vsplit¶

paddle. vsplit ( x: Tensor, num_or_indices: int | Sequence[int], name: str | None = None ) → list[Tensor] [source]

vsplit Full name Vertical Split, splits the input Tensor into multiple sub-Tensors along the vertical axis, which is equivalent to paddle.tensor_split with axis=0.

1. When the number of Tensor dimensions is equal to 2:

2. When the number of Tensor dimensions is greater than 2:

Note

Make sure that the number of Tensor dimensions transformed using paddle.vsplit must be not less than 2.

Parameters

• x (Tensor) – A Tensor whose dimension must be greater than 1. The data type is bool, bfloat16, float16, float32, float64, uint8, int32 or int64.

• num_or_indices (int|list|tuple) – If num_or_indices is an int n, x is split into n sections. If num_or_indices is a list or tuple of integer indices, x is split at each of the indices.

• name (str, optional) – The default value is None. Normally there is no need for user to set this property. For more information, please refer to Name .

Returns

list[Tensor], The list of segmented Tensors.

Examples

>>> import paddle

>>> # x is a Tensor of shape [8, 6, 7]
>>> x = paddle.rand([8, 6, 7])
>>> out0, out1 = paddle.vsplit(x, num_or_indices=2)
>>> print(out0.shape)
[4, 6, 7]
>>> print(out1.shape)
[4, 6, 7]

>>> out0, out1, out2 = paddle.vsplit(x, num_or_indices=[1, 4])
>>> print(out0.shape)
[1, 6, 7]
>>> print(out1.shape)
[3, 6, 7]
>>> print(out2.shape)
[4, 6, 7]