ML Community Day is November 9! Join us for updates from TensorFlow, JAX, and more Learn more


Concats input tensor across all dimensions.

An op which merges slices the input tensor based on the given num_splits attribute, strips paddings optionally, and returns the merged tensor without paddings.

This op may be generated via the TPU bridge.

For example, with input tensor:

[[0, 1],
 [4, 5]]
[[2, 3],
 [6, 7]]
[[8, 9],
 [12, 13]]
[[10, 11],
 [14, 15]]


[2, 2]

and paddings:

[1, 1]

the expected outputs is:

[[0, 1, 2],
 [4, 5, 6],
 [8, 9, 10]]

inputs A list of at least 1 Tensor objects with the same type. Input tensor slices in row-major order to merge across all dimensions. All inputs must have the same shape. } out_arg { name: "output" description: <
num_concats A list of ints. Number of ways to merge per dimension.
paddings An optional list of ints. Defaults to []. Optional list of right paddings per dimension to strip from the final merged tensor. These paddings must not exceed the dimension size of the merged result prior to stripping paddings.
name A name for the operation (optional).

A Tensor. Has the same type as inputs.