org.nd4j.autodiff.samediff.ops

Class SDLoss

java.lang.Object

org.nd4j.autodiff.samediff.ops.SDOps

org.nd4j.autodiff.samediff.ops.SDLoss

public class SDLoss
extends SDOps

SameDiff loss functions
Accessible via SameDiff.loss()

Author:

Alex Black

Field Summary

Fields inherited from class org.nd4j.autodiff.samediff.ops.SDOps

sd

Constructor Summary

Constructors

Constructor and Description
SDLoss(SameDiff sameDiff)

Method Summary

Modifier and Type	Method and Description
SDVariable	absoluteDifference(String name, SDVariable label, SDVariable predictions) See absoluteDifference(String, SDVariable, SDVariable, SDVariable, LossReduce).
SDVariable	absoluteDifference(String name, SDVariable label, SDVariable predictions, LossReduce lossReduce) See absoluteDifference(String, SDVariable, SDVariable, SDVariable, LossReduce).
SDVariable	absoluteDifference(String name, SDVariable label, SDVariable predictions, SDVariable weights, LossReduce lossReduce) Absolute difference loss: {@code sum_i abs( label[i] - predictions[i] )
SDVariable	cosineDistance(String name, SDVariable label, SDVariable predictions, int dimension) See cosineDistance(String, SDVariable, SDVariable, SDVariable, LossReduce, int).
SDVariable	cosineDistance(String name, SDVariable label, SDVariable predictions, LossReduce lossReduce, int dimension) See cosineDistance(String, SDVariable, SDVariable, SDVariable, LossReduce, int).
SDVariable	cosineDistance(String name, SDVariable label, SDVariable predictions, SDVariable weights, LossReduce lossReduce, int dimension) Cosine distance loss: 1 - cosineSimilarity(x,y) or 1 - sum_i label[i] * prediction[i], which is equivalent to cosine distance when both the predictions and labels are normalized. Note: This loss function assumes that both the predictions and labels are normalized to have unit l2 norm.
SDVariable	hingeLoss(String name, SDVariable label, SDVariable predictions) See hingeLoss(String, SDVariable, SDVariable, SDVariable, LossReduce).
SDVariable	hingeLoss(String name, SDVariable label, SDVariable predictions, LossReduce lossReduce) See hingeLoss(String, SDVariable, SDVariable, SDVariable, LossReduce).
SDVariable	hingeLoss(String name, SDVariable label, SDVariable predictions, SDVariable weights, LossReduce lossReduce) Hinge loss: a loss function used for training classifiers.
SDVariable	huberLoss(String name, SDVariable label, SDVariable predictions, double delta) See huberLoss(String, SDVariable, SDVariable, SDVariable, LossReduce, double).
SDVariable	huberLoss(String name, SDVariable label, SDVariable predictions, LossReduce lossReduce, double delta) See huberLoss(String, SDVariable, SDVariable, SDVariable, LossReduce, double).
SDVariable	huberLoss(String name, SDVariable label, SDVariable predictions, SDVariable weights, LossReduce lossReduce, double delta) Huber loss function, used for robust regression.
SDVariable	l2Loss(SDVariable var) L2 loss: 1/2 * sum(x^2)
SDVariable	l2Loss(String name, SDVariable var) L2 loss: 1/2 * sum(x^2)
SDVariable	logLoss(String name, SDVariable label, SDVariable predictions) See logLoss(String, SDVariable, SDVariable, SDVariable, LossReduce, double).
SDVariable	logLoss(String name, SDVariable label, SDVariable predictions, LossReduce lossReduce) See logLoss(String, SDVariable, SDVariable, SDVariable, LossReduce, double).
SDVariable	logLoss(String name, SDVariable label, SDVariable predictions, SDVariable weights, LossReduce lossReduce, double epsilon) Log loss, i.e., binary cross entropy loss, usually used for binary multi-label classification.
SDVariable	logPoisson(String name, SDVariable label, SDVariable predictions) See logPoisson(String, SDVariable, SDVariable, SDVariable, LossReduce).
SDVariable	logPoisson(String name, SDVariable label, SDVariable predictions, LossReduce lossReduce) See logPoisson(String, SDVariable, SDVariable, SDVariable, LossReduce).
SDVariable	logPoisson(String name, SDVariable label, SDVariable predictions, SDVariable weights, LossReduce lossReduce) Log poisson loss: a loss function used for training classifiers.
SDVariable	logPoissonFull(String name, SDVariable label, SDVariable predictions) See logPoissonFull(String, SDVariable, SDVariable, SDVariable, LossReduce).
SDVariable	logPoissonFull(String name, SDVariable label, SDVariable predictions, LossReduce lossReduce) See logPoissonFull(String, SDVariable, SDVariable, SDVariable, LossReduce).
SDVariable	logPoissonFull(String name, SDVariable label, SDVariable predictions, SDVariable weights, LossReduce lossReduce) Log poisson loss: a loss function used for training classifiers.
SDVariable	meanPairwiseSquaredError(String name, SDVariable label, SDVariable predictions, LossReduce lossReduce) See meanPairwiseSquaredError(String, SDVariable, SDVariable, SDVariable, LossReduce).
SDVariable	meanPairwiseSquaredError(String name, SDVariable label, SDVariable predictions, SDVariable weights, LossReduce lossReduce) Mean pairwise squared error. MPWSE loss calculates the difference between pairs of consecutive elements in the predictions and labels arrays.
SDVariable	meanSquaredError(String name, SDVariable label, SDVariable predictions) See meanSquaredError(String, SDVariable, SDVariable, SDVariable, LossReduce).
SDVariable	meanSquaredError(String name, SDVariable label, SDVariable predictions, LossReduce lossReduce) See meanSquaredError(String, SDVariable, SDVariable, SDVariable, LossReduce).
SDVariable	meanSquaredError(String name, SDVariable label, SDVariable predictions, SDVariable weights, LossReduce lossReduce) Mean squared error loss function.
SDVariable	sigmoidCrossEntropy(String name, SDVariable label, SDVariable predictions) See sigmoidCrossEntropy(String, SDVariable, SDVariable, SDVariable, LossReduce, double).
SDVariable	sigmoidCrossEntropy(String name, SDVariable label, SDVariable predictions, LossReduce lossReduce) See sigmoidCrossEntropy(String, SDVariable, SDVariable, SDVariable, LossReduce, double).
SDVariable	sigmoidCrossEntropy(String name, SDVariable label, SDVariable predictionLogits, SDVariable weights, LossReduce lossReduce, double labelSmoothing) Sigmoid cross entropy: applies the sigmoid activation function on the input logits (input "pre-sigmoid preductions") and implements the binary cross entropy loss function.
SDVariable	softmaxCrossEntropy(String name, SDVariable label, SDVariable predictions) See softmaxCrossEntropy(String, SDVariable, SDVariable, SDVariable, LossReduce, double).
SDVariable	softmaxCrossEntropy(String name, SDVariable label, SDVariable predictions, LossReduce lossReduce) See softmaxCrossEntropy(String, SDVariable, SDVariable, SDVariable, LossReduce, double).
SDVariable	softmaxCrossEntropy(String name, SDVariable oneHotLabels, SDVariable logitPredictions, SDVariable weights, LossReduce lossReduce, double labelSmoothing) Applies the softmax activation function to the input, then implement multi-class cross entropy: -sum_classes label[i] * log(p[c]) where p = softmax(logits) If LossReduce.NONE is used, returned shape is [numExamples] out for [numExamples, numClasses] predicitons/labels; otherwise, the output is a scalar.
SDVariable	sparseSoftmaxCrossEntropy(SDVariable logits, SDVariable labels) See sparseSoftmaxCrossEntropy(String, SDVariable, SDVariable)
SDVariable	sparseSoftmaxCrossEntropy(String name, SDVariable logits, SDVariable labels) As per softmaxCrossEntropy(String, SDVariable, SDVariable, LossReduce) but the labels variable is represented as an integer array instead of the equivalent one-hot array. i.e., if logits are rank N, then labels have rank N-1
SDVariable	weightedCrossEntropyWithLogits(SDVariable targets, SDVariable inputs, SDVariable weights) TODO
SDVariable	weightedCrossEntropyWithLogits(String name, SDVariable targets, SDVariable inputs, SDVariable weights) TODO

Methods inherited from class org.nd4j.autodiff.samediff.ops.SDOps

f, updateVariableNameAndReference

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

SDLoss

public SDLoss(SameDiff sameDiff)

Method Detail

absoluteDifference

public SDVariable absoluteDifference(String name,
                                     @NonNull
                                     SDVariable label,
                                     @NonNull
                                     SDVariable predictions)

See absoluteDifference(String, SDVariable, SDVariable, SDVariable, LossReduce).

absoluteDifference

public SDVariable absoluteDifference(String name,
                                     @NonNull
                                     SDVariable label,
                                     @NonNull
                                     SDVariable predictions,
                                     SDVariable weights,
                                     @NonNull
                                     LossReduce lossReduce)

Absolute difference loss: {@code sum_i abs( label[i] - predictions[i] )

Parameters:

name - Name of the operation

label - Label array

predictions - Predictions array

weights - Weights array. May be null. If null, a weight of 1.0 is used

lossReduce - Reduction type for the loss. See LossReduce for more details. Default: LossReduce.MEAN_BY_NONZERO_WEIGHT_COUNT

Returns:

Loss variable

absoluteDifference

public SDVariable absoluteDifference(String name,
                                     @NonNull
                                     SDVariable label,
                                     @NonNull
                                     SDVariable predictions,
                                     @NonNull
                                     LossReduce lossReduce)

See absoluteDifference(String, SDVariable, SDVariable, SDVariable, LossReduce).

cosineDistance

public SDVariable cosineDistance(String name,
                                 @NonNull
                                 SDVariable label,
                                 @NonNull
                                 SDVariable predictions,
                                 int dimension)

See cosineDistance(String, SDVariable, SDVariable, SDVariable, LossReduce, int).

cosineDistance

public SDVariable cosineDistance(String name,
                                 @NonNull
                                 SDVariable label,
                                 @NonNull
                                 SDVariable predictions,
                                 SDVariable weights,
                                 @NonNull
                                 LossReduce lossReduce,
                                 int dimension)

Cosine distance loss: 1 - cosineSimilarity(x,y) or 1 - sum_i label[i] * prediction[i], which is equivalent to cosine distance when both the predictions and labels are normalized.
Note: This loss function assumes that both the predictions and labels are normalized to have unit l2 norm. If this is not the case, you should normalize them first by dividing by SDBaseOps.norm2(String, SDVariable, boolean, int...) along the cosine distance dimension (with keepDims=true).

Parameters:

name - Name of the operation

label - Label array

predictions - Predictions array

weights - Weights array. May be null. If null, a weight of 1.0 is used

lossReduce - Reduction type for the loss. See LossReduce for more details. Default: LossReduce.MEAN_BY_NONZERO_WEIGHT_COUNT

dimension - Dimension to perform the cosine distance over

Returns:

Cosine distance loss variable

cosineDistance

public SDVariable cosineDistance(String name,
                                 @NonNull
                                 SDVariable label,
                                 @NonNull
                                 SDVariable predictions,
                                 @NonNull
                                 LossReduce lossReduce,
                                 int dimension)

See cosineDistance(String, SDVariable, SDVariable, SDVariable, LossReduce, int).

hingeLoss

public SDVariable hingeLoss(String name,
                            @NonNull
                            SDVariable label,
                            @NonNull
                            SDVariable predictions)

See hingeLoss(String, SDVariable, SDVariable, SDVariable, LossReduce).

hingeLoss

public SDVariable hingeLoss(String name,
                            @NonNull
                            SDVariable label,
                            @NonNull
                            SDVariable predictions,
                            SDVariable weights,
                            @NonNull
                            LossReduce lossReduce)

Hinge loss: a loss function used for training classifiers. Implements L = max(0, 1 - t * predictions) where t is the label values after internally converting to {-1,1} from the user specified {0,1}. Note that Labels should be provided with values {0,1}.

Parameters:

name - Name of the operation

label - Label array. Each value should be 0.0 or 1.0 (internally -1 to 1 is used)

predictions - Predictions array

weights - Weights array. May be null. If null, a weight of 1.0 is used

lossReduce - Reduction type for the loss. See LossReduce for more details. Default: LossReduce.MEAN_BY_NONZERO_WEIGHT_COUNT

Returns:

Loss variable

hingeLoss

public SDVariable hingeLoss(String name,
                            @NonNull
                            SDVariable label,
                            @NonNull
                            SDVariable predictions,
                            @NonNull
                            LossReduce lossReduce)

See hingeLoss(String, SDVariable, SDVariable, SDVariable, LossReduce).

huberLoss

public SDVariable huberLoss(String name,
                            @NonNull
                            SDVariable label,
                            @NonNull
                            SDVariable predictions,
                            double delta)

See huberLoss(String, SDVariable, SDVariable, SDVariable, LossReduce, double).

huberLoss

public SDVariable huberLoss(String name,
                            @NonNull
                            SDVariable label,
                            @NonNull
                            SDVariable predictions,
                            SDVariable weights,
                            @NonNull
                            LossReduce lossReduce,
                            double delta)

Huber loss function, used for robust regression. It is similar both squared error loss and absolute difference loss, though is less sensitive to outliers than squared error.
Huber loss implements:

  L = 0.5 * (label[i] - predictions[i])^2 if abs(label[i] - predictions[i]) < delta
  L = delta * abs(label[i] - predictions[i]) - 0.5 * delta^2 otherwise
     
 

Parameters:

name - Name of the operation

label - Label array

predictions - Predictions array

weights - Weights array. May be null. If null, a weight of 1.0 is used

lossReduce - Reduction type for the loss. See LossReduce for more details. Default: LossReduce.MEAN_BY_NONZERO_WEIGHT_COUNT

delta - Loss function delta value

Returns:

Huber loss variable

huberLoss

public SDVariable huberLoss(String name,
                            @NonNull
                            SDVariable label,
                            @NonNull
                            SDVariable predictions,
                            @NonNull
                            LossReduce lossReduce,
                            double delta)

See huberLoss(String, SDVariable, SDVariable, SDVariable, LossReduce, double).

l2Loss

public SDVariable l2Loss(@NonNull
                         SDVariable var)

L2 loss: 1/2 * sum(x^2)

Parameters:

var - Variable to calculate L2 loss of

Returns:

L2 loss

l2Loss

public SDVariable l2Loss(String name,
                         @NonNull
                         SDVariable var)

L2 loss: 1/2 * sum(x^2)

Parameters:

name - Name of the output variable

var - Variable to calculate L2 loss of

Returns:

L2 loss

logLoss

public SDVariable logLoss(String name,
                          @NonNull
                          SDVariable label,
                          @NonNull
                          SDVariable predictions)

See logLoss(String, SDVariable, SDVariable, SDVariable, LossReduce, double).

logLoss

public SDVariable logLoss(String name,
                          @NonNull
                          SDVariable label,
                          @NonNull
                          SDVariable predictions,
                          SDVariable weights,
                          @NonNull
                          LossReduce lossReduce,
                          double epsilon)

Log loss, i.e., binary cross entropy loss, usually used for binary multi-label classification. Implements: -1/numExamples * sum_i (labels[i] * log(predictions[i] + epsilon) + (1-labels[i]) * log(1-predictions[i] + epsilon))

Parameters:

name - Name of the operation

label - Label array

predictions - Predictions array

weights - Weights array. May be null. If null, a weight of 1.0 is used

lossReduce - Reduction type for the loss. See LossReduce for more details. Default: LossReduce.MEAN_BY_NONZERO_WEIGHT_COUNT

Returns:

Log loss variable

logLoss

public SDVariable logLoss(String name,
                          @NonNull
                          SDVariable label,
                          @NonNull
                          SDVariable predictions,
                          @NonNull
                          LossReduce lossReduce)

See logLoss(String, SDVariable, SDVariable, SDVariable, LossReduce, double).

logPoisson

public SDVariable logPoisson(String name,
                             @NonNull
                             SDVariable label,
                             @NonNull
                             SDVariable predictions)

See logPoisson(String, SDVariable, SDVariable, SDVariable, LossReduce).

logPoisson

public SDVariable logPoisson(String name,
                             @NonNull
                             SDVariable label,
                             @NonNull
                             SDVariable predictions,
                             SDVariable weights,
                             @NonNull
                             LossReduce lossReduce)

Log poisson loss: a loss function used for training classifiers. Implements L = exp(c) - z * c where c is log(predictions) and z is labels.

Parameters:

name - Name of the operation

label - Label array. Each value should be 0.0 or 1.0

predictions - Predictions array (has to be log(x) of actual predictions)

weights - Weights array. May be null. If null, a weight of 1.0 is used

lossReduce - Reduction type for the loss. See LossReduce for more details. Default: LossReduce.MEAN_BY_NONZERO_WEIGHT_COUNT

Returns:

Loss variable

logPoisson

public SDVariable logPoisson(String name,
                             @NonNull
                             SDVariable label,
                             @NonNull
                             SDVariable predictions,
                             @NonNull
                             LossReduce lossReduce)

See logPoisson(String, SDVariable, SDVariable, SDVariable, LossReduce).

logPoissonFull

public SDVariable logPoissonFull(String name,
                                 @NonNull
                                 SDVariable label,
                                 @NonNull
                                 SDVariable predictions)

See logPoissonFull(String, SDVariable, SDVariable, SDVariable, LossReduce).

logPoissonFull

public SDVariable logPoissonFull(String name,
                                 @NonNull
                                 SDVariable label,
                                 @NonNull
                                 SDVariable predictions,
                                 SDVariable weights,
                                 @NonNull
                                 LossReduce lossReduce)

Log poisson loss: a loss function used for training classifiers. Implements L = exp(c) - z * c + z * log(z) - z + 0.5 * log(2 * pi * z) where c is log(predictions) and z is labels.

Parameters:

name - Name of the operation

label - Label array. Each value should be 0.0 or 1.0

predictions - Predictions array (has to be log(x) of actual predictions)

weights - Weights array. May be null. If null, a weight of 1.0 is used

lossReduce - Reduction type for the loss. See LossReduce for more details. Default: LossReduce.MEAN_BY_NONZERO_WEIGHT_COUNT

Returns:

Loss variable

logPoissonFull

public SDVariable logPoissonFull(String name,
                                 @NonNull
                                 SDVariable label,
                                 @NonNull
                                 SDVariable predictions,
                                 @NonNull
                                 LossReduce lossReduce)

See logPoissonFull(String, SDVariable, SDVariable, SDVariable, LossReduce).

meanPairwiseSquaredError

public SDVariable meanPairwiseSquaredError(String name,
                                           @NonNull
                                           SDVariable label,
                                           @NonNull
                                           SDVariable predictions,
                                           @NonNull
                                           LossReduce lossReduce)

See meanPairwiseSquaredError(String, SDVariable, SDVariable, SDVariable, LossReduce).

meanPairwiseSquaredError

public SDVariable meanPairwiseSquaredError(String name,
                                           @NonNull
                                           SDVariable label,
                                           @NonNull
                                           SDVariable predictions,
                                           SDVariable weights,
                                           @NonNull
                                           LossReduce lossReduce)

Mean pairwise squared error.
MPWSE loss calculates the difference between pairs of consecutive elements in the predictions and labels arrays. For example, if predictions = [p0, p1, p2] and labels are [l0, l1, l2] then MPWSE is: [((p0-p1) - (l0-l1))^2 + ((p0-p2) - (l0-l2))^2 + ((p1-p2) - (l1-l2))^2] / 3

Parameters:

name - Name of the operation

label - Label array

predictions - Predictions array

weights - Weights array. May be null. If null, a weight of 1.0 is used. Must be either null, scalar, or have shape [batchSize]

Returns:

Loss variable, scalar output

meanSquaredError

public SDVariable meanSquaredError(String name,
                                   @NonNull
                                   SDVariable label,
                                   @NonNull
                                   SDVariable predictions)

See meanSquaredError(String, SDVariable, SDVariable, SDVariable, LossReduce).

meanSquaredError

public SDVariable meanSquaredError(String name,
                                   @NonNull
                                   SDVariable label,
                                   @NonNull
                                   SDVariable predictions,
                                   SDVariable weights,
                                   @NonNull
                                   LossReduce lossReduce)

Mean squared error loss function. Implements (label[i] - prediction[i])^2 - i.e., squared error on a per-element basis. When averaged (using LossReduce.MEAN_BY_WEIGHT or LossReduce.MEAN_BY_NONZERO_WEIGHT_COUNT (the default)) this is the mean squared error loss function.

Parameters:

name - Name of the operation

label - Label array

predictions - Predictions array

weights - Weights array. May be null. If null, a weight of 1.0 is used

lossReduce - Reduction type for the loss. See LossReduce for more details. Default: LossReduce.MEAN_BY_NONZERO_WEIGHT_COUNT

Returns:

Loss variable

meanSquaredError

public SDVariable meanSquaredError(String name,
                                   @NonNull
                                   SDVariable label,
                                   @NonNull
                                   SDVariable predictions,
                                   @NonNull
                                   LossReduce lossReduce)

See meanSquaredError(String, SDVariable, SDVariable, SDVariable, LossReduce).

sigmoidCrossEntropy

public SDVariable sigmoidCrossEntropy(String name,
                                      @NonNull
                                      SDVariable label,
                                      @NonNull
                                      SDVariable predictions)

See sigmoidCrossEntropy(String, SDVariable, SDVariable, SDVariable, LossReduce, double).

sigmoidCrossEntropy

public SDVariable sigmoidCrossEntropy(String name,
                                      @NonNull
                                      SDVariable label,
                                      @NonNull
                                      SDVariable predictionLogits,
                                      SDVariable weights,
                                      @NonNull
                                      LossReduce lossReduce,
                                      double labelSmoothing)

Sigmoid cross entropy: applies the sigmoid activation function on the input logits (input "pre-sigmoid preductions") and implements the binary cross entropy loss function. This implementation is numerically more stable than using standard (but separate) sigmoid activation function and log loss (binary cross entropy) loss function.
Implements: -1/numExamples * sum_i (labels[i] * log(sigmoid(logits[i])) + (1-labels[i]) * log(1-sigmoid(logits[i]))) though this is done in a mathematically equivalent but more numerical stable form.

When label smoothing is > 0, the following label smoothing is used:

  numClasses = labels.size(1);
 label = (1.0 - labelSmoothing) * label + 0.5 * labelSmoothing
 

Parameters:

name - Name of the operation

label - Label array

predictionLogits - Predictions array

weights - Weights array. May be null. If null, a weight of 1.0 is used

lossReduce - Reduction type for the loss. See LossReduce for more details. Default: LossReduce.MEAN_BY_NONZERO_WEIGHT_COUNT

Returns:

Loss variable

sigmoidCrossEntropy

public SDVariable sigmoidCrossEntropy(String name,
                                      @NonNull
                                      SDVariable label,
                                      @NonNull
                                      SDVariable predictions,
                                      @NonNull
                                      LossReduce lossReduce)

See sigmoidCrossEntropy(String, SDVariable, SDVariable, SDVariable, LossReduce, double).

softmaxCrossEntropy

public SDVariable softmaxCrossEntropy(String name,
                                      @NonNull
                                      SDVariable label,
                                      @NonNull
                                      SDVariable predictions)

See softmaxCrossEntropy(String, SDVariable, SDVariable, SDVariable, LossReduce, double).

softmaxCrossEntropy

public SDVariable softmaxCrossEntropy(String name,
                                      @NonNull
                                      SDVariable oneHotLabels,
                                      @NonNull
                                      SDVariable logitPredictions,
                                      SDVariable weights,
                                      @NonNull
                                      LossReduce lossReduce,
                                      double labelSmoothing)

Applies the softmax activation function to the input, then implement multi-class cross entropy:
-sum_classes label[i] * log(p[c]) where p = softmax(logits)
If LossReduce.NONE is used, returned shape is [numExamples] out for [numExamples, numClasses] predicitons/labels; otherwise, the output is a scalar.

When label smoothing is > 0, the following label smoothing is used:

  numClasses = labels.size(1);
 oneHotLabel = (1.0 - labelSmoothing) * oneHotLabels + labelSmoothing/numClasses
 

Parameters:

name - Name of the operation

oneHotLabels - Label array. Should be one-hot per example and same shape as predictions (for example, [mb, nOut])

logitPredictions - Predictions array (pre-softmax)

weights - Weights array. May be null. If null, a weight of 1.0 is used

lossReduce - Reduction type for the loss. See LossReduce for more details. Default: LossReduce.MEAN_BY_NONZERO_WEIGHT_COUNT

labelSmoothing - Label smoothing value. Default value: 0

Returns:

Loss variable

softmaxCrossEntropy

public SDVariable softmaxCrossEntropy(String name,
                                      @NonNull
                                      SDVariable label,
                                      @NonNull
                                      SDVariable predictions,
                                      @NonNull
                                      LossReduce lossReduce)

See softmaxCrossEntropy(String, SDVariable, SDVariable, SDVariable, LossReduce, double).

sparseSoftmaxCrossEntropy

public SDVariable sparseSoftmaxCrossEntropy(@NonNull
                                            SDVariable logits,
                                            @NonNull
                                            SDVariable labels)

See sparseSoftmaxCrossEntropy(String, SDVariable, SDVariable)

sparseSoftmaxCrossEntropy

public SDVariable sparseSoftmaxCrossEntropy(String name,
                                            @NonNull
                                            SDVariable logits,
                                            @NonNull
                                            SDVariable labels)

As per softmaxCrossEntropy(String, SDVariable, SDVariable, LossReduce) but the labels variable is represented as an integer array instead of the equivalent one-hot array.
i.e., if logits are rank N, then labels have rank N-1

Parameters:

name - Name of the output variable. May be null

logits - Logits array ("pre-softmax activations")

labels - Labels array. Must be an integer type.

Returns:

Softmax cross entropy

weightedCrossEntropyWithLogits

public SDVariable weightedCrossEntropyWithLogits(SDVariable targets,
                                                 SDVariable inputs,
                                                 SDVariable weights)

TODO

Parameters:

targets -

inputs -

weights -

Returns:

weightedCrossEntropyWithLogits

public SDVariable weightedCrossEntropyWithLogits(String name,
                                                 SDVariable targets,
                                                 SDVariable inputs,
                                                 SDVariable weights)

TODO

Parameters:

name -

targets -

inputs -

weights -

Returns: