SparseTensor

Represents a sparse tensor.

TensorFlow represents a sparse tensor as three separate dense tensors: indices, values, and denseShape. In Scala, the three tensors are collected into a SparseTensor class for ease of use. If you have separate indices, values, and denseShape tensors, wrap them in a SparseTensor object before passing to the relevant sparse tensor manipulation.

Concretely, the sparse tensor SparseTensor(indices, values, denseShape) comprises the following components, where N and rank are the number of values and number of dimensions in the SparseTensor, respectively:

indices: Two-dimensional INT64 tensor with shape [N, rank], which specifies the indices of the elements in the sparse tensor that have nonzero values (elements are zero-indexed). For example, indices = 3], [2, 4 specifies that the elements with indexes [1, 3] and [2, 4] have nonzero values.
values: One-dimensional tensor of any type, with shape [N], which supplies the values for each element in indices. For example, given indices = 3], [2, 4, the parameter values = [18, 3.6] specifies that element [1, 3] of the sparse tensor has a value of 18, and element [2, 4] of the tensor has a value of 3.6.
denseShape: One-dimensional INT64 tensor with shape [rank], which specifies the dense shape of the sparse tensor. For example, denseShape = [3, 6] specifies a two-dimensional 3x6 tensor, denseShape = [2, 3, 4] specifies a three-dimensional 2x3x4 tensor, and denseShape = [9] specifies a one-dimensional tensor with 9 elements.

The corresponding dense tensor, dense, satisfies:

dense.shape == denseShape
dense(indices(i)) = values(i) // Using a somewhat loose notation with respect to indexing.

IMPORTANT NOTE: By convention, indices should be sorted in row-major order (or equivalently lexicographic order on indices(i)). This is not enforced when SparseTensor objects are constructed, but most ops assume correct ordering. If the ordering of sparse tensor st is wrong, a fixed version can be obtained by calling sparseReorder(st).

For example, the sparse tensor SparseTensor(indices = 0], [1, 2, values = [1, 2], denseShape = [3, 4]), represents the dense tensor 0, 0, 0], [0, 0, 2, 0], [0, 0, 0, 0.

indices: Two-dimensional INT32 tensor with shape [N, rank].
values: One-dimensional tensor with shape [N].
denseShape: One-dimensional INT32 tensor with shape [rank].

Linear Supertypes

Serializable, Serializable, Product, Equals, TensorLike[D], AnyRef, Any

Instance Constructors

new SparseTensor(indices: Tensor[types.INT64], values: Tensor[D], denseShape: Tensor[types.INT64])

indices
Two-dimensional INT32 tensor with shape [N, rank].
values
One-dimensional tensor with shape [N].
denseShape
One-dimensional INT32 tensor with shape [rank].

Value Members

final def !=(arg0: Any): Boolean

Definition Classes
AnyRef → Any
final def ##(): Int

Definition Classes
AnyRef → Any
final def ==(arg0: Any): Boolean

Definition Classes
AnyRef → Any
final def asInstanceOf[T0]: T0

Definition Classes
Any
def clone(): AnyRef

Attributes
protected[java.lang]
Definition Classes
AnyRef
Annotations
@throws( ... )
val dataType: D

Data type of this sparse op output.
Data type of this sparse op output.

Definition Classes
SparseTensor → TensorLike
val denseShape: Tensor[types.INT64]

One-dimensional INT32 tensor with shape [rank].
val device: String

Device on which this sparse op output will be placed.
Device on which this sparse op output will be placed.

Definition Classes
SparseTensor → TensorLike
final def eq(arg0: AnyRef): Boolean

Definition Classes
AnyRef
def finalize(): Unit

Attributes
protected[java.lang]
Definition Classes
AnyRef
Annotations
@throws( classOf[java.lang.Throwable] )
final def getClass(): Class[_]

Definition Classes
AnyRef → Any
val indices: Tensor[types.INT64]

Two-dimensional INT32 tensor with shape [N, rank].
final def isInstanceOf[T0]: Boolean

Definition Classes
Any
final def ne(arg0: AnyRef): Boolean

Definition Classes
AnyRef
final def notify(): Unit

Definition Classes
AnyRef
final def notifyAll(): Unit

Definition Classes
AnyRef
val shape: core.Shape

Shape of this sparse tensor.
Shape of this sparse tensor.

Definition Classes
SparseTensor → TensorLike
final def synchronized[T0](arg0: ⇒ T0): T0

Definition Classes
AnyRef
def toString(): String

Definition Classes
SparseTensor → AnyRef → Any
def toTensor[DD >: D <: types.DataType](defaultValue: Tensor[DD] = Tensor.zeros(dataType, Shape()), validateIndices: Boolean = true): Tensor[DD]

Converts this sparse tensor to a dense tensor.
Converts this sparse tensor to a dense tensor.
The constructed op builds a tensor dense with shape input.denseShape, such that:
```
// If input.indices is scalar:
dense(i) ==> (i == input.indices ? input.values : defaultValue)

// If input.indices is a vector, then for each i:
dense(input.indices(i)) ==> input.values(i)

// If input.indices is an n by d matrix, then for each i in [0, n):
dense(input.indices(i)(0), ..., input.indices(i)(d-1)) ==> input.values(i)
```
All other values in dense are set to defaultValue. If input.values is a scalar, then all sparse indices are set to that single value.
input.indices should be sorted in lexicographic order and they must not contain any repeats. If validateIndices is true, then these properties are checked during execution.
defaultValue
Scalar tensor with the same data type as input.values, containing the value set for indices that are not specified in input.indices.
validateIndices
If true, the indices in input.indices are checked to make sure that they are sorted in lexicographic order and that there are no repeats.
returns
Result as a new tensor, with the same data type as input.values and shape input.denseShape.
def toTensor: Tensor[D]

Returns the Tensor that this TensorLike object represents.
Returns the Tensor that this TensorLike object represents.

Definition Classes
SparseTensor → TensorLike
def toTensorIndexedSlices: TensorIndexedSlices[D]

Returns an TensorIndexedSlices that has the same value as this TensorLike.
Returns an TensorIndexedSlices that has the same value as this TensorLike.
returns
TensorIndexedSlices that has the same value as this TensorLike.

Definition Classes
SparseTensor → TensorLike
Annotations
@throws( ... )
val values: Tensor[D]

One-dimensional tensor with shape [N].
final def wait(): Unit

Definition Classes
AnyRef
Annotations
@throws( ... )
final def wait(arg0: Long, arg1: Int): Unit

Definition Classes
AnyRef
Annotations
@throws( ... )
final def wait(arg0: Long): Unit

Definition Classes
AnyRef
Annotations
@throws( ... )

Related Doc: package tensors

final case class SparseTensor[+D <: types.DataType](indices: Tensor[types.INT64], values: Tensor[D], denseShape: Tensor[types.INT64]) extends TensorLike[D] with Product with Serializable

Instance Constructors

new SparseTensor(indices: Tensor[types.INT64], values: Tensor[D], denseShape: Tensor[types.INT64])

Value Members

final def !=(arg0: Any): Boolean

final def ##(): Int

final def ==(arg0: Any): Boolean

final def asInstanceOf[T0]: T0

def clone(): AnyRef

val dataType: D

val denseShape: Tensor[types.INT64]

val device: String

final def eq(arg0: AnyRef): Boolean

def finalize(): Unit

final def getClass(): Class[_]

val indices: Tensor[types.INT64]

final def isInstanceOf[T0]: Boolean

final def ne(arg0: AnyRef): Boolean

final def notify(): Unit

final def notifyAll(): Unit

val shape: core.Shape

final def synchronized[T0](arg0: ⇒ T0): T0

def toString(): String

def toTensor[DD >: D <: types.DataType](defaultValue: Tensor[DD] = Tensor.zeros(dataType, Shape()), validateIndices: Boolean = true): Tensor[DD]

def toTensor: Tensor[D]

def toTensorIndexedSlices: TensorIndexedSlices[D]

val values: Tensor[D]

final def wait(): Unit

final def wait(arg0: Long, arg1: Int): Unit

final def wait(arg0: Long): Unit

Inherited from Serializable

Inherited from Serializable

Inherited from Product

Inherited from Equals

Inherited from TensorLike[D]

Inherited from AnyRef

Inherited from Any

Ungrouped