T
- data type for output()
output@Operator public final class Gather<T> extends PrimitiveOp implements Operand<T>
`indices` must be an integer tensor of any dimension (usually 0-D or 1-D). Produces an output tensor with shape `params.shape[:axis] + indices.shape + params.shape[axis + 1:]` where:
# Scalar indices (output is rank(params) - 1).
output[a_0, ..., a_n, b_0, ..., b_n] =
params[a_0, ..., a_n, indices, b_0, ..., b_n]
# Vector indices (output is rank(params)).
output[a_0, ..., a_n, i, b_0, ..., b_n] =
params[a_0, ..., a_n, indices[i], b_0, ..., b_n]
# Higher rank indices (output is rank(params) + rank(indices) - 1).
output[a_0, ..., a_n, i, ..., j, b_0, ... b_n] =
params[a_0, ..., a_n, indices[i, ..., j], b_0, ..., b_n]
Note that on CPU, if an out of bound index is found, an error is returned. On GPU, if an out of bound index is found, a 0 is stored in the corresponding output value.
See also `tf.batch_gather` and `tf.gather_nd`.
Modifier and Type | Class and Description |
---|---|
static class |
Gather.Options
Optional attributes for
Gather |
operation
Modifier and Type | Method and Description |
---|---|
Output<T> |
asOutput()
Returns the symbolic handle of a tensor.
|
static Gather.Options |
batchDims(Long batchDims) |
static <T,U extends Number,V extends Number> |
create(Scope scope,
Operand<T> params,
Operand<U> indices,
Operand<V> axis,
Gather.Options... options)
Factory method to create a class wrapping a new Gather operation.
|
Output<T> |
output()
Values from `params` gathered from indices given by `indices`, with
shape `params.shape[:axis] + indices.shape + params.shape[axis + 1:]`.
|
equals, hashCode, op, toString
public static <T,U extends Number,V extends Number> Gather<T> create(Scope scope, Operand<T> params, Operand<U> indices, Operand<V> axis, Gather.Options... options)
scope
- current scopeparams
- The tensor from which to gather values. Must be at least rank
`axis + 1`.indices
- Index tensor. Must be in range `[0, params.shape[axis])`.axis
- The axis in `params` to gather `indices` from. Defaults to the first
dimension. Supports negative indexes.options
- carries optional attributes valuespublic static Gather.Options batchDims(Long batchDims)
batchDims
- public Output<T> output()
public Output<T> asOutput()
Operand
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.
asOutput
in interface Operand<T>
OperationBuilder.addInput(Output)
Copyright © 2022. All rights reserved.