TFloat16

public interface TFloat16

IEEE-754 half-precision 16-bit float tensor type.

Since there is no floating-point type that fits in 16 bits in Java, a conversion (with potentially a precision loss) is required for each 32 bits value written or read on a tensor of this type from the JVM. Therefore, if a lot of I/O operations are to be expected on a tensor, performances will be improved by working with TFloat32 or TFloat64 data types whenever possible.

Also, TFloat16 tensors normally perform better if they are located in GPU memory since most CPUs do not support this format natively. For CPU computation on 16-bit floats, the TBfloat16 tensor type might be a better option.

Public Methods

abstract static TFloat16	scalarOf(float value) Allocates a new tensor for storing a single float value.
abstract static TFloat16	tensorOf(Shape shape) Allocates a new tensor of the given shape.
abstract static TFloat16	tensorOf(Shape shape, FloatDataBuffer data) Allocates a new tensor of the given shape, initialized with the provided data.
abstract static TFloat16	tensorOf(Shape shape, Consumer<TFloat16> dataInit) Allocates a new tensor of the given shape and initialize its data.
abstract static TFloat16	tensorOf(NdArray<Float> src) Allocates a new tensor which is a copy of a given array of floats.
abstract static TFloat16	vectorOf(float... values) Allocates a new tensor for storing a vector of floats.

Inherited Methods

From interface org.tensorflow.ndarray.FloatNdArray

abstract FloatNdArray	copyTo(NdArray<Float> dst)
abstract NdArraySequence<FloatNdArray>	elements(int dimensionIdx) Returns a sequence of all elements at a given dimension.
abstract FloatNdArray	get(long... coordinates) Returns the N-dimensional element of this array at the given coordinates.
abstract float	getFloat(long... coordinates) Returns the float value of the scalar found at the given coordinates.
abstract Float	getObject(long... coordinates) Returns the value of the scalar found at the given coordinates.
abstract FloatNdArray	read(FloatDataBuffer dst)
abstract FloatNdArray	read(DataBuffer<Float> dst)
abstract NdArraySequence<FloatNdArray>	scalars() Returns a sequence of all scalars in this array.
abstract FloatNdArray	set(NdArray<Float> src, long... coordinates)
abstract FloatNdArray	setFloat(float value, long... coordinates) Assigns the float value of the scalar found at the given coordinates.
abstract FloatNdArray	setObject(Float value, long... coordinates)
abstract FloatNdArray	slice(Index... coordinates) Creates a multi-dimensional view (or slice) of this array by mapping one or more dimensions to the given index selectors.
abstract FloatNdArray	write(DataBuffer<Float> src)
abstract FloatNdArray	write(FloatDataBuffer src)

From interface org.tensorflow.ndarray.NdArray

abstract NdArray<Float>	copyTo(NdArray<Float> dst) Copy the content of this array to the destination array.
abstract NdArraySequence<? extends NdArray<T>>	elements(int dimensionIdx) Returns a sequence of all elements at a given dimension.
abstract boolean	equals(Object obj) Checks equality between n-dimensional arrays.
abstract NdArray<Float>	get(long... coordinates) Returns the N-dimensional element of this array at the given coordinates.
abstract Float	getObject(long... coordinates) Returns the value of the scalar found at the given coordinates.
abstract NdArray<Float>	read(DataBuffer<Float> dst) Read the content of this N-dimensional array into the destination buffer.
abstract NdArraySequence<? extends NdArray<T>>	scalars() Returns a sequence of all scalars in this array.
abstract NdArray<Float>	set(NdArray<Float> src, long... coordinates) Assigns the value of the N-dimensional element found at the given coordinates.
abstract NdArray<Float>	setObject(Float value, long... coordinates) Assigns the value of the scalar found at the given coordinates.
abstract NdArray<Float>	slice(Index... indices) Creates a multi-dimensional view (or slice) of this array by mapping one or more dimensions to the given index selectors.
abstract NdArray<Float>	write(DataBuffer<Float> src) Write the content of this N-dimensional array from the source buffer.

From interface org.tensorflow.ndarray.Shaped

abstract int	rank()
abstract Shape	shape()
abstract long	size() Computes and returns the total size of this container, in number of values.

From interface org.tensorflow.types.family.TType

abstract void	close() Release resources associated with the Tensor.
abstract DataType	dataType() Returns the DataType of elements stored in the tensor.
abstract long	numBytes() Returns the size, in bytes, of the tensor data.
abstract Class<? extends TType>	type() Returns the type of this tensor as a registered subclass of TType

From interface org.tensorflow.Tensor

abstract RawTensor	asRawTensor() Returns a raw (untyped) representation of this tensor
abstract void	close() Release resources associated with the Tensor.
abstract DataType	dataType() Returns the DataType of elements stored in the tensor.
abstract long	numBytes() Returns the size, in bytes, of the tensor data.
abstract static <T extends TType> T	of(Class<T> type, Shape shape, long size, Consumer<T> dataInitializer) Allocates a tensor of a given datatype, shape and size.
abstract static <T extends TType> T	of(Class<T> type, Shape shape) Allocates a tensor of a given datatype and shape.
abstract static <T extends TType> T	of(Class<T> type, Shape shape, long size) Allocates a tensor of a given datatype, shape and size.
abstract static <T extends TType> T	of(Class<T> type, Shape shape, Consumer<T> dataInitializer) Allocates and initialize a tensor of a given datatype and shape.
abstract static <T extends TType> T	of(Class<T> type, Shape shape, ByteDataBuffer rawData) Creates a Tensor of any type from the raw data provided by the given buffer.
abstract Shape	shape() Returns the shape of the tensor.

From interface java.lang.AutoCloseable

abstract void

close()

Public Methods

public static abstract TFloat16 scalarOf (float value)

Allocates a new tensor for storing a single float value.

Parameters

value	float to store in the new tensor

Returns

• the new tensor

public static abstract TFloat16 tensorOf (Shape shape)

Allocates a new tensor of the given shape.

Parameters

shape	shape of the tensor to allocate

Returns

• the new tensor

public static abstract TFloat16 tensorOf (Shape shape, FloatDataBuffer data)

Allocates a new tensor of the given shape, initialized with the provided data.

Parameters

shape	shape of the tensor to allocate
data	buffer of floats to initialize the tensor with

Returns

• the new tensor

public static abstract TFloat16 tensorOf (Shape shape, Consumer<TFloat16> dataInit)

Allocates a new tensor of the given shape and initialize its data.

Parameters

shape	shape of the tensor to allocate
dataInit	tensor data initializer

Returns

• the new tensor

Throws

TensorFlowException	if the tensor cannot be allocated or initialized

public static abstract TFloat16 tensorOf (NdArray<Float> src)

Allocates a new tensor which is a copy of a given array of floats.

The tensor will have the same shape as the source array and its data will be copied.

Parameters

src	the source array giving the shape and data to the new tensor

Returns

• the new tensor

public static abstract TFloat16 vectorOf (float... values)

Allocates a new tensor for storing a vector of floats.

Parameters

values	floats to store in the new tensor

Returns

• the new tensor