Warning: This API is deprecated and will be removed in a future version of TensorFlow after the replacement is stable.

ResourceScatterMax

Stay organized with collections Save and categorize content based on your preferences.
public final class ResourceScatterMax

Reduces sparse updates into the variable referenced by `resource` using the `max` operation.

This operation computes

# Scalar indices ref[indices, ...] = max(ref[indices, ...], updates[...])

# Vector indices (for each i) ref[indices[i], ...] = max(ref[indices[i], ...], updates[i, ...])

# High rank indices (for each i, ..., j) ref[indices[i, ..., j], ...] = max(ref[indices[i, ..., j], ...], updates[i, ..., j, ...])

Duplicate entries are handled correctly: if multiple `indices` reference the same location, their contributions are combined.

Requires `updates.shape = indices.shape + ref.shape[1:]` or `updates.shape = []`.

Public Methods

static <T extends Number, U> ResourceScatterMax
create(Scope scope, Operand<?> resource, Operand<T> indices, Operand<U> updates)
Factory method to create a class wrapping a new ResourceScatterMax operation.

Inherited Methods

Public Methods

public static ResourceScatterMax create (Scope scope, Operand<?> resource, Operand<T> indices, Operand<U> updates)

Factory method to create a class wrapping a new ResourceScatterMax operation.

Parameters
scope current scope
resource Should be from a `Variable` node.
indices A tensor of indices into the first dimension of `ref`.
updates A tensor of updated values to add to `ref`.
Returns
  • a new instance of ResourceScatterMax