_vx_nn_deconvolution_params_t Struct Reference

_vx_nn_deconvolution_params_t Struct Reference#

MIVisionX: _vx_nn_deconvolution_params_t Struct Reference
Data Fields
_vx_nn_deconvolution_params_t Struct Reference
Extension: Deep Convolutional Networks API

Input parameters for a deconvolution operation. More...

#include <vx_khr_nn.h>

Data Fields

vx_size padding_x
 Number of elements subtracted at each side in the x dimension of the output.
 
vx_size padding_y
 Number of elements subtracted at each side in the y dimension of the output.
 
vx_enum overflow_policy
 A VX_TYPE_ENUM of the vx_convert_policy_e enumeration.
 
vx_enum rounding_policy
 A VX_TYPE_ENUM of the vx_round_policy_e enumeration.
 
vx_size a_x
 user-specified quantity used to distinguish between the $upscale_x$ different possible output sizes.
 
vx_size a_y
 user-specified quantity used to distinguish between the $upscale_y$ different possible output sizes.
 

Detailed Description

Input parameters for a deconvolution operation.

The documentation for this struct was generated from the following file:
  • /home/docs/checkouts/readthedocs.org/user_builds/advanced-micro-devices-mivisionx/checkouts/docs-5.7.0/amd_openvx/openvx/include/VX/vx_khr_nn.h