GlrmLoss (h2o-genmodel 3.46.0 API)

java.lang.Object
- java.lang.Enum<GlrmLoss>
- - hex.genmodel.algos.glrm.GlrmLoss

All Implemented Interfaces:

java.io.Serializable, java.lang.Comparable<GlrmLoss>
```
public enum GlrmLoss
extends java.lang.Enum<GlrmLoss>
```
Loss function for the GLRM algorithm.

Enum Constant Summary

Enum Constants
Enum Constant and Description

Absolute

Categorical

Hinge

Huber

Logistic

Ordinal

Periodic

Poisson

Quadratic

Enum Constants
Enum Constant and Description
`Absolute`
`Categorical`
`Hinge`
`Huber`
`Logistic`
`Ordinal`
`Periodic`
`Poisson`
`Quadratic`

Method Summary

All Methods Static Methods Instance Methods Abstract Methods Concrete Methods
Modifier and Type	Method and Description
`double`	`impute(double u)` \argmin_a L(u, a): Data imputation for real numeric values
`abstract boolean`	`isForBinary()`
`abstract boolean`	`isForCategorical()`
`abstract boolean`	`isForNumeric()`
`double`	`lgrad(double u, double a)` \grad_u L(u,a): Derivative of the numeric loss function with respect to u
`double`	`loss(double u, double a)` Loss function for numeric variables
`int`	`mimpute(double[] u)` \argmin_a L(u, a): Data imputation for categorical values {0, 1, 2, ...}
`double[]`	`mlgrad(double[] u, int a)` \grad_u L(u,a): Gradient of multidimensional loss function with respect to u
`double[]`	`mlgrad(double[] u, int a, double[] prod, int u_len)` \grad_u L(u,a): Gradient of multidimensional loss function with respect to u.
`double`	`mloss(double[] u, int a)` Loss function for categorical variables where the size of u represents the true column length.
`double`	`mloss(double[] u, int a, int u_len)` Loss function for categorical variables performing same function as mloss above.
`void`	`setParameters(int p)` Initialize additional parameters on the loss function.
`static GlrmLoss`	`valueOf(java.lang.String name)` Returns the enum constant of this type with the specified name.
`static GlrmLoss[]`	`values()` Returns an array containing the constants of this enum type, in the order they are declared.

Methods inherited from class java.lang.Enum
clone, compareTo, equals, finalize, getDeclaringClass, hashCode, name, ordinal, toString, valueOf

Methods inherited from class java.lang.Object
getClass, notify, notifyAll, wait, wait, wait

- Enum Constant Detail
  - Quadratic
```
public static final GlrmLoss Quadratic
```
  - Absolute
```
public static final GlrmLoss Absolute
```
  - Huber
```
public static final GlrmLoss Huber
```
  - Poisson
```
public static final GlrmLoss Poisson
```
  - Periodic
```
public static final GlrmLoss Periodic
```
  - Logistic
```
public static final GlrmLoss Logistic
```
  - Hinge
```
public static final GlrmLoss Hinge
```
  - Categorical
```
public static final GlrmLoss Categorical
```
  - Ordinal
```
public static final GlrmLoss Ordinal
```
- Method Detail
  - values
```
public static GlrmLoss[] values()
```
    Returns an array containing the constants of this enum type, in the order they are declared. This method may be used to iterate over the constants as follows:
```
for (GlrmLoss c : GlrmLoss.values())
    System.out.println(c);
```
    Returns:
    
    an array containing the constants of this enum type, in the order they are declared
  - valueOf
```
public static GlrmLoss valueOf(java.lang.String name)
```
    Returns the enum constant of this type with the specified name. The string must match exactly an identifier used to declare an enum constant in this type. (Extraneous whitespace characters are not permitted.)
    
    Parameters:
    
    name - the name of the enum constant to be returned.
    
    Returns:
    
    the enum constant with the specified name
    
    Throws:
    
    java.lang.IllegalArgumentException - if this enum type has no constant with the specified name
    
    java.lang.NullPointerException - if the argument is null
  - isForNumeric
```
public abstract boolean isForNumeric()
```
  - isForCategorical
```
public abstract boolean isForCategorical()
```
  - isForBinary
```
public abstract boolean isForBinary()
```
  - loss
```
public double loss(double u,
                   double a)
```
    Loss function for numeric variables
  - lgrad
```
public double lgrad(double u,
                    double a)
```
    \grad_u L(u,a): Derivative of the numeric loss function with respect to u
  - impute
```
public double impute(double u)
```
    \argmin_a L(u, a): Data imputation for real numeric values
  - mloss
```
public double mloss(double[] u,
                    int a)
```
    Loss function for categorical variables where the size of u represents the true column length.
  - mloss
```
public double mloss(double[] u,
                    int a,
                    int u_len)
```
    Loss function for categorical variables performing same function as mloss above. However, in this case, the size of u can be much bigger than what is needed. The actual length of u is now specified in u_len.
  - mlgrad
```
public double[] mlgrad(double[] u,
                       int a)
```
    \grad_u L(u,a): Gradient of multidimensional loss function with respect to u
  - mlgrad
```
public double[] mlgrad(double[] u,
                       int a,
                       double[] prod,
                       int u_len)
```
    \grad_u L(u,a): Gradient of multidimensional loss function with respect to u. This method avoids the memory allocation compared to the method above by passing in a array prod which can be longer than the actual column length. The actual column length for prod is now specified by u_len.
  - mimpute
```
public int mimpute(double[] u)
```
    \argmin_a L(u, a): Data imputation for categorical values {0, 1, 2, ...}
  - setParameters
```
public void setParameters(int p)
```
    Initialize additional parameters on the loss function. Currently used by Periodic class only.

Enum GlrmLoss

Enum Constant Summary

Method Summary

Methods inherited from class java.lang.Enum

Methods inherited from class java.lang.Object

Enum Constant Detail

Quadratic

Absolute

Huber

Poisson

Periodic

Logistic

Hinge

Categorical

Ordinal

Method Detail

values

valueOf

isForNumeric

isForCategorical

isForBinary

loss

lgrad

impute

mloss

mloss

mlgrad

mlgrad

mimpute

setParameters