concatenate¶

nvtripy.concatenate(tensors: Sequence[Tensor], dim: int) → Tensor[source]¶

Returns a copy of the input tensor on the target device.

Parameters:

tensors (Sequence[Tensor]) – [dtype=T1] Sequence of tensors of the same type and having the same shape except in the concatenated dimension.
dim (int) – the dimension over which the tensors are concatenated.

Returns:

[dtype=T1] Concatenated tensor with shape along dim axis equal to sum of dimensions at dim axis for all inputs.

Return type:

Tensor

TYPE CONSTRAINTS:

T1: bool, float16, bfloat16, float32, int4, int8, int32, int64

Example

a = tp.iota((1, 2), dtype=tp.float32)
b = tp.iota((2, 2), dtype=tp.float32)

output = tp.concatenate([a, b], dim=0)

Local Variables¶

>>> a
tensor(
    [[0.0000, 0.0000]], 
    dtype=float32, loc=gpu:0, shape=(1, 2))

>>> b
tensor(
    [[0.0000, 0.0000],
     [1.0000, 1.0000]], 
    dtype=float32, loc=gpu:0, shape=(2, 2))

>>> output
tensor(
    [[0.0000, 0.0000],
     [0.0000, 0.0000],
     [1.0000, 1.0000]], 
    dtype=float32, loc=gpu:0, shape=(3, 2))