utils

Type Members

case class ConfigEncoding(keys: List[String]) extends Encoder[Map[String, Double], DenseVector[Double]] with Product with Serializable

An encoding which converts a hyper-parameter configuration from a Map to a breeze DenseVector and back.
An encoding which converts a hyper-parameter configuration from a Map to a breeze DenseVector and back.
keys
A list of hyper-parameter strings.
case class GaussianScaler(mean: DenseVector[Double], sigma: DenseVector[Double]) extends ReversibleScaler[DenseVector[Double]] with Product with Serializable

Scales attributes of a vector pattern using the sample mean and variance of each dimension.
Scales attributes of a vector pattern using the sample mean and variance of each dimension. This assumes that there is no covariance between the data dimensions.
mean
Sample mean of the data
sigma
Sample variance of each data dimension
abstract class HasHyperParameters extends AnyRef
case class MVGaussianScaler(mean: DenseVector[Double], sigma: DenseMatrix[Double]) extends ReversibleScaler[DenseVector[Double]] with Product with Serializable

Scales the attributes of a data pattern using the sample mean and covariance matrix calculated on the data set.
Scales the attributes of a data pattern using the sample mean and covariance matrix calculated on the data set. This allows standardization of multivariate data sets where the covariance of individual data dimensions is not negligible.
mean
Sample mean of data
sigma
Sample covariance matrix of data.
case class MeanScaler(center: DenseVector[Double]) extends ReversibleScaler[DenseVector[Double]] with Product with Serializable
case class MinMaxScaler(min: DenseVector[Double], max: DenseVector[Double]) extends ReversibleScaler[DenseVector[Double]] with Product with Serializable
case class PCAScaler(center: DenseVector[Double], covmat: DenseMatrix[Double]) extends ReversibleScaler[DenseVector[Double]] with Product with Serializable

Transforms data by projecting on the principal components (eigen-vectors) of the sample covariance matrix.
Transforms data by projecting on the principal components (eigen-vectors) of the sample covariance matrix.
center
The empirical mean of the data features
covmat
The empirical covariance matrix of the data features
case class UnivariateGaussianScaler(mean: Double, sigma: Double) extends ReversibleScaler[Double] with Product with Serializable
case class UnivariateMeanScaler(center: Double) extends ReversibleScaler[Double] with Product with Serializable

Value Members

def H(x: Double): Double

Calculates the Harmonic number function for positive real arguments.
object MinMaxAccumulator extends AccumulatorParam[DenseVector[Double]]
object Renjin

Encapsulates renjin script engine and its capabilities.
def chebyshev(n: Int, x: Double, kind: Int = 1): Double

Calculates the Chebyshev polynomials of the first and second kind, in a tail recursive manner, using their recurrence relations.
def combine[A](xs: Traversable[Traversable[A]]): Seq[Seq[A]]
def diagonal[M <: PartitionedMatrix](pm: M): PartitionedMatrix

Extract the diagonal elements of a PartitionedMatrix
def diagonal(m: DenseMatrix[Double]): DenseMatrix[Double]

Extract the diagonal elements of a breeze DenseMatrix
def downloadURL(url: String, saveAs: String): Unit
def extractColumns(lines: Stream[String], sep: String, columns: List[Int], naStrings: Map[Int, String]): Stream[String]
def factorial(n: Int, accumulator: Long = 1): Long

Annotations
@tailrec()
def getCSVReader(file: String, delim: Char): CSVReader

Get a CSVReader object from a file name and delimiter character.
Get a CSVReader object from a file name and delimiter character.
file
The file pathname as a String
delim
The delimiter character used to split the csv file.
returns
A CSVReader object which can be iterated for each line.
def getMinMax(data: List[DenseVector[Double]]): (DenseVector[Double], DenseVector[Double])
def getPriorMapDistr(d: Map[String, ContinuousDistr[Double]]): ContinuousDistr[Map[String, Double]]

Convert a hyper-prior specification to a continuous distribution over Map
def getStats(data: List[DenseVector[Double]]): (DenseVector[Double], DenseVector[Double])

Get the mean and variance of a data set which is a List of DenseVector.
Get the mean and variance of a data set which is a List of DenseVector.
data
The data set.
returns
A Tuple2 containing the mean and variance * n-1.
def getStatsMult(data: List[DenseVector[Double]]): (DenseVector[Double], DenseMatrix[Double])

Get the mean and variance of a data set which is a List of DenseVector.
Get the mean and variance of a data set which is a List of DenseVector.
data
The data set.
returns
A Tuple2 containing the mean and variance.
def getTypeTag[T](obj: T)(implicit arg0: scala.reflect.api.JavaUniverse.TypeTag[T]): scala.reflect.api.JavaUniverse.TypeTag[T]
def haarMatrix(n: Int): DenseMatrix[Double]

Construct a Haar transform matrix of size n
Construct a Haar transform matrix of size n
NOTE: n must be a power of 2.
def hermite(n: Int, x: Double): Double

Calculate the value of the hermite polynomials tail recursively.
Calculate the value of the hermite polynomials tail recursively. This is needed to calculate the Gaussian derivatives at a point x.
def isSquareMatrix[V](mat: Matrix[V]): Unit
def isSymmetricMatrix[V](mat: Matrix[V]): Unit
def legendre(n: Int, x: Double): Double

Calculate the value of the Legendre polynomials tail recursively.
val log1pExp: (Double) ⇒ Double
def median(list: Stream[Double]): Double
object mvlgamma extends UFunc
def productField[Domain, Domain1](ev: Field[Domain], ev1: Field[Domain1])(implicit eqq: Eq[Domain], eqq1: Eq[Domain1]): Field[(Domain, Domain1)]
def quickselect(list: Stream[Double], k: Int): Double

Implementation of the quick-select algorithm.
def replace(find: String)(replace: String)(input: String): String
def strReplace(fileName: String)(findStringRegex: String, replaceString: String): Stream[String]
def textFileToStream(fileName: String): Stream[String]
def transformData(transform: (String) ⇒ String)(lines: Stream[String]): Stream[String]
def writeToFile(destination: String)(lines: Stream[String]): Unit

package utils

Type Members

case class ConfigEncoding(keys: List[String]) extends Encoder[Map[String, Double], DenseVector[Double]] with Product with Serializable

case class GaussianScaler(mean: DenseVector[Double], sigma: DenseVector[Double]) extends ReversibleScaler[DenseVector[Double]] with Product with Serializable

abstract class HasHyperParameters extends AnyRef

case class MVGaussianScaler(mean: DenseVector[Double], sigma: DenseMatrix[Double]) extends ReversibleScaler[DenseVector[Double]] with Product with Serializable

case class MeanScaler(center: DenseVector[Double]) extends ReversibleScaler[DenseVector[Double]] with Product with Serializable

case class MinMaxScaler(min: DenseVector[Double], max: DenseVector[Double]) extends ReversibleScaler[DenseVector[Double]] with Product with Serializable

case class PCAScaler(center: DenseVector[Double], covmat: DenseMatrix[Double]) extends ReversibleScaler[DenseVector[Double]] with Product with Serializable

case class UnivariateGaussianScaler(mean: Double, sigma: Double) extends ReversibleScaler[Double] with Product with Serializable

case class UnivariateMeanScaler(center: Double) extends ReversibleScaler[Double] with Product with Serializable

Value Members

def H(x: Double): Double

object MinMaxAccumulator extends AccumulatorParam[DenseVector[Double]]

object Renjin

def chebyshev(n: Int, x: Double, kind: Int = 1): Double

def combine[A](xs: Traversable[Traversable[A]]): Seq[Seq[A]]

def diagonal[M <: PartitionedMatrix](pm: M): PartitionedMatrix

def diagonal(m: DenseMatrix[Double]): DenseMatrix[Double]

def downloadURL(url: String, saveAs: String): Unit

def extractColumns(lines: Stream[String], sep: String, columns: List[Int], naStrings: Map[Int, String]): Stream[String]

def factorial(n: Int, accumulator: Long = 1): Long

def getCSVReader(file: String, delim: Char): CSVReader

def getMinMax(data: List[DenseVector[Double]]): (DenseVector[Double], DenseVector[Double])

def getPriorMapDistr(d: Map[String, ContinuousDistr[Double]]): ContinuousDistr[Map[String, Double]]

def getStats(data: List[DenseVector[Double]]): (DenseVector[Double], DenseVector[Double])

def getStatsMult(data: List[DenseVector[Double]]): (DenseVector[Double], DenseMatrix[Double])

def getTypeTag[T](obj: T)(implicit arg0: scala.reflect.api.JavaUniverse.TypeTag[T]): scala.reflect.api.JavaUniverse.TypeTag[T]

def haarMatrix(n: Int): DenseMatrix[Double]

def hermite(n: Int, x: Double): Double

def isSquareMatrix[V](mat: Matrix[V]): Unit

def isSymmetricMatrix[V](mat: Matrix[V]): Unit

def legendre(n: Int, x: Double): Double

val log1pExp: (Double) ⇒ Double

def median(list: Stream[Double]): Double

object mvlgamma extends UFunc

def productField[Domain, Domain1](ev: Field[Domain], ev1: Field[Domain1])(implicit eqq: Eq[Domain], eqq1: Eq[Domain1]): Field[(Domain, Domain1)]

def quickselect(list: Stream[Double], k: Int): Double

def replace(find: String)(replace: String)(input: String): String

def strReplace(fileName: String)(findStringRegex: String, replaceString: String): Stream[String]

def textFileToStream(fileName: String): Stream[String]

def transformData(transform: (String) ⇒ String)(lines: Stream[String]): Stream[String]

def writeToFile(destination: String)(lines: Stream[String]): Unit

Inherited from AnyRef

Inherited from Any

Ungrouped