add nn ops for sigmoidCrossEntropyWithLogits, softmaxCrossEntropyWith…

…Logits and sparseSoftmaxCrossEntropyWithLogits

tensorflow-core/tensorflow-core-api/src/gen/annotations/org/tensorflow/op/NnOps.java

@@ -1811,14 +1811,14 @@ public <T extends TNumber> Softmax<T> softmax(Operand<T> logits) {
 
   /**
    * Computes softmax cross entropy cost and gradients to backpropagate.
-   *  <p>
    *  Inputs are the logits, not probabilities.
    *
-   * @param <T> data type for {@code loss()} output
+   * @param <T> data type for {@code loss} output
    * @param features batch_size x num_classes matrix
    * @param labels batch_size x num_classes matrix
    *  The caller must ensure that each batch of labels represents a valid
    *  probability distribution.
+   * @param <T> data type for {@code SoftmaxCrossEntropyWithLogits} output and operands
    * @return a new instance of SoftmaxCrossEntropyWithLogits
    */
   public <T extends TNumber> SoftmaxCrossEntropyWithLogits<T> softmaxCrossEntropyWithLogits(
@@ -2011,18 +2011,17 @@ public <T extends TType> SpaceToDepth<T> spaceToDepth(Operand<T> input, Long blo
 
   /**
    * Computes softmax cross entropy cost and gradients to backpropagate.
-   *  <p>
-   *  Unlike `SoftmaxCrossEntropyWithLogits`, this operation does not accept
+   *  Unlike {@code SoftmaxCrossEntropyWithLogits}, this operation does not accept
    *  a matrix of label probabilities, but rather a single label per row
    *  of features.  This label is considered to have probability 1.0 for the
    *  given row.
-   *  <p>
-   *  Inputs are the logits, not probabilities.
+   *  <p>Inputs are the logits, not probabilities.
    *
-   * @param <T> data type for {@code loss()} output
+   * @param <T> data type for {@code loss} output
    * @param features batch_size x num_classes matrix
    * @param labels batch_size vector with values in [0, num_classes).
    *  This is the label for the given minibatch entry.
+   * @param <T> data type for {@code SparseSoftmaxCrossEntropyWithLogits} output and operands
    * @return a new instance of SparseSoftmaxCrossEntropyWithLogits
    */
   public <T extends TNumber> SparseSoftmaxCrossEntropyWithLogits<T> sparseSoftmaxCrossEntropyWithLogits(

tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/SoftmaxCrossEntropyWithLogits.java

@@ -29,57 +29,68 @@
 
 /**
  * Computes softmax cross entropy cost and gradients to backpropagate.
- * <p>
  * Inputs are the logits, not probabilities.
- * 
- * @param <T> data type for {@code loss()} output
+ *
+ * @param <T> data type for {@code loss} output
  */
-@Operator(group = "nn")
+@Operator(
+    group = "nn"
+)
 public final class SoftmaxCrossEntropyWithLogits<T extends TNumber> extends RawOp {
-
+  /**
+   * The name of this op, as known by TensorFlow core engine
+   */
+  public static final String OP_NAME = "SoftmaxCrossEntropyWithLogits";
+
+  private Output<T> loss;
+
+  private Output<T> backprop;
+
+  private SoftmaxCrossEntropyWithLogits(Operation operation) {
+    super(operation);
+    int outputIdx = 0;
+    loss = operation.output(outputIdx++);
+    backprop = operation.output(outputIdx++);
+  }
+
   /**
    * Factory method to create a class wrapping a new SoftmaxCrossEntropyWithLogits operation.
-   * 
+   *
    * @param scope current scope
    * @param features batch_size x num_classes matrix
    * @param labels batch_size x num_classes matrix
    * The caller must ensure that each batch of labels represents a valid
    * probability distribution.
+   * @param <T> data type for {@code SoftmaxCrossEntropyWithLogits} output and operands
    * @return a new instance of SoftmaxCrossEntropyWithLogits
    */
-  @Endpoint(describeByClass = true)
-  public static <T extends TNumber> SoftmaxCrossEntropyWithLogits<T> create(Scope scope, Operand<T> features, Operand<T> labels) {
+  @Endpoint(
+      describeByClass = true
+  )
+  public static <T extends TNumber> SoftmaxCrossEntropyWithLogits<T> create(Scope scope,
+      Operand<T> features, Operand<T> labels) {
     OperationBuilder opBuilder = scope.env().opBuilder("SoftmaxCrossEntropyWithLogits", scope.makeOpName("SoftmaxCrossEntropyWithLogits"));
     opBuilder.addInput(features.asOutput());
     opBuilder.addInput(labels.asOutput());
     opBuilder = scope.apply(opBuilder);
     return new SoftmaxCrossEntropyWithLogits<>(opBuilder.build());
   }
-  
+
   /**
+   * Gets loss.
    * Per example loss (batch_size vector).
+   * @return loss.
    */
   public Output<T> loss() {
     return loss;
   }
-  
+
   /**
+   * Gets backprop.
    * backpropagated gradients (batch_size x num_classes matrix).
+   * @return backprop.
    */
   public Output<T> backprop() {
     return backprop;
   }
-
-  /** The name of this op, as known by TensorFlow core engine */
-  public static final String OP_NAME = "SoftmaxCrossEntropyWithLogits";
-
-  private Output<T> loss;
-  private Output<T> backprop;
-
-  private SoftmaxCrossEntropyWithLogits(Operation operation) {
-    super(operation);
-    int outputIdx = 0;
-    loss = operation.output(outputIdx++);
-    backprop = operation.output(outputIdx);
-  }
 }

tensorflow-core/tensorflow-core-api/src/gen/java/org/tensorflow/op/nn/SparseSoftmaxCrossEntropyWithLogits.java

@@ -29,61 +29,71 @@
 
 /**
  * Computes softmax cross entropy cost and gradients to backpropagate.
- * <p>
- * Unlike `SoftmaxCrossEntropyWithLogits`, this operation does not accept
+ * Unlike {@code SoftmaxCrossEntropyWithLogits}, this operation does not accept
  * a matrix of label probabilities, but rather a single label per row
  * of features.  This label is considered to have probability 1.0 for the
  * given row.
- * <p>
- * Inputs are the logits, not probabilities.
- * 
- * @param <T> data type for {@code loss()} output
+ * <p>Inputs are the logits, not probabilities.
+ *
+ * @param <T> data type for {@code loss} output
  */
-@Operator(group = "nn")
+@Operator(
+    group = "nn"
+)
 public final class SparseSoftmaxCrossEntropyWithLogits<T extends TNumber> extends RawOp {
-
+  /**
+   * The name of this op, as known by TensorFlow core engine
+   */
+  public static final String OP_NAME = "SparseSoftmaxCrossEntropyWithLogits";
+
+  private Output<T> loss;
+
+  private Output<T> backprop;
+
+  private SparseSoftmaxCrossEntropyWithLogits(Operation operation) {
+    super(operation);
+    int outputIdx = 0;
+    loss = operation.output(outputIdx++);
+    backprop = operation.output(outputIdx++);
+  }
+
   /**
    * Factory method to create a class wrapping a new SparseSoftmaxCrossEntropyWithLogits operation.
-   * 
+   *
    * @param scope current scope
    * @param features batch_size x num_classes matrix
    * @param labels batch_size vector with values in [0, num_classes).
    * This is the label for the given minibatch entry.
+   * @param <T> data type for {@code SparseSoftmaxCrossEntropyWithLogits} output and operands
    * @return a new instance of SparseSoftmaxCrossEntropyWithLogits
    */
-  @Endpoint(describeByClass = true)
-  public static <T extends TNumber> SparseSoftmaxCrossEntropyWithLogits<T> create(Scope scope, Operand<T> features, Operand<? extends TNumber> labels) {
+  @Endpoint(
+      describeByClass = true
+  )
+  public static <T extends TNumber> SparseSoftmaxCrossEntropyWithLogits<T> create(Scope scope,
+      Operand<T> features, Operand<? extends TNumber> labels) {
     OperationBuilder opBuilder = scope.env().opBuilder("SparseSoftmaxCrossEntropyWithLogits", scope.makeOpName("SparseSoftmaxCrossEntropyWithLogits"));
     opBuilder.addInput(features.asOutput());
     opBuilder.addInput(labels.asOutput());
     opBuilder = scope.apply(opBuilder);
     return new SparseSoftmaxCrossEntropyWithLogits<>(opBuilder.build());
   }
-  
+
   /**
+   * Gets loss.
    * Per example loss (batch_size vector).
+   * @return loss.
    */
   public Output<T> loss() {
     return loss;
   }
-  
+
   /**
+   * Gets backprop.
    * backpropagated gradients (batch_size x num_classes matrix).
+   * @return backprop.
    */
   public Output<T> backprop() {
     return backprop;
   }
-
-  /** The name of this op, as known by TensorFlow core engine */
-  public static final String OP_NAME = "SparseSoftmaxCrossEntropyWithLogits";
-
-  private Output<T> loss;
-  private Output<T> backprop;
-
-  private SparseSoftmaxCrossEntropyWithLogits(Operation operation) {
-    super(operation);
-    int outputIdx = 0;
-    loss = operation.output(outputIdx++);
-    backprop = operation.output(outputIdx);
-  }
 }

tensorflow-framework/src/main/java/org/tensorflow/framework/op/NnOps.java

@@ -87,7 +87,7 @@ public class NnOps {
    * @param logits the logits of type float32 or float64
    * @param <T> the type of labels and logits
    * @return the component-wise logistic losses.
-   * @throws IllegalArgumentException if logits' and labels' do not have the same shape
+   * @throws IllegalArgumentException if logits and labels do not have the same shape
    */
   public <T extends TNumber> Operand<T> sigmoidCrossEntropyWithLogits(
       Operand<T> labels, Operand<T> logits) {
@@ -139,7 +139,6 @@ public <T extends TNumber> Operand<T> sigmoidCrossEntropyWithLogits(
    * @return the softmax cross entropy loss. Its type is the same as {@code logits} and its shape is
    *     the same as {@code labels} except that it does not have the last dimension of {@code
    *     labels}.
-   *
    */
   public <T extends TNumber, U extends TNumber> Operand<T> softmaxCrossEntropyWithLogits(
       Operand<U> labels, Operand<T> logits, int axis) {
@@ -181,14 +180,14 @@ public <T extends TNumber, U extends TNumber> Operand<T> softmaxCrossEntropyWith
    * @param logits Per-label activations (typically a linear output) of shape {@code [d_0, d_1, ...,
    *     d_{r-1}, numClasses]} and dataType of {@code TFloat16}, {@code TFloat32}, or {@code
    *     TFloat64}. These activation energies are interpreted as unnormalized log probabilities.
-   * @param <T> The data type for the labels
-   * @param <U> The data type for the logits and loss
+   * @param <U> the data type for the labels
+   * @param <T> the data tyoe for the loss and logits.
    * @return the loss
-   * @throws IllegalArgumentException If logits are scalars (need to have {@code rank >= 1}) or if the rank
-   *     of the labels is not equal to the rank of the logits minus one.
+   * @throws IllegalArgumentException If logits are scalars (need to have {@code rank >= 1}) or if
+   *     the rank of the labels is not equal to the rank of the logits minus one.
    */
-  public <T extends TNumber, U extends TNumber> Operand<U> sparseSoftmaxCrossEntropyWithLogits(
-      Operand<T> labels, Operand<U> logits) {
+  public <T extends TNumber, U extends TNumber> Operand<T> sparseSoftmaxCrossEntropyWithLogits(
+      Operand<U> labels, Operand<T> logits) {
     return SparseSoftmaxCrossEntropyWithLogits.sparseSoftmaxCrossEntropyWithLogits(
         scope, labels, logits);
   }

tensorflow-framework/src/main/java/org/tensorflow/framework/op/nn/SigmoidCrossEntropyWithLogits.java

@@ -26,8 +26,7 @@ public class SigmoidCrossEntropyWithLogits {
    * independent and not mutually exclusive. For instance, one could perform multilabel
    * classification where a picture can contain both an elephant and a dog at the same time.
    *
-   * <p>For brevity, let {@code x = logits}, {@code z = labels}. The logistic loss in
-   * pseudo-code is
+   * <p>For brevity, let {@code x = logits}, {@code z = labels}. The logistic loss in pseudo-code is
    *
    * <pre>
    * z * -log(sigmoid(x)) + (1 - z) * -log(1 - sigmoid(x))