Split the data from the input value into TensorArray elements.
Assuming that `lengths` takes on values
(n0, n1, ..., n(T-1))
(n0 + n1 + ... + n(T-1) x d0 x d1 x ...)
this splits values into a TensorArray with T tensors.
TensorArray index t will be the subtensor of values with starting position
(n0 + n1 + ... + n(t-1), 0, 0, ...)
nt x d0 x d1 x ...
Public Methods
Output<Float> |
asOutput()
Returns the symbolic handle of a tensor.
|
static <T> TensorArraySplit | |
Output<Float> |
flowOut()
A float scalar that enforces proper chaining of operations.
|
Inherited Methods
Public Methods
public Output<Float> 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.
public static TensorArraySplit create (Scope scope, Operand<?> handle, Operand<T> value, Operand<Long> lengths, Operand<Float> flowIn)
Factory method to create a class wrapping a new TensorArraySplit operation.
Parameters
scope | current scope |
---|---|
handle | The handle to a TensorArray. |
value | The concatenated tensor to write to the TensorArray. |
lengths | The vector of lengths, how to split the rows of value into the TensorArray. |
flowIn | A float scalar that enforces proper chaining of operations. |
Returns
- a new instance of TensorArraySplit