megengine.functional.concat¶

concat(inps, axis=0, device=None)[source]¶

Concat some tensors

Parameters

inps (Iterable[Tensor]) – input tensors to concat.
axis (int) – over which dimension the tensors are concatenated. Default: 0
device – which device output will be. Default: None

Return type

Tensor

Returns

output tensor.

Examples

>>> import numpy as np
>>> data1 = Tensor(np.arange(0, 6, dtype=np.float32).reshape((2, 3)))
>>> data2 = Tensor(np.arange(6, 12, dtype=np.float32).reshape((2, 3)))
>>> out = F.concat([data1, data2])
>>> out.numpy()
array([[ 0.,  1.,  2.],
       [ 3.,  4.,  5.],
       [ 6.,  7.,  8.],
       [ 9., 10., 11.]], dtype=float32)

megengine.functional.squeeze

megengine.functional.stack