tf.raw_ops.ResourceScatterMax  |  TensorFlow v2.16.1 (original) (raw)

tf.raw_ops.ResourceScatterMax

Stay organized with collections Save and categorize content based on your preferences.

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

View aliases

Compat aliases for migration

SeeMigration guide for more details.

tf.compat.v1.raw_ops.ResourceScatterMax

tf.raw_ops.ResourceScatterMax(
    resource, indices, updates, name=None
)

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 = [].