.. index:: pair: struct; dnnl::convolution_backward_weights::desc .. _doxid-structdnnl_1_1convolution__backward__weights_1_1desc: struct dnnl::convolution_backward_weights::desc =============================================== .. toctree:: :hidden: Overview ~~~~~~~~ Descriptor for a convolution weights gradient primitive. :ref:`More...` .. ref-code-block:: cpp :class: doxyrest-overview-code-block #include struct desc { // fields :ref:`dnnl_convolution_desc_t` :target:`data`; // construction :ref:`desc`( :ref:`algorithm` aalgorithm, const :ref:`memory::desc`& src_desc, const :ref:`memory::desc`& diff_weights_desc, const :ref:`memory::desc`& diff_bias_desc, const :ref:`memory::desc`& diff_dst_desc, const :ref:`memory::dims`& strides, const :ref:`memory::dims`& padding_l, const :ref:`memory::dims`& padding_r ); :ref:`desc`( :ref:`algorithm` aalgorithm, const :ref:`memory::desc`& src_desc, const :ref:`memory::desc`& diff_weights_desc, const :ref:`memory::desc`& diff_dst_desc, const :ref:`memory::dims`& strides, const :ref:`memory::dims`& padding_l, const :ref:`memory::dims`& padding_r ); :ref:`desc`( :ref:`algorithm` aalgorithm, const :ref:`memory::desc`& src_desc, const :ref:`memory::desc`& diff_weights_desc, const :ref:`memory::desc`& diff_bias_desc, const :ref:`memory::desc`& diff_dst_desc, const :ref:`memory::dims`& strides, const :ref:`memory::dims`& dilates, const :ref:`memory::dims`& padding_l, const :ref:`memory::dims`& padding_r ); :ref:`desc`( :ref:`algorithm` aalgorithm, const :ref:`memory::desc`& src_desc, const :ref:`memory::desc`& diff_weights_desc, const :ref:`memory::desc`& diff_dst_desc, const :ref:`memory::dims`& strides, const :ref:`memory::dims`& dilates, const :ref:`memory::dims`& padding_l, const :ref:`memory::dims`& padding_r ); }; .. _details-structdnnl_1_1convolution__backward__weights_1_1desc: Detailed Documentation ~~~~~~~~~~~~~~~~~~~~~~ Descriptor for a convolution weights gradient primitive. Construction ------------ .. index:: pair: function; desc .. _doxid-structdnnl_1_1convolution__backward__weights_1_1desc_1ad72317d76925fa15f96c9f2459bea055: .. ref-code-block:: cpp :class: doxyrest-title-code-block desc( :ref:`algorithm` aalgorithm, const :ref:`memory::desc`& src_desc, const :ref:`memory::desc`& diff_weights_desc, const :ref:`memory::desc`& diff_bias_desc, const :ref:`memory::desc`& diff_dst_desc, const :ref:`memory::dims`& strides, const :ref:`memory::dims`& padding_l, const :ref:`memory::dims`& padding_r ) Constructs a descriptor for a convolution weights gradient primitive with bias. .. note:: All the memory descriptors may be initialized with the :ref:`dnnl::memory::format_tag::any ` value of ``format_tag``. Arrays ``strides``, ``padding_l``, and ``padding_r`` contain values for spatial dimensions only and hence must have the same number of elements as there are spatial dimensions. The order of values is the same as in the tensor: depth (for 3D tensors), height (for 3D and 2D tensors), and width. .. rubric:: Parameters: .. list-table:: :widths: 20 80 * - aalgorithm - Convolution algorithm. Possible values are :ref:`dnnl::algorithm::convolution_direct `, :ref:`dnnl::algorithm::convolution_winograd `, and :ref:`dnnl::algorithm::convolution_auto `. * - src_desc - Source memory descriptor. * - diff_weights_desc - Diff weights memory descriptor. * - diff_bias_desc - Diff bias memory descriptor. Passing zero memory descriptor disables the bias term. * - diff_dst_desc - Diff destination memory descriptor. * - strides - Strides for each spatial dimension. * - padding_l - Vector of padding values for low indices for each spatial dimension ``([[front,] top,] left)``. * - padding_r - Vector of padding values for high indices for each spatial dimension ``([[back,] bottom,] right)``. .. index:: pair: function; desc .. _doxid-structdnnl_1_1convolution__backward__weights_1_1desc_1a22f11db6486e320a2f478ac576600959: .. ref-code-block:: cpp :class: doxyrest-title-code-block desc( :ref:`algorithm` aalgorithm, const :ref:`memory::desc`& src_desc, const :ref:`memory::desc`& diff_weights_desc, const :ref:`memory::desc`& diff_dst_desc, const :ref:`memory::dims`& strides, const :ref:`memory::dims`& padding_l, const :ref:`memory::dims`& padding_r ) Constructs a descriptor for a convolution weights gradient primitive without bias. .. note:: All the memory descriptors may be initialized with the :ref:`dnnl::memory::format_tag::any ` value of ``format_tag``. Arrays ``strides``, ``padding_l``, and ``padding_r`` contain values for spatial dimensions only and hence must have the same number of elements as there are spatial dimensions. The order of values is the same as in the tensor: depth (for 3D tensors), height (for 3D and 2D tensors), and width. .. rubric:: Parameters: .. list-table:: :widths: 20 80 * - aalgorithm - Convolution algorithm. Possible values are :ref:`dnnl::algorithm::convolution_direct `, :ref:`dnnl::algorithm::convolution_winograd `, and :ref:`dnnl::algorithm::convolution_auto `. * - src_desc - Source memory descriptor. * - diff_weights_desc - Diff weights memory descriptor. * - diff_dst_desc - Diff destination memory descriptor. * - strides - Strides for each spatial dimension. * - padding_l - Vector of padding values for low indices for each spatial dimension ``([[front,] top,] left)``. * - padding_r - Vector of padding values for high indices for each spatial dimension ``([[back,] bottom,] right)``. .. index:: pair: function; desc .. _doxid-structdnnl_1_1convolution__backward__weights_1_1desc_1a2d660abe71de37bbc75fab3fe227da20: .. ref-code-block:: cpp :class: doxyrest-title-code-block desc( :ref:`algorithm` aalgorithm, const :ref:`memory::desc`& src_desc, const :ref:`memory::desc`& diff_weights_desc, const :ref:`memory::desc`& diff_bias_desc, const :ref:`memory::desc`& diff_dst_desc, const :ref:`memory::dims`& strides, const :ref:`memory::dims`& dilates, const :ref:`memory::dims`& padding_l, const :ref:`memory::dims`& padding_r ) Constructs a descriptor for a dilated convolution weights gradient primitive with bias. .. note:: All the memory descriptors may be initialized with the :ref:`dnnl::memory::format_tag::any ` value of ``format_tag``. Arrays ``strides``, ``dilates``, ``padding_l``, and ``padding_r`` contain values for spatial dimensions only and hence must have the same number of elements as there are spatial dimensions. The order of values is the same as in the tensor: depth (for 3D tensors), height (for 3D and 2D tensors), and width. .. rubric:: Parameters: .. list-table:: :widths: 20 80 * - aalgorithm - Convolution algorithm. Possible values are :ref:`dnnl::algorithm::convolution_direct `, :ref:`dnnl::algorithm::convolution_winograd `, and :ref:`dnnl::algorithm::convolution_auto `. * - src_desc - Source memory descriptor. * - diff_weights_desc - Diff weights memory descriptor. * - diff_bias_desc - Diff bias memory descriptor. Passing zero memory descriptor disables the bias term. * - diff_dst_desc - Diff destination memory descriptor. * - strides - Strides for each spatial dimension. * - dilates - Dilations for each spatial dimension. A zero value means no dilation in the corresponding dimension. * - padding_l - Vector of padding values for low indices for each spatial dimension ``([[front,] top,] left)``. * - padding_r - Vector of padding values for high indices for each spatial dimension ``([[back,] bottom,] right)``. .. index:: pair: function; desc .. _doxid-structdnnl_1_1convolution__backward__weights_1_1desc_1a4cc6f6b8188002b343c5620ece8f8a72: .. ref-code-block:: cpp :class: doxyrest-title-code-block desc( :ref:`algorithm` aalgorithm, const :ref:`memory::desc`& src_desc, const :ref:`memory::desc`& diff_weights_desc, const :ref:`memory::desc`& diff_dst_desc, const :ref:`memory::dims`& strides, const :ref:`memory::dims`& dilates, const :ref:`memory::dims`& padding_l, const :ref:`memory::dims`& padding_r ) Constructs a descriptor for a dilated convolution weights gradient primitive without bias. .. note:: All the memory descriptors may be initialized with the :ref:`dnnl::memory::format_tag::any ` value of ``format_tag``. Arrays ``strides``, ``dilates``, ``padding_l``, and ``padding_r`` contain values for spatial dimensions only and hence must have the same number of elements as there are spatial dimensions. The order of values is the same as in the tensor: depth (for 3D tensors), height (for 3D and 2D tensors), and width. .. rubric:: Parameters: .. list-table:: :widths: 20 80 * - aalgorithm - Convolution algorithm. Possible values are :ref:`dnnl::algorithm::convolution_direct `, :ref:`dnnl::algorithm::convolution_winograd `, and :ref:`dnnl::algorithm::convolution_auto `. * - src_desc - Source memory descriptor. * - diff_weights_desc - Diff weights memory descriptor. * - diff_dst_desc - Diff destination memory descriptor. * - strides - Strides for each spatial dimension. * - dilates - Dilations for each spatial dimension. A zero value means no dilation in the corresponding dimension. * - padding_l - Vector of padding values for low indices for each spatial dimension ``([[front,] top,] left)``. * - padding_r - Vector of padding values for high indices for each spatial dimension ``([[back,] bottom,] right)``.