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 module
pad_gpu
Implements GPU-specific tensor padding kernels with constant, reflect, and repeat (edge-fill) strategies.
Functions
-
get_padding_output_shape: Computes the output shape produced by paddinginput_shapewith the before/after amounts given inpaddings. -
pad_constant: Filloutputwith values frominput, and edges padded withconstantbased onpaddings. -
pad_reflect: Filloutputwith values frominput, and edges padded with reflected values from the unpadded region. -
pad_repeat: Filloutputwith values frominput, and edges padded with the boundary values from the unpadded region. -
padded_copy_kernel: -
padded_index_mode_kernel: GPU kernel shared by the reflect and edge (repeat) pad modes.