Intel® oneAPI Deep Neural Network Developer Guide and Reference
A newer version of this document is available. Customers should click here to go to the newest version.
enum dnnl_graph_op_attr_t
Overview
Attributes of operations. More…
#include <dnnl_graph_types.h>
enum dnnl_graph_op_attr_t
{
dnnl_graph_op_attr_undef = 0,
dnnl_graph_op_attr_alpha = 0x1,
dnnl_graph_op_attr_beta,
dnnl_graph_op_attr_epsilon,
dnnl_graph_op_attr_max,
dnnl_graph_op_attr_min,
dnnl_graph_op_attr_momentum,
dnnl_graph_op_attr_scales = 0x20,
dnnl_graph_op_attr_axis = 0x30,
dnnl_graph_op_attr_begin_norm_axis,
dnnl_graph_op_attr_groups,
dnnl_graph_op_attr_axes = 0x40,
dnnl_graph_op_attr_dilations,
dnnl_graph_op_attr_dst_shape,
dnnl_graph_op_attr_kernel,
dnnl_graph_op_attr_order,
dnnl_graph_op_attr_output_padding,
dnnl_graph_op_attr_pads_begin,
dnnl_graph_op_attr_pads_end,
dnnl_graph_op_attr_shape,
dnnl_graph_op_attr_sizes,
dnnl_graph_op_attr_src_shape,
dnnl_graph_op_attr_strides,
dnnl_graph_op_attr_weights_shape,
dnnl_graph_op_attr_zps,
dnnl_graph_op_attr_exclude_pad = 0x60,
dnnl_graph_op_attr_keep_dims,
dnnl_graph_op_attr_keep_stats,
dnnl_graph_op_attr_per_channel_broadcast,
dnnl_graph_op_attr_special_zero,
dnnl_graph_op_attr_transpose_a,
dnnl_graph_op_attr_transpose_b,
dnnl_graph_op_attr_use_affine,
dnnl_graph_op_attr_use_dst,
dnnl_graph_op_attr_auto_broadcast = 0x80,
dnnl_graph_op_attr_auto_pad,
dnnl_graph_op_attr_coordinate_transformation_mode,
dnnl_graph_op_attr_data_format,
dnnl_graph_op_attr_mode,
dnnl_graph_op_attr_qtype,
dnnl_graph_op_attr_rounding_type,
dnnl_graph_op_attr_weights_format,
};Detailed Documentation
Attributes of operations.
Enum Values
dnnl_graph_op_attr_undefUndefined op attribute.
dnnl_graph_op_attr_alphaSpecifies an alpha attribute to an op.
dnnl_graph_op_attr_betaSpecifies an beta attribute to an op.
dnnl_graph_op_attr_epsilonSpecifies an epsilon attribute to an op.
dnnl_graph_op_attr_maxSpecifies a max attribute to an op.
dnnl_graph_op_attr_minSpecifies a min attribute to an op.
dnnl_graph_op_attr_momentumSpecifies a momentum attribute to an op.
dnnl_graph_op_attr_scalesSpecifies a scales attribute to an op.
dnnl_graph_op_attr_axisSpecifies an axis attribute to an op.
dnnl_graph_op_attr_begin_norm_axisSpecifies a begin_norm_axis attribute to an op.
dnnl_graph_op_attr_groupsSpecifies a groups attribute to an op.
dnnl_graph_op_attr_axesSpecifies an axes attribute to an op.
dnnl_graph_op_attr_dilationsSpecifies a dilations attribute to an op.
dnnl_graph_op_attr_dst_shapeSpecifies an dst_shape attribute to an op.
dnnl_graph_op_attr_kernelSpecifies a kernel attribute to an op.
dnnl_graph_op_attr_orderSpecifies an order attribute to an op.
dnnl_graph_op_attr_output_paddingSpecifies an output_padding attribute to an op.
dnnl_graph_op_attr_pads_beginSpecifies a pads_begin attribute to an op.
dnnl_graph_op_attr_pads_endSpecifies a pads_end attribute to an op.
dnnl_graph_op_attr_shapeSpecifies a shape attribute to an op.
dnnl_graph_op_attr_sizesSpecifies a sizes attribute to an op.
dnnl_graph_op_attr_src_shapeSpecifies a input_shape attribute to an op.
dnnl_graph_op_attr_stridesSpecifies a strides attribute to an op.
dnnl_graph_op_attr_weights_shapeSpecifies a weight_shape attribute to an op.
dnnl_graph_op_attr_zpsSpecifies a zps attribute to an op.
dnnl_graph_op_attr_exclude_padSpecifies an exclude_pad attribute to an op.
dnnl_graph_op_attr_keep_dimsSpecifies a keep_dims attribute to an op.
dnnl_graph_op_attr_keep_statsSpecifies a keep_stats attribute to an op.
dnnl_graph_op_attr_per_channel_broadcastSpecifies a per_channel_broadcast attribute to an op.
dnnl_graph_op_attr_special_zeroSpecifies a special_zero attribute to an op.
dnnl_graph_op_attr_transpose_aSpecifies a transpose_a attribute to an op.
dnnl_graph_op_attr_transpose_bSpecifies a transpose_b attribute to an op.
dnnl_graph_op_attr_use_affineSpecifies an use_affine attribute to an op.
dnnl_graph_op_attr_use_dstSpecifies an use_dst attribute to an op.
dnnl_graph_op_attr_auto_broadcastSpecifies an auto_broadcast attribute to an op.
The value can be “none” or “numpy”.
dnnl_graph_op_attr_auto_padSpecifies an auto_pad attribute to an op.
The value can be “none”, “same_upper”, “same_lower”, or “valid”.
dnnl_graph_op_attr_coordinate_transformation_modeSpecifies an coordinate_transformation_mode attribute to an op.
The value can be “half_pixel” or “align_corners”. The attribute is defined for Interpolate operations.
dnnl_graph_op_attr_data_formatSpecifies a data_format of an op. The value can be “NCX” or “NXC”.
dnnl_graph_op_attr_modeSpecifies a mode attribute of an op.
The value can be “nearest”, “linear”, “bilinear”, or “trilinear”. The attribute is defined for Interpolate operations.
dnnl_graph_op_attr_qtypeSpecifies a qtype attribute to an op.
The value can be “per_channel” or “per_tensor”. The attribute is defined for quantization operations.
dnnl_graph_op_attr_rounding_typeSpecifies a rounding_type attribute to an op.
The value can be “ceil” or “floor”.
dnnl_graph_op_attr_weights_formatSpecifies a weights_format of an op.
The value can be “OIX”, “XIO”, “IOX”, or “XOI”. Different operations may support different values.