Mojo function

pool_shape_impl

pool_shape_impl[input_type: DType, filter_type: DType, strides_type: DType, dilations_type: DType, paddings_type: DType, single_thread_blocking_override: Bool, ceil_mode: Bool](input_buf: TileTensor[input_type, LayoutType, origin, address_space=address_space, linear_idx_type=linear_idx_type, element_shape_types=element_shape_types], filter_buf: TileTensor[filter_type, LayoutType, origin, address_space=address_space, linear_idx_type=linear_idx_type, element_shape_types=element_shape_types], strides_buf: TileTensor[strides_type, LayoutType, origin, address_space=address_space, linear_idx_type=linear_idx_type, element_shape_types=element_shape_types], dilations_buf: TileTensor[dilations_type, LayoutType, origin, address_space=address_space, linear_idx_type=linear_idx_type, element_shape_types=element_shape_types], paddings_buf: TileTensor[paddings_type, LayoutType, origin, address_space=address_space, linear_idx_type=linear_idx_type, element_shape_types=element_shape_types]) -> IndexList[TileTensor[input_type, LayoutType, origin, address_space=address_space, linear_idx_type=linear_idx_type, element_shape_types=element_shape_types].rank]

Compute the output shape of a pooling operation, and assert the inputs are compatible. Works for 2D pool operations only in the NHWC format.

Parameters:

input_type (DType): Type of the input tensor.
filter_type (DType): Type of the filter tensor.
strides_type (DType): Type of the strides tensor.
dilations_type (DType): Type of the dilations tensor.
paddings_type (DType): Type of the paddings tensor.
single_thread_blocking_override (Bool): If True, then the operation is run synchronously using a single thread.
ceil_mode (Bool): Define rounding mode for shape calculation.

Args:

input_buf (TileTensor): The input tensor.
filter_buf (TileTensor): The filter size buffer.
strides_buf (TileTensor): The strides size buffer.
dilations_buf (TileTensor): The dilations size buffer.
paddings_buf (TileTensor): The paddings size buffer.

Returns:

IndexList: The output shape.