MathUtils (deeplearning4j-core 0.5.0 API)

java.lang.Object
- org.deeplearning4j.util.MathUtils

```
public class MathUtils
extends Object
```
This is a math utils class.

Author:

Adam Gibson

Field Summary

Fields
Modifier and Type Field and Description

static double log2
The natural logarithm of 2.

static double SMALL
The small deviation allowed in double comparisons.

Fields
Modifier and Type	Field and Description
`static double`	`log2` The natural logarithm of 2.
`static double`	`SMALL` The small deviation allowed in double comparisons.

Constructor Summary

Constructors
Constructor and Description

MathUtils()

Constructors
Constructor and Description
`MathUtils()`

Method Summary

All Methods Static Methods Instance Methods Concrete Methods
Modifier and Type	Method and Description
`static double`	`adjustedrSquared(double rSquared, int numRegressors, int numDataPoints)` This calculates the adjusted r^2 including degrees of freedom.
`static double`	`bernoullis(double n, double k, double successProb)` This will return the bernoulli trial for the given event.
`static int`	`binomial(org.apache.commons.math3.random.RandomGenerator rng, int n, double p)` Generates a binomial distributed number using the given rng
`org.apache.commons.math3.linear.CholeskyDecomposition`	`choleskyFromMatrix(org.apache.commons.math3.linear.RealMatrix m)` This will return the cholesky decomposition of the given matrix
`static int`	`clamp(int value, int min, int max)` Clamps the value to a discrete value
`static double`	`combination(double n, double r)` This returns the combination of n choose r
`static List<double[]>`	`coordSplit(double[] vector)` This returns the coordinate split in a list of coordinates such that the values for ret[0] are the x values and ret[1] are the y values
`static List<double[]>`	`coordSplit(List<Double> vector)` This returns the coordinate split in a list of coordinates such that the values for ret[0] are the x values and ret[1] are the y values
`static double`	`correlation(double[] residuals, double[] targetAttribute)` Returns the correlation coefficient of two double vectors.
`static double`	`determinationCoefficient(double[] y1, double[] y2, int n)` This returns the determination coefficient of two vectors given a length
`static int`	`discretize(double value, double min, double max, int binCount)` Discretize the given value
`static int`	`distanceFinderZValue(double[] vector)` This will translate a vector in to an equivalent integer
`static double`	`entropy(double[] vector)` This returns the entropy (information gain, or uncertainty of a random variable).
`static double`	`errorFor(double actual, double prediction)`
`static double`	`euclideanDistance(double[] p, double[] q)` This returns the distance of two vectors sum(i=1,n) (q_i - p_i)^2
`static double`	`euclideanDistance(float[] p, float[] q)` This returns the distance of two vectors sum(i=1,n) (q_i - p_i)^2
`static double`	`factorial(double n)` This will return the factorial of the given number n.
`static double[]`	`fromString(String data, String separator)` This will take a given string and separator and convert it to an equivalent double array.
`static double[]`	`generateUniform(int l)` This will generate a series of uniformally distributed numbers between l times
`static boolean`	`gr(double a, double b)` Tests if a is greater than b.
`static double`	`hypotenuse(double a, double b)` sqrt(a^2 + b^2) without under/overflow.
`static double`	`idf(double totalDocs, double numTimesWordAppearedInADocument)` Inverse document frequency: the total docs divided by the number of times the word appeared in a document
`static double`	`information(double[] probabilities)` This returns the entropy for a given vector of probabilities.
`static int`	`kroneckerDelta(double i, double j)` This returns the kronecker delta of two doubles.
`static double`	`log2(double a)` Returns the logarithm of a for base 2.
`static double[]`	`logs2probs(double[] a)` Converts an array containing the natural logarithms of probabilities stored in a vector back into probabilities.
`static double`	`manhattanDistance(double[] p, double[] q)` This will calculate the Manhattan distance between two sets of points.
`static double`	`max(double[] doubles)`
`static int`	`maxIndex(double[] doubles)` Returns index of maximum element in a given array of doubles.
`static double`	`mean(double[] vector)` Computes the mean for an array of doubles.
`static double[]`	`mergeCoords(double[] x, double[] y)` This will merge the coordinates of the given coordinate system.
`static List<Double>`	`mergeCoords(List<Double> x, List<Double> y)` This will merge the coordinates of the given coordinate system.
`static double`	`min(double[] doubles)`
`static long`	`nextPowOf2(long v)` See: http://stackoverflow.com/questions/466204/rounding-off-to-nearest-power-of-2
`static void`	`normalize(double[] doubles, double sum)` Normalizes the doubles in the array using the given value.
`static double`	`normalize(double val, double min, double max)` Normalize a value (val - min) / (max - min)
`static double[]`	`normalizeToOne(double[] doubles)`
`static List<List<Double>>`	`partitionVariable(List<Double> arr, int chunk)` This will partition the given whole variable data applyTransformToDestination in to the specified chunk number.
`static double`	`permutation(double n, double r)` This returns the permutation of n choose r.
`static int`	`probRound(double value, Random rand)` Rounds a double to the next nearest integer value in a probabilistic fashion (e.g.
`static double`	`probToLogOdds(double prob)` Returns the log-odds for a given probability.
`static double`	`randomDoubleBetween(double begin, double end)`
`static float`	`randomFloatBetween(float begin, float end)`
`static int`	`randomNumberBetween(double begin, double end)` Generates a random integer between the specified numbers
`static int`	`randomNumberBetween(double begin, double end, org.apache.commons.math3.random.RandomGenerator rng)` Generates a random integer between the specified numbers
`static double`	`rootMeansSquaredError(double[] real, double[] predicted)` This returns the root mean squared error of two data sets
`static int`	`round(double value)` Rounds a double to the next nearest integer value.
`static double`	`roundDouble(double value, int afterDecimalPoint)` Rounds a double to the given number of decimal places.
`static float`	`roundFloat(float value, int afterDecimalPoint)` Rounds a double to the given number of decimal places.
`static double[]`	`sampleDoublesInInterval(double[][] doubles, int l)`
`static void`	`shuffleArray(int[] array, long rngSeed)`
`static void`	`shuffleArray(int[] array, Random rng)`
`static double`	`sigmoid(double x)` 1 / 1 + exp(-x)
`double`	`slope(double x1, double x2, double y1, double y2)` This returns the slope of the given points.
`static boolean`	`sm(double a, double b)` Tests if a is smaller than b.
`static double`	`squaredLoss(double[] x, double[] y, double w_0, double w_1)` This will return the squared loss of the given points
`static double`	`ssError(double[] predictedValues, double[] targetAttribute)` How much of the variance is NOT explained by the regression
`static double`	`ssReg(double[] residuals, double[] targetAttribute)` How much of the variance is explained by the regression
`static double`	`ssTotal(double[] residuals, double[] targetAttribute)` Total variance in target attribute
`static double`	`stringSimilarity(String... strings)` Calculate string similarity with tfidf weights relative to each character frequency and how many times a character appears in a given string
`static double`	`sum(double[] nums)` This returns the sum of the given array.
`static double`	`sumOfMeanDifferences(double[] vector, double[] vector2)` Used for calculating top part of simple regression for beta 1
`static double`	`sumOfMeanDifferencesOnePoint(double[] vector)` Used for calculating top part of simple regression for beta 1
`static double`	`sumOfProducts(double[]... nums)` This returns the sum of products for the given numbers.
`static double`	`sumOfSquares(double[] vector)` This returns the sum of squares for the given vector.
`static double`	`tf(int count, int documentLength)` Term frequency: 1+ log10(count)
`static double`	`tfidf(double tf, double idf)` Return td * idf
`static double`	`times(double[] nums)` This returns the product of all numbers in the given array.
`static int`	`toDecimal(String binary)` This will convert the given binary string to a decimal based integer
`static double`	`uniform(Random rng, double min, double max)` Generate a uniform random number from the given rng
`static double`	`variance(double[] vector)`
`static double`	`vectorLength(double[] vector)` Returns the vector length (sqrt(sum(x_i))
`static double`	`w_0(double[] x, double[] y, int n)`
`static double`	`w_1(double[] x, double[] y, int n)`
`static double[]`	`weightsFor(double[] vector)` This returns the minimized loss values for a given vector.
`static double[]`	`weightsFor(List<Double> vector)` This returns the minimized loss values for a given vector.
`static double[]`	`xVals(double[] vector)` This returns the x values of the given vector.
`static double[]`	`yVals(double[] vector)` This returns the odd indexed values for the given vector

Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

- Field Detail
  - log2
```
public static double log2
```
    The natural logarithm of 2.
  - SMALL
```
public static double SMALL
```
    The small deviation allowed in double comparisons.
- Constructor Detail
  - MathUtils
```
public MathUtils()
```
- Method Detail
  - normalize
```
public static double normalize(double val,
                               double min,
                               double max)
```
    Normalize a value (val - min) / (max - min)
    
    Parameters:
    
    val - value to normalize
    
    max - max value
    
    min - min value
    
    Returns:
    
    the normalized value
  - clamp
```
public static int clamp(int value,
                        int min,
                        int max)
```
    Clamps the value to a discrete value
    
    Parameters:
    
    value - the value to clamp
    
    min - min for the probability distribution
    
    max - max for the probability distribution
    
    Returns:
    
    the discrete value
  - discretize
```
public static int discretize(double value,
                             double min,
                             double max,
                             int binCount)
```
    Discretize the given value
    
    Parameters:
    
    value - the value to discretize
    
    min - the min of the distribution
    
    max - the max of the distribution
    
    binCount - the number of bins
    
    Returns:
    
    the discretized value
  - nextPowOf2
```
public static long nextPowOf2(long v)
```
    See: http://stackoverflow.com/questions/466204/rounding-off-to-nearest-power-of-2
    
    Parameters:
    
    v - the number to getFromOrigin the next power of 2 for
    
    Returns:
    
    the next power of 2 for the passed in value
  - binomial
```
public static int binomial(org.apache.commons.math3.random.RandomGenerator rng,
                           int n,
                           double p)
```
    Generates a binomial distributed number using the given rng
    
    Parameters:
    
    rng -
    
    n -
    
    p -
    
    Returns:
  - uniform
```
public static double uniform(Random rng,
                             double min,
                             double max)
```
    Generate a uniform random number from the given rng
    
    Parameters:
    
    rng - the rng to use
    
    min - the min num
    
    max - the max num
    
    Returns:
    
    a number uniformly distributed between min and max
  - correlation
```
public static double correlation(double[] residuals,
                                 double[] targetAttribute)
```
    Returns the correlation coefficient of two double vectors.
    
    Parameters:
    
    residuals - residuals
    
    targetAttribute - target attribute vector
    
    Returns:
    
    the correlation coefficient or r
  - sigmoid
```
public static double sigmoid(double x)
```
    1 / 1 + exp(-x)
    
    Parameters:
    
    x -
    
    Returns:
  - ssReg
```
public static double ssReg(double[] residuals,
                           double[] targetAttribute)
```
    How much of the variance is explained by the regression
    
    Parameters:
    
    residuals - error
    
    targetAttribute - data for target attribute
    
    Returns:
    
    the sum squares of regression
  - ssError
```
public static double ssError(double[] predictedValues,
                             double[] targetAttribute)
```
    How much of the variance is NOT explained by the regression
    
    Parameters:
    
    predictedValues - predicted values
    
    targetAttribute - data for target attribute
    
    Returns:
    
    the sum squares of regression
  - stringSimilarity
```
public static double stringSimilarity(String... strings)
```
    Calculate string similarity with tfidf weights relative to each character frequency and how many times a character appears in a given string
    
    Parameters:
    
    strings - the strings to calculate similarity for
    
    Returns:
    
    the cosine similarity between the strings
  - vectorLength
```
public static double vectorLength(double[] vector)
```
    Returns the vector length (sqrt(sum(x_i))
    
    Parameters:
    
    vector - the vector to return the vector length for
    
    Returns:
    
    the vector length of the passed in array
  - idf
```
public static double idf(double totalDocs,
                         double numTimesWordAppearedInADocument)
```
    Inverse document frequency: the total docs divided by the number of times the word appeared in a document
    
    Parameters:
    
    totalDocs - the total documents for the data applyTransformToDestination
    
    numTimesWordAppearedInADocument - the number of times the word occurred in a document
    
    Returns:
    
    log(10) (totalDocs/numTImesWordAppearedInADocument)
  - tf
```
public static double tf(int count,
                        int documentLength)
```
    Term frequency: 1+ log10(count)
    
    Parameters:
    
    count - the count of a word or character in a given string or document
    
    Returns:
    
    1+ log(10) count
  - tfidf
```
public static double tfidf(double tf,
                           double idf)
```
    Return td * idf
    
    Parameters:
    
    tf - the term frequency (assumed calculated)
    
    idf - inverse document frequency (assumed calculated)
    
    Returns:
    
    td * idf
  - ssTotal
```
public static double ssTotal(double[] residuals,
                             double[] targetAttribute)
```
    Total variance in target attribute
    
    Parameters:
    
    residuals - error
    
    targetAttribute - data for target attribute
    
    Returns:
    
    Total variance in target attribute
  - sum
```
public static double sum(double[] nums)
```
    This returns the sum of the given array.
    
    Parameters:
    
    nums - the array of numbers to sum
    
    Returns:
    
    the sum of the given array
  - mergeCoords
```
public static double[] mergeCoords(double[] x,
                                   double[] y)
```
    This will merge the coordinates of the given coordinate system.
    
    Parameters:
    
    x - the x coordinates
    
    y - the y coordinates
    
    Returns:
    
    a vector such that each (x,y) pair is at ret[i],ret[i+1]
  - mergeCoords
```
public static List<Double> mergeCoords(List<Double> x,
                                       List<Double> y)
```
    This will merge the coordinates of the given coordinate system.
    
    Parameters:
    
    x - the x coordinates
    
    y - the y coordinates
    
    Returns:
    
    a vector such that each (x,y) pair is at ret[i],ret[i+1]
  - weightsFor
```
public static double[] weightsFor(List<Double> vector)
```
    This returns the minimized loss values for a given vector. It is assumed that the x, y pairs are at vector[i], vector[i+1]
    
    Parameters:
    
    vector - the vector of numbers to getFromOrigin the weights for
    
    Returns:
    
    a double array with w_0 and w_1 are the associated indices.
  - squaredLoss
```
public static double squaredLoss(double[] x,
                                 double[] y,
                                 double w_0,
                                 double w_1)
```
    This will return the squared loss of the given points
    
    Parameters:
    
    x - the x coordinates to use
    
    y - the y coordinates to use
    
    w_0 - the first weight
    
    w_1 - the second weight
    
    Returns:
    
    the squared loss of the given points
  - w_1
```
public static double w_1(double[] x,
                         double[] y,
                         int n)
```
  - w_0
```
public static double w_0(double[] x,
                         double[] y,
                         int n)
```
  - weightsFor
```
public static double[] weightsFor(double[] vector)
```
    This returns the minimized loss values for a given vector. It is assumed that the x, y pairs are at vector[i], vector[i+1]
    
    Parameters:
    
    vector - the vector of numbers to getFromOrigin the weights for
    
    Returns:
    
    a double array with w_0 and w_1 are the associated indices.
  - errorFor
```
public static double errorFor(double actual,
                              double prediction)
```
  - sumOfMeanDifferences
```
public static double sumOfMeanDifferences(double[] vector,
                                          double[] vector2)
```
    Used for calculating top part of simple regression for beta 1
    
    Parameters:
    
    vector - the x coordinates
    
    vector2 - the y coordinates
    
    Returns:
    
    the sum of mean differences for the input vectors
  - sumOfMeanDifferencesOnePoint
```
public static double sumOfMeanDifferencesOnePoint(double[] vector)
```
    Used for calculating top part of simple regression for beta 1
    
    Parameters:
    
    vector - the x coordinates
    
    Returns:
    
    the sum of mean differences for the input vectors
  - variance
```
public static double variance(double[] vector)
```
  - times
```
public static double times(double[] nums)
```
    This returns the product of all numbers in the given array.
    
    Parameters:
    
    nums - the numbers to multiply over
    
    Returns:
    
    the product of all numbers in the array, or 0 if the length is or nums i null
  - sumOfProducts
```
public static double sumOfProducts(double[]... nums)
```
    This returns the sum of products for the given numbers.
    
    Parameters:
    
    nums - the sum of products for the give numbers
    
    Returns:
    
    the sum of products for the given numbers
  - coordSplit
```
public static List<double[]> coordSplit(double[] vector)
```
    This returns the coordinate split in a list of coordinates such that the values for ret[0] are the x values and ret[1] are the y values
    
    Parameters:
    
    vector - the vector to split with x and y values/
    
    Returns:
    
    a coordinate split for the given vector of values. if null, is passed in null is returned
  - partitionVariable
```
public static List<List<Double>> partitionVariable(List<Double> arr,
                                                   int chunk)
```
    This will partition the given whole variable data applyTransformToDestination in to the specified chunk number.
    
    Parameters:
    
    arr - the data applyTransformToDestination to pass in
    
    chunk - the number to separate by
    
    Returns:
    
    a partition data applyTransformToDestination relative to the passed in chunk number
  - coordSplit
```
public static List<double[]> coordSplit(List<Double> vector)
```
    This returns the coordinate split in a list of coordinates such that the values for ret[0] are the x values and ret[1] are the y values
    
    Parameters:
    
    vector - the vector to split with x and y values Note that the list will be more stable due to the size operator. The array version will have extraneous values if not monitored properly.
    
    Returns:
    
    a coordinate split for the given vector of values. if null, is passed in null is returned
  - xVals
```
public static double[] xVals(double[] vector)
```
    This returns the x values of the given vector. These are assumed to be the even values of the vector.
    
    Parameters:
    
    vector - the vector to getFromOrigin the values for
    
    Returns:
    
    the x values of the given vector
  - yVals
```
public static double[] yVals(double[] vector)
```
    This returns the odd indexed values for the given vector
    
    Parameters:
    
    vector - the odd indexed values of rht egiven vector
    
    Returns:
    
    the y values of the given vector
  - sumOfSquares
```
public static double sumOfSquares(double[] vector)
```
    This returns the sum of squares for the given vector.
    
    Parameters:
    
    vector - the vector to obtain the sum of squares for
    
    Returns:
    
    the sum of squares for this vector
  - determinationCoefficient
```
public static double determinationCoefficient(double[] y1,
                                              double[] y2,
                                              int n)
```
    This returns the determination coefficient of two vectors given a length
    
    Parameters:
    
    y1 - the first vector
    
    y2 - the second vector
    
    n - the length of both vectors
    
    Returns:
    
    the determination coefficient or r^2
  - log2
```
public static double log2(double a)
```
    Returns the logarithm of a for base 2.
    
    Parameters:
    
    a - a double
    
    Returns:
    
    the logarithm for base 2
  - slope
```
public double slope(double x1,
                    double x2,
                    double y1,
                    double y2)
```
    This returns the slope of the given points.
    
    Parameters:
    
    x1 - the first x to use
    
    x2 - the end x to use
    
    y1 - the begin y to use
    
    y2 - the end y to use
    
    Returns:
    
    the slope of the given points
  - rootMeansSquaredError
```
public static double rootMeansSquaredError(double[] real,
                                           double[] predicted)
```
    This returns the root mean squared error of two data sets
    
    Parameters:
    
    real - the real values
    
    predicted - the predicted values
    
    Returns:
    
    the root means squared error for two data sets
  - entropy
```
public static double entropy(double[] vector)
```
    This returns the entropy (information gain, or uncertainty of a random variable).
    
    Parameters:
    
    vector - the vector of values to getFromOrigin the entropy for
    
    Returns:
    
    the entropy of the given vector
  - kroneckerDelta
```
public static int kroneckerDelta(double i,
                                 double j)
```
    This returns the kronecker delta of two doubles.
    
    Parameters:
    
    i - the first number to compare
    
    j - the second number to compare
    
    Returns:
    
    1 if they are equal, 0 otherwise
  - adjustedrSquared
```
public static double adjustedrSquared(double rSquared,
                                      int numRegressors,
                                      int numDataPoints)
```
    This calculates the adjusted r^2 including degrees of freedom. Also known as calculating "strength" of a regression
    
    Parameters:
    
    rSquared - the r squared value to calculate
    
    numRegressors - number of variables
    
    numDataPoints - size of the data applyTransformToDestination
    
    Returns:
    
    an adjusted r^2 for degrees of freedom
  - normalizeToOne
```
public static double[] normalizeToOne(double[] doubles)
```
  - min
```
public static double min(double[] doubles)
```
  - max
```
public static double max(double[] doubles)
```
  - normalize
```
public static void normalize(double[] doubles,
                             double sum)
```
    Normalizes the doubles in the array using the given value.
    
    Parameters:
    
    doubles - the array of double
    
    sum - the value by which the doubles are to be normalized
    
    Throws:
    
    IllegalArgumentException - if sum is zero or NaN
  - logs2probs
```
public static double[] logs2probs(double[] a)
```
    Converts an array containing the natural logarithms of probabilities stored in a vector back into probabilities. The probabilities are assumed to sum to one.
    
    Parameters:
    
    a - an array holding the natural logarithms of the probabilities
    
    Returns:
    
    the converted array
  - information
```
public static double information(double[] probabilities)
```
    This returns the entropy for a given vector of probabilities.
    
    Parameters:
    
    probabilities - the probabilities to getFromOrigin the entropy for
    
    Returns:
    
    the entropy of the given probabilities.
  - maxIndex
```
public static int maxIndex(double[] doubles)
```
    Returns index of maximum element in a given array of doubles. First maximum is returned.
    
    Parameters:
    
    doubles - the array of doubles
    
    Returns:
    
    the index of the maximum element
  - factorial
```
public static double factorial(double n)
```
    This will return the factorial of the given number n.
    
    Parameters:
    
    n - the number to getFromOrigin the factorial for
    
    Returns:
    
    the factorial for this number
  - probToLogOdds
```
public static double probToLogOdds(double prob)
```
    Returns the log-odds for a given probability.
    
    Parameters:
    
    prob - the probability
    
    Returns:
    
    the log-odds after the probability has been mapped to [Utils.SMALL, 1-Utils.SMALL]
  - round
```
public static int round(double value)
```
    Rounds a double to the next nearest integer value. The JDK version of it doesn't work properly.
    
    Parameters:
    
    value - the double value
    
    Returns:
    
    the resulting integer value
  - permutation
```
public static double permutation(double n,
                                 double r)
```
    This returns the permutation of n choose r.
    
    Parameters:
    
    n - the n to choose
    
    r - the number of elements to choose
    
    Returns:
    
    the permutation of these numbers
  - combination
```
public static double combination(double n,
                                 double r)
```
    This returns the combination of n choose r
    
    Parameters:
    
    n - the number of elements overall
    
    r - the number of elements to choose
    
    Returns:
    
    the amount of possible combinations for this applyTransformToDestination of elements
  - hypotenuse
```
public static double hypotenuse(double a,
                                double b)
```
    sqrt(a^2 + b^2) without under/overflow.
  - probRound
```
public static int probRound(double value,
                            Random rand)
```
    Rounds a double to the next nearest integer value in a probabilistic fashion (e.g. 0.8 has a 20% chance of being rounded down to 0 and a 80% chance of being rounded up to 1). In the limit, the average of the rounded numbers generated by this procedure should converge to the original double.
    
    Parameters:
    
    value - the double value
    
    rand - the random number generator
    
    Returns:
    
    the resulting integer value
  - roundDouble
```
public static double roundDouble(double value,
                                 int afterDecimalPoint)
```
    Rounds a double to the given number of decimal places.
    
    Parameters:
    
    value - the double value
    
    afterDecimalPoint - the number of digits after the decimal point
    
    Returns:
    
    the double rounded to the given precision
  - roundFloat
```
public static float roundFloat(float value,
                               int afterDecimalPoint)
```
    Rounds a double to the given number of decimal places.
    
    Parameters:
    
    value - the double value
    
    afterDecimalPoint - the number of digits after the decimal point
    
    Returns:
    
    the double rounded to the given precision
  - bernoullis
```
public static double bernoullis(double n,
                                double k,
                                double successProb)
```
    This will return the bernoulli trial for the given event. A bernoulli trial is a mechanism for detecting the probability of a given event occurring k times in n independent trials
    
    Parameters:
    
    n - the number of trials
    
    k - the number of times the target event occurs
    
    successProb - the probability of the event happening
    
    Returns:
    
    the probability of the given event occurring k times.
  - sm
```
public static boolean sm(double a,
                         double b)
```
    Tests if a is smaller than b.
    
    Parameters:
    
    a - a double
    
    b - a double
  - gr
```
public static boolean gr(double a,
                         double b)
```
    Tests if a is greater than b.
    
    Parameters:
    
    a - a double
    
    b - a double
  - fromString
```
public static double[] fromString(String data,
                                  String separator)
```
    This will take a given string and separator and convert it to an equivalent double array.
    
    Parameters:
    
    data - the data to separate
    
    separator - the separator to use
    
    Returns:
    
    the new double array based on the given data
  - mean
```
public static double mean(double[] vector)
```
    Computes the mean for an array of doubles.
    
    Parameters:
    
    vector - the array
    
    Returns:
    
    the mean
  - choleskyFromMatrix
```
public org.apache.commons.math3.linear.CholeskyDecomposition choleskyFromMatrix(org.apache.commons.math3.linear.RealMatrix m)
                                                                         throws Exception
```
    This will return the cholesky decomposition of the given matrix
    
    Parameters:
    
    m - the matrix to convert
    
    Returns:
    
    the cholesky decomposition of the given matrix. See: http://en.wikipedia.org/wiki/Cholesky_decomposition
    
    Throws:
    
    org.apache.commons.math3.linear.NonSquareMatrixException
    
    Exception
  - toDecimal
```
public static int toDecimal(String binary)
```
    This will convert the given binary string to a decimal based integer
    
    Parameters:
    
    binary - the binary string to convert
    
    Returns:
    
    an equivalent base 10 number
  - distanceFinderZValue
```
public static int distanceFinderZValue(double[] vector)
```
    This will translate a vector in to an equivalent integer
    
    Parameters:
    
    vector - the vector to translate
    
    Returns:
    
    a z value such that the value is the interleaved lsd to msd for each double in the vector
  - euclideanDistance
```
public static double euclideanDistance(double[] p,
                                       double[] q)
```
    This returns the distance of two vectors sum(i=1,n) (q_i - p_i)^2
    
    Parameters:
    
    p - the first vector
    
    q - the second vector
    
    Returns:
    
    the distance between two vectors
  - euclideanDistance
```
public static double euclideanDistance(float[] p,
                                       float[] q)
```
    This returns the distance of two vectors sum(i=1,n) (q_i - p_i)^2
    
    Parameters:
    
    p - the first vector
    
    q - the second vector
    
    Returns:
    
    the distance between two vectors
  - generateUniform
```
public static double[] generateUniform(int l)
```
    This will generate a series of uniformally distributed numbers between l times
    
    Parameters:
    
    l - the number of numbers to generate
    
    Returns:
    
    l uniformally generated numbers
  - manhattanDistance
```
public static double manhattanDistance(double[] p,
                                       double[] q)
```
    This will calculate the Manhattan distance between two sets of points. The Manhattan distance is equivalent to: 1_sum_n |p_i - q_i|
    
    Parameters:
    
    p - the first point vector
    
    q - the second point vector
    
    Returns:
    
    the Manhattan distance between two object
  - sampleDoublesInInterval
```
public static double[] sampleDoublesInInterval(double[][] doubles,
                                               int l)
```
  - randomNumberBetween
```
public static int randomNumberBetween(double begin,
                                      double end)
```
    Generates a random integer between the specified numbers
    
    Parameters:
    
    begin - the begin of the interval
    
    end - the end of the interval
    
    Returns:
    
    an int between begin and end
  - randomNumberBetween
```
public static int randomNumberBetween(double begin,
                                      double end,
                                      org.apache.commons.math3.random.RandomGenerator rng)
```
    Generates a random integer between the specified numbers
    
    Parameters:
    
    begin - the begin of the interval
    
    end - the end of the interval
    
    Returns:
    
    an int between begin and end
  - randomFloatBetween
```
public static float randomFloatBetween(float begin,
                                       float end)
```
  - randomDoubleBetween
```
public static double randomDoubleBetween(double begin,
                                         double end)
```
  - shuffleArray
```
public static void shuffleArray(int[] array,
                                long rngSeed)
```
  - shuffleArray
```
public static void shuffleArray(int[] array,
                                Random rng)
```

Class MathUtils

Field Summary

Constructor Summary

Method Summary

Methods inherited from class java.lang.Object

Field Detail

log2

SMALL

Constructor Detail

MathUtils

Method Detail

normalize

clamp

discretize

nextPowOf2

binomial

uniform

correlation

sigmoid

ssReg

ssError

stringSimilarity

vectorLength

idf

tf

tfidf

ssTotal

sum

mergeCoords

mergeCoords

weightsFor

squaredLoss

w_1

w_0

weightsFor

errorFor

sumOfMeanDifferences

sumOfMeanDifferencesOnePoint

variance

times

sumOfProducts

coordSplit

partitionVariable

coordSplit

xVals

yVals

sumOfSquares

determinationCoefficient

log2

slope

rootMeansSquaredError

entropy

kroneckerDelta

adjustedrSquared

normalizeToOne

min

max

normalize

logs2probs

information

maxIndex

factorial

probToLogOdds

round

permutation

combination

hypotenuse

probRound

roundDouble

roundFloat

bernoullis

sm

gr

fromString

mean

choleskyFromMatrix

toDecimal

distanceFinderZValue

euclideanDistance

euclideanDistance