Join the newly launched Discord
community for real-time discussions, peer support, and direct interaction with the Meridian team!
meridian.backend.split
Stay organized with collections
Save and categorize content based on your preferences.
Splits a tensor value
into a list of sub tensors.
meridian
.
backend
.
split
(
value
,
num_or_size_splits
,
axis
=
0
,
num
=
None
,
name
=
'split'
)
See also tf.unstack
.
If num_or_size_splits
is an int
, then it splits value
along the
dimension axis
into num_or_size_splits
smaller tensors. This requires that value.shape[axis]
is divisible by num_or_size_splits
.
If num_or_size_splits
is a 1-D Tensor (or list), then value
is split into len(num_or_size_splits)
elements. The shape of the i
-th
element has the same size as the value
except along dimension axis
where
the size is num_or_size_splits[i]
.
For example:
>>>
x
=
tf
.
Variable
(
tf
.
random
.
uniform
(
[
5
,
30
]
,
-
1
,
1
))
>>>
>>>
# Split `x` into 3 tensors along dimension 1
>>>
s0
,
s1
,
s2
=
tf
.
split
(
x
,
num_or_size_splits
=
3
,
axis
=
1
)
>>>
tf
.
shape
(
s0
).
numpy
()
array
(
[
5
,
10
]
,
dtype
=
int32
)
>>>
>>>
# Split `x` into 3 tensors with sizes [4, 15, 11] along dimension 1
>>>
split0
,
split1
,
split2
=
tf
.
split
(
x
,
[
4
,
15
,
11
]
,
1
)
>>>
tf
.
shape
(
split0
).
numpy
()
array
(
[
5
,
4
]
,
dtype
=
int32
)
>>>
tf
.
shape
(
split1
).
numpy
()
array
(
[
5
,
15
]
,
dtype
=
int32
)
>>>
tf
.
shape
(
split2
).
numpy
()
array
(
[
5
,
11
]
,
dtype
=
int32
)
Either an int
indicating the number of splits
along axis
or a 1-D integer Tensor
or Python list containing the sizes
of each output tensor along axis
. If an int
, then it must evenly
divide value.shape[axis]
; otherwise the sum of sizes along the split
axis must match that of the value
.
An int
or scalar int32
Tensor
. The dimension along which
to split. Must be in the range [-rank(value), rank(value))
. Defaults to
0.
Optional, an int
, used to specify the number of outputs when it
cannot be inferred from the shape of size_splits
.
A name for the operation (optional).
if num_or_size_splits
is an int
returns a list of num_or_size_splits
Tensor
objects; if num_or_size_splits
is a 1-D
list or 1-D Tensor
returns num_or_size_splits.get_shape[0]
Tensor
objects resulting from splitting value
.
If num
is unspecified and cannot be inferred.
If num_or_size_splits
is a scalar Tensor
.
Except as otherwise noted, the content of this page is licensed under the Creative Commons Attribution 4.0 License
, and code samples are licensed under the Apache 2.0 License
. For details, see the Google Developers Site Policies
. Java is a registered trademark of Oracle and/or its affiliates.
Last updated 2025-09-05 UTC.
[[["Easy to understand","easyToUnderstand","thumb-up"],["Solved my problem","solvedMyProblem","thumb-up"],["Other","otherUp","thumb-up"]],[["Missing the information I need","missingTheInformationINeed","thumb-down"],["Too complicated / too many steps","tooComplicatedTooManySteps","thumb-down"],["Out of date","outOfDate","thumb-down"],["Samples / code issue","samplesCodeIssue","thumb-down"],["Other","otherDown","thumb-down"]],["Last updated 2025-09-05 UTC."],[],[],null,["\u003cbr /\u003e\n\nSplits a tensor `value` into a list of sub tensors. \n\n meridian.backend.split(\n value, num_or_size_splits, axis=0, num=None, name='split'\n )\n\nSee also `tf.unstack`.\n\nIf `num_or_size_splits` is an `int`, then it splits `value` along the\ndimension `axis` into `num_or_size_splits` smaller tensors. This requires that\n`value.shape[axis]` is divisible by `num_or_size_splits`.\n\nIf `num_or_size_splits` is a 1-D Tensor (or list), then `value` is split into\n`len(num_or_size_splits)` elements. The shape of the `i`-th\nelement has the same size as the `value` except along dimension `axis` where\nthe size is `num_or_size_splits[i]`.\n\nFor example: \n\n \u003e\u003e\u003e x = tf.Variable(tf.random.uniform([5, 30], -1, 1))\n \u003e\u003e\u003e\n \u003e\u003e\u003e # Split `x` into 3 tensors along dimension 1\n \u003e\u003e\u003e s0, s1, s2 = tf.split(x, num_or_size_splits=3, axis=1)\n \u003e\u003e\u003e tf.shape(s0).numpy()\n array([ 5, 10], dtype=int32)\n \u003e\u003e\u003e\n \u003e\u003e\u003e # Split `x` into 3 tensors with sizes [4, 15, 11] along dimension 1\n \u003e\u003e\u003e split0, split1, split2 = tf.split(x, [4, 15, 11], 1)\n \u003e\u003e\u003e tf.shape(split0).numpy()\n array([5, 4], dtype=int32)\n \u003e\u003e\u003e tf.shape(split1).numpy()\n array([ 5, 15], dtype=int32)\n \u003e\u003e\u003e tf.shape(split2).numpy()\n array([ 5, 11], dtype=int32)\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n| Args ||\n|----------------------|-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|\n| `value` | The `Tensor` to split. |\n| `num_or_size_splits` | Either an `int` indicating the number of splits along `axis` or a 1-D integer `Tensor` or Python list containing the sizes of each output tensor along `axis`. If an `int`, then it must evenly divide `value.shape[axis]`; otherwise the sum of sizes along the split axis must match that of the `value`. |\n| `axis` | An `int` or scalar `int32` `Tensor`. The dimension along which to split. Must be in the range `[-rank(value), rank(value))`. Defaults to 0. |\n| `num` | Optional, an `int`, used to specify the number of outputs when it cannot be inferred from the shape of `size_splits`. |\n| `name` | A name for the operation (optional). |\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n| Returns ||\n|---|---|\n| if `num_or_size_splits` is an `int` returns a list of `num_or_size_splits` `Tensor` objects; if `num_or_size_splits` is a 1-D list or 1-D `Tensor` returns `num_or_size_splits.get_shape[0]` `Tensor` objects resulting from splitting `value`. ||\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n| Raises ||\n|--------------|-------------------------------------------------|\n| `ValueError` | If `num` is unspecified and cannot be inferred. |\n| `ValueError` | If `num_or_size_splits` is a scalar `Tensor`. |\n\n\u003cbr /\u003e"]]
