Partitions `data` into `num_partitions` tensors using indices from `partitions`.
For each index tuple `js` of size `partitions.ndim`, the slice `data[js, ...]` becomes part of `outputs[partitions[js]]`. The slices with `partitions[js] = i` are placed in `outputs[i]` in lexicographic order of `js`, and the first dimension of `outputs[i]` is the number of entries in `partitions` equal to `i`. In detail,
outputs[i].shape = [sum(partitions == i)] + data.shape[partitions.ndim:]
outputs[i] = pack([data[js, ...] for js if partitions[js] == i])
`data.shape` must start with `partitions.shape`.
For example:
# Scalar partitions.
partitions = 1
num_partitions = 2
data = [10, 20]
outputs[0] = [] # Empty with shape [0, 2]
outputs[1] = [[10, 20]]
# Vector partitions.
partitions = [0, 0, 1, 1, 0]
num_partitions = 2
data = [10, 20, 30, 40, 50]
outputs[0] = [10, 20, 50]
outputs[1] = [30, 40]
See `dynamic_stitch` for an example on how to merge partitions back.
![](https://www.tensorflow.org/images/DynamicPartition.png?authuser=19)
Public Methods
static <T> DynamicPartition <T> | |
Iterator< Operand <T>> |
iterator
()
|
List< Output <T>> |
outputs
()
|
Inherited Methods
Public Methods
public static DynamicPartition <T> create ( Scope scope, Operand <T> data, Operand <Integer> partitions, Long numPartitions)
Factory method to create a class wrapping a new DynamicPartition operation.
Parameters
scope | current scope |
---|---|
partitions | Any shape. Indices in the range `[0, num_partitions)`. |
numPartitions | The number of partitions to output. |
Returns
- a new instance of DynamicPartition