Ta strona została przetłumaczona przez Cloud Translation API.

Operand

publiczny interfejs Operand

Znane podklasy pośrednie

All , AllToAll <T>, AnonymousHashTable , AnonymousIteratorV3 , AnonymousMultiDeviceIteratorV3 , AnonymousMutableDenseHashTable , AnonymousMutableHashTable , AnonymousMutableHashTableOfTensors , Any , ApplyAdagradV2 <T>, AssertCardinalityDatasetSet <T> Assert , Asert54 , asert54 .

Interface implemented by operands of a TensorFlow operation.

Example usage:

// The "decodeJpeg" operation can be used as an operand to the "cast" operation
 Operand&lt;UInt8&gt; decodeJpeg = ops.image().decodeJpeg(...);
 ops.math().cast(decodeJpeg, DataType.FLOAT);

 // The output "y" of the "unique" operation can be used as an operand to the "cast" operation
 Output&lt;Integer&gt; y = ops.array().unique(...).y();
 ops.math().cast(y, Float.class);

 // The "split" operation can be used as operand list to the "concat" operation
 Iterable&lt;? extends Operand&lt;Float&gt;&gt; split = ops.array().split(...);
 ops.array().concat(0, split);

Public Methods

abstract Output <T>

asOutput ()

Returns the symbolic handle of a tensor.

Public Methods

public abstract Output <T> asOutput ()

Returns the symbolic handle of a tensor.

Inputs to TensorFlow operations are outputs of another TensorFlow operation. This method is used to obtain a symbolic handle that represents the computation of the input.

Operand

Public Methods

Public Methods

public abstract Output <T> asOutput ()

See Also