(
@@ -813,6 +817,7 @@ class _RobustScalerParams(HasInputCol, HasOutputCol, HasRelativeError):
  upper: Param[float]
  withCentering: Param[bool]
  withScaling: Param[bool]
+ def __init__(self, *args: Any): ...
  def getLower(self) -> float: ...
  def getUpper(self) -> float: ...
  def getWithCentering(self) -> bool: ...
@@ -913,6 +918,7 @@ class SQLTransformer(JavaTransformer, JavaMLReadable[SQLTransformer], JavaMLWrit
 class _StandardScalerParams(HasInputCol, HasOutputCol):
  withMean: Param[bool]
  withStd: Param[bool]
+ def __init__(self, *args: Any): ...
  def getWithMean(self) -> bool: ...
  def getWithStd(self) -> bool: ...
 
@@ -1178,6 +1184,7 @@ class VectorAssembler(
 class _VectorIndexerParams(HasInputCol, HasOutputCol, HasHandleInvalid):
  maxCategories: Param[int]
  handleInvalid: Param[str]
+ def __init__(self, *args: Any): ...
  def getMaxCategories(self) -> int: ...
 
 class VectorIndexer(
@@ -1256,6 +1263,7 @@ class _Word2VecParams(HasStepSize, HasMaxIter, HasSeed, HasInputCol, HasOutputCo
  minCount: Param[int]
  windowSize: Param[int]
  maxSentenceLength: Param[int]
+ def __init__(self, *args: Any): ...
  def getVectorSize(self) -> int: ...
  def getNumPartitions(self) -> int: ...
  def getMinCount(self) -> int: ...
@@ -1358,6 +1366,7 @@ class _RFormulaParams(HasFeaturesCol, HasLabelCol, HasHandleInvalid):
  forceIndexLabel: Param[bool]
  stringIndexerOrderType: Param[str]
  handleInvalid: Param[str]
+ def __init__(self, *args: Any): ...
  def getFormula(self) -> str: ...
  def getForceIndexLabel(self) -> bool: ...
  def getStringIndexerOrderType(self) -> str: ...
@@ -1406,6 +1415,7 @@ class _SelectorParams(HasFeaturesCol, HasOutputCol, HasLabelCol):
  fpr: Param[float]
  fdr: Param[float]
  fwe: Param[float]
+ def __init__(self, *args: Any): ...
  def getSelectorType(self) -> str: ...
  def getNumTopFeatures(self) -> int: ...
  def getPercentile(self) -> float: ...

third_party/3/pyspark/ml/fpm.pyi

@@ -19,7 +19,7 @@
 # Stubs for pyspark.ml.base (Python 3)
 #
 
-from typing import Optional
+from typing import Any, Optional
 
 from pyspark.ml._typing import P
 from pyspark.ml.util import *
@@ -32,6 +32,7 @@ class _FPGrowthParams(HasPredictionCol):
  minSupport: Param[float]
  numPartitions: Param[int]
  minConfidence: Param[float]
+ def __init__(self, *args: Any): ...
  def getItemsCol(self) -> str: ...
  def getMinSupport(self) -> float: ...
  def getNumPartitions(self) -> int: ...

third_party/3/pyspark/ml/recommendation.pyi

@@ -46,6 +46,7 @@ class _ALSParams(
  nonnegative: Param[bool]
  intermediateStorageLevel: Param[str]
  finalStorageLevel: Param[str]
+ def __init__(self, *args: Any): ...
  def getRank(self) -> int: ...
  def getNumUserBlocks(self) -> int: ...
  def getNumItemBlocks(self) -> int: ...

third_party/3/pyspark/ml/regression.pyi

@@ -76,6 +76,7 @@ class _LinearRegressionParams(
  solver: Param[str]
  loss: Param[str]
  epsilon: Param[float]
+ def __init__(self, *args: Any): ...
  def getEpsilon(self) -> float: ...
 
 class LinearRegression(
@@ -251,7 +252,8 @@ class IsotonicRegressionModel(
 
 class _DecisionTreeRegressorParams(
  _DecisionTreeParams, _TreeRegressorParams, HasVarianceCol
-): ...
+):
+ def __init__(self, *args: Any): ...
 
 class DecisionTreeRegressor(
  _JavaRegressor[DecisionTreeRegressionModel],
@@ -323,7 +325,8 @@ class DecisionTreeRegressionModel(
  @property
  def featureImportances(self) -> Vector: ...
 
-class _RandomForestRegressorParams(_RandomForestParams, _TreeRegressorParams): ...
+class _RandomForestRegressorParams(_RandomForestParams, _TreeRegressorParams):
+ def __init__(self, *args: Any): ...
 
 class RandomForestRegressor(
  _JavaRegressor[RandomForestRegressionModel],
@@ -406,6 +409,7 @@ class RandomForestRegressionModel(
 class _GBTRegressorParams(_GBTParams, _TreeRegressorParams):
  supportedLossTypes: List[str]
  lossType: Param[str]
+ def __init__(self, *args: Any): ...
  def getLossType(self) -> str: ...
 
 class GBTRegressor(
@@ -508,6 +512,7 @@ class _AFTSurvivalRegressionParams(
  censorCol: Param[str]
  quantileProbabilities: Param[List[float]]
  quantilesCol: Param[str]
+ def __init__(self, *args: Any): ...
  def getCensorCol(self) -> str: ...
  def getQuantileProbabilities(self) -> List[float]: ...
  def getQuantilesCol(self) -> str: ...
@@ -593,6 +598,7 @@ class _GeneralizedLinearRegressionParams(
  linkPower: Param[float]
  solver: Param[str]
  offsetCol: Param[str]
+ def __init__(self, *args: Any): ...
  def getFamily(self) -> str: ...
  def getLinkPredictionCol(self) -> str: ...
  def getLink(self) -> str: ...
@@ -722,12 +728,14 @@ class _FactorizationMachinesParams(
  HasSeed,
  HasFitIntercept,
  HasRegParam,
+ HasWeightCol,
 ):
  factorSize: Param[int]
  fitLinear: Param[bool]
  miniBatchFraction: Param[float]
  initStd: Param[float]
  solver: Param[str]
+ def __init__(self, *args: Any): ...
  def getFactorSize(self): ...
  def getFitLinear(self): ...
  def getMiniBatchFraction(self): ...

third_party/3/pyspark/ml/tuning.pyi

@@ -48,6 +48,7 @@ class _ValidatorParams(HasSeed):
 class _CrossValidatorParams(_ValidatorParams):
  numFolds: Param[int]
  foldCol: Param[str]
+ def __init__(self, *args: Any): ...
  def getNumFolds(self) -> int: ...
  def getFoldCol(self) -> str: ...
 
@@ -115,6 +116,7 @@ class CrossValidatorModel(
 
 class _TrainValidationSplitParams(_ValidatorParams):
  trainRatio: Param[float]
+ def __init__(self, *args: Any): ...
  def getTrainRatio(self) -> float: ...
 
 class TrainValidationSplit(