For the complete documentation index, see llms.txt. Markdown versions of all pages are available by appending .md to any URL (e.g. /max/get-started.md).

Mojo function

resize_bicubic

def resize_bicubic[dtype: DType, //, target: StringSlice[ImmStaticOrigin]](output: TileTensor[dtype, Storage=output.Storage, linear_idx_type=output.linear_idx_type], input: TileTensor[dtype, Storage=input.Storage, linear_idx_type=input.linear_idx_type], ctx: DeviceContext)

Perform bicubic interpolation.

Parameters:

dtype (DType): Element type of the input and output tensors (inferred).
target (StringSlice[ImmStaticOrigin]): StaticString identifying the execution platform, used to select between the GPU and CPU code paths.

Args:

output (TileTensor[dtype, Storage=output.Storage, linear_idx_type=output.linear_idx_type]): Output tensor with desired dimensions on host or device.
input (TileTensor[dtype, Storage=input.Storage, linear_idx_type=input.linear_idx_type]): Input tensor of shape [B, C, H, W] on host or device.
ctx (DeviceContext): Device context to enqueue GPU kernels on.