IMPORTANT: To view this page as Markdown, append `.md` to the URL (e.g. /get-started.md). For the complete documentation index, see llms.txt.
Skip to main content
For the complete documentation index, see llms.txt. Markdown versions of all pages are available by appending .md to any URL (e.g. /get-started.md).

Mojo function

max_pool_gpu_impl

def max_pool_gpu_impl[dtype: DType, int_type: DType, simd_width: Int](ctx: DeviceContext, input: TileTensor[dtype, Engine=input.Engine, address_space=input.address_space, linear_idx_type=input.linear_idx_type], filter: TileTensor[int_type, Engine=filter.Engine, address_space=filter.address_space, linear_idx_type=filter.linear_idx_type], strides: TileTensor[int_type, Engine=strides.Engine, address_space=strides.address_space, linear_idx_type=strides.linear_idx_type], dilations: TileTensor[int_type, Engine=dilations.Engine, address_space=dilations.address_space, linear_idx_type=dilations.linear_idx_type], paddings: TileTensor[int_type, Engine=paddings.Engine, address_space=paddings.address_space, linear_idx_type=paddings.linear_idx_type], output: TileTensor[dtype, Engine=output.Engine, address_space=output.address_space, linear_idx_type=output.linear_idx_type], ceil_mode: Bool = False)

Computes max pooling on GPU with a fixed channel vector width.

Parameters:

  • ​dtype (DType): Data type of the input and output tensors.
  • ​int_type (DType): Data type of the filter, strides, dilations, and paddings tensors.
  • ​simd_width (Int): Channels pooled per thread. Must divide the channel count.

Args:

Was this page helpful?