tf.math.segment_max | TensorFlow v2.16.1 (original) (raw)

tf.math.segment_max

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

Computes the maximum along segments of a tensor.

View aliases

Compat aliases for migration

tf.compat.v1.math.segment_max, tf.compat.v1.segment_max

tf.math.segment_max(
    data: Annotated[Any, tf.raw_ops.Any],
    segment_ids: Annotated[Any, tf.raw_ops.Any],
    name=None
) -> Annotated[Any, tf.raw_ops.Any]

Readthe section on segmentationfor an explanation of segments.

Computes a tensor such that \(output_i = \max_j(data_j)\) where max is over j such that segment_ids[j] == i.

If the max is empty for a given segment ID i, output[i] = 0.

For example:

c = tf.constant([[1,2,3,4], [4, 3, 2, 1], [5,6,7,8]]) tf.math.segment_max(c, tf.constant([0, 0, 1])).numpy() array([[4, 3, 3, 4], [5, 6, 7, 8]], dtype=int32)

Args
data	A Tensor. Must be one of the following types: float32, float64, int32, uint8, int16, int8, int64, bfloat16, uint16, half, uint32, uint64.
segment_ids	A Tensor. Must be one of the following types: int32, int64. A 1-D tensor whose size is equal to the size of data's first dimension. Values should be sorted and can be repeated.
name	A name for the operation (optional).

Returns
A Tensor. Has the same type as data.