resize¶

nvtripy.resize(*args, **kwargs) → Tensor[source]¶

This function has multiple overloads:

nvtripy.resize(input: nvtripy.Tensor, mode: str, output_shape: Sequence[int | nvtripy.DimensionSize], align_corners: bool = False) -> nvtripy.Tensor

Resizes the input tensor.

Parameters:

input – The input tensor.
mode – The resize operation’s algorithm. Must be one of: [“cubic”, linear”, “nearest”].
output_shape – The output shape of the resize operation.
align_corners – If set to True, the input and output tensors are aligned by the center points of their corner pixels, preserving the values at the corner pixels. If set to False, the input and output tensors are aligned by the corner points of their corner pixels. Only in effect when mode is "cubic" or "linear".

Returns:

The output tensor after the resize operation.

Return type:

Tensor

Example

input = tp.reshape(tp.arange(16, dtype=tp.float32), (1, 1, 4, 4))
output = tp.resize(input, "nearest", output_shape=(1, 1, 8, 8))

Local Variables¶

>>> input
tensor(
    [[[[0.0000, 1.0000, 2.0000, 3.0000],
       [4.0000, 5.0000, 6.0000, 7.0000],
       [8.0000, 9.0000, 10.0000, 11.0000],
       [12.0000, 13.0000, 14.0000, 15.0000]]]], 
    dtype=float32, loc=gpu:0, shape=(1, 1, 4, 4))

>>> output
tensor(
    [[[[0.0000, 0.0000, 1.0000, ..., 2.0000, 3.0000, 3.0000],
       [0.0000, 0.0000, 1.0000, ..., 2.0000, 3.0000, 3.0000],
       [4.0000, 4.0000, 5.0000, ..., 6.0000, 7.0000, 7.0000],
       ...,
       [8.0000, 8.0000, 9.0000, ..., 10.0000, 11.0000, 11.0000],
       [12.0000, 12.0000, 13.0000, ..., 14.0000, 15.0000, 15.0000],
       [12.0000, 12.0000, 13.0000, ..., 14.0000, 15.0000, 15.0000]]]], 
    dtype=float32, loc=gpu:0, shape=(1, 1, 8, 8))

nvtripy.resize(input: nvtripy.Tensor, mode: str, scales: Sequence[numbers.Number], align_corners: bool = False) -> nvtripy.Tensor