tf.tensor_scatter_nd_max

Apply a sparse update to a tensor taking the element-wise maximum.

tf.tensor_scatter_nd_max(
    tensor: _atypes.TensorFuzzingAnnotation[TV_TensorScatterMax_T],
    indices: _atypes.TensorFuzzingAnnotation[TV_TensorScatterMax_Tindices],
    updates: _atypes.TensorFuzzingAnnotation[TV_TensorScatterMax_T],
    name=None
) -> _atypes.TensorFuzzingAnnotation[TV_TensorScatterMax_T]

Returns a new tensor copied from tensor whose values are element-wise maximum between tensor and updates according to the indices.

tensor = [0, 0, 0, 0, 0, 0, 0, 0]
indices = [[1], [4], [5]]
updates = [1, -1, 1]
tf.tensor_scatter_nd_max(tensor, indices, updates).numpy()
array([0, 1, 0, 0, 0, 1, 0, 0], dtype=int32)

Refer to tf.tensor_scatter_nd_update for more details.

tensor A Tensor. Tensor to update.
indices A Tensor. Must be one of the following types: int32, int64. Index tensor.
updates A Tensor. Must have the same type as tensor. Updates to scatter into output.
name A name for the operation (optional).

A Tensor. Has the same type as tensor.