copy

nvtripy.copy(input: Tensor, device: device) → Tensor

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

Parameters:

• input (Tensor) – [dtype=T1] Tensor that will be copied

• device (device) – The target device.

Returns:

[dtype=T1] A copy of input tensor on target device.

Return type:

Tensor

TYPE CONSTRAINTS:

• T1: bool, float8, float16, bfloat16, float32, int4, int8, int32, int64

Example

input = tp.Tensor([1, 2], device=tp.device("gpu"))
output = tp.copy(input, tp.device("cpu"))

Local Variables

>>> input
tensor([1, 2], dtype=int32, loc=gpu:0, shape=(2,))

>>> output
tensor([1, 2], dtype=int32, loc=cpu:0, shape=(2,))