| ▶Ncldnn | |
| ▶Ndetail | |
| Cbuild_option_traits | Helper template to convert build_option_type value to particular build_option class |
| ▶Ndetails | Helper structs used in tensor constructor with dim_vec_kinds |
| Cdim_vec_kind_init | Template class used in tensor constructor using dim_vec_kinds |
| Cdim_vec_limits | Template class with max_dimensionalities and dimension offset for dimension kinds |
| Cdim_vec_limits< dim_vec_kind::batch > | |
| Cdim_vec_limits< dim_vec_kind::feature > | |
| Cdim_vec_limits< dim_vec_kind::spatial > | |
| Cmemory_c_to_cpp_converter | |
| ▶Ninstrumentation | |
| Cprofiling_info | Represents list of profiling_interval |
| Cprofiling_interval | Represents prifiling interval as its name and value |
| Cprofiling_period | Abstract class to represent profiling period |
| Cprofiling_period_basic | Basic profiling_period implementation which stores data as an simple period value |
| Ctimer | Helper class to calculate time periods |
| ▶Nmeta | |
| Call | |
| Call< Val, Values... > | |
| Calways_false | |
| Calways_false_ty_val | |
| Cis_any_of | |
| Cis_any_of< T, U, Rest... > | |
| Cval_tuple | |
| Cactivation | Activation using rectified linear unit or parameterized rectified linear unit |
| Cbatch_norm | Batch normalization primitive |
| Cbuild_option | Represents user-provided program build option |
| Cbuild_option_bool | build_option specialization for boolean options |
| Cbuild_option_directory | build_option specialization for selecting a directory |
| Cbuild_option_outputs | build_option specialization for program outputs list |
| Cbuild_option_tuning_config | build_option specialization for tuning config |
| Cbuild_options | Represents program build options list |
| Cconcatenation | |
| Cconvolution | Performs forward spatial convolution with weight sharing. Also supports built-in Relu cldnn_activation_desc available by setting it in arguments |
| Ccrop | Performs crop operation on input |
| Ccustom_gpu_primitive | This primitive executes a custom kernel provided by the application |
| Cdata | Provides input data to topology |
| Cdata_type_to_type | Converts data_types to C++ type |
| Cdata_type_traits | Helper class to identify key properties for data_types |
| Cdeconvolution | Performs transposed convolution. Also supports built-in Relu activation available by setting it in arguments |
| Cdetection_output | Generates a list of detections based on location and confidence predictions by doing non maximum suppression |
| Celtwise | Performs elementwise operations (sum, subtract, max or product) on two input primitives Also supports built-in Relu activation available by setting it in arguments. |
| Cengine | Represents clDNN engine object |
| Cengine_configuration | Configuration parameters for created engine |
| Cerror | ClDNN specific exception type |
| Cevent | Represents an clDNN Event object |
| Cformat | Represents memory formats (orders).
In CNN most of data is described as 4 dimensional blocks. In Intel(R) clDNN library we describe memory with 4 letters |
| Cformat_traits | Format information helper class |
| Cfully_connected | Performs forward fully connected layer (inner product). Also supports built-in Relu cldnn_activation_desc available by setting it in arguments. |
| Chalf_impl | |
| Cinput_layout | Provides input layout for a data to be passed later to network |
| Clayout | Describes memory layout |
| Clrn | Local response normalization |
| Cmemory | Represents buffer with particular layout |
| Cnetwork | Executable network allocated from program |
| Cnetwork_output | Represents network output returned by network::get_output() |
| Cnormalize | Normalizes the input using an L2 norm and multiplies the output with scale value. The scale can be equal for all channels or one scale per channel |
| Cpadding | Represents data padding information |
| Cpermute | Permutes data in the memory, with respect to provided order |
| Cpointer | Helper class to get an access memory data |
| Cpooling | Performs "pooling" operation which is a form of non-linear down-sampling |
| ▶Cprimitive | Base class of network primitive description |
| Cfixed_size_vector_ref | Initialize fields common for all primitives |
| Cprimitive_id_arr | |
| Cprimitive_base | Base class for all primitives implementations |
| Cprior_box | Generates a set of default bounding boxes with different sizes and aspect ratios |
| Cprogram | Compiled program build from topology by engine |
| Cproposal | |
| Cregion_yolo | Normalizes results so they sum to 1 |
| Creorder | Changes how data is ordered in memory. Value type is not changed & all information is preserved |
| Creorg_yolo | Normalizes results so they sum to 1 |
| Creshape | Changes information about inputs's layout effectively creating new memory which share underlaying buffer but is interpreted in a different way (different shape) |
| Croi_pooling | |
| Cscale | Performs elementwise product of input and scale_input |
| Csoftmax | Normalizes results so they sum to 1 |
| Csplit | Performs split operation on input |
| Ctensor | N-dimensional vector. Mostly used to represent memory size |
| Ctopology | Network topology to be defined by user |
| Ctuning_config_options | Tuning configuration |
| Ctype_to_data_type | Converts C++ type to data_types |
| Cupsampling | Performs nearest neighbor/bilinear upsampling Also supports built-in Relu activation available by setting it in arguments |
| Ccldnn_activation_additional_params | Activation additional params |
| Ccldnn_activation_desc | Activation using rectified linear unit or parameterized rectified linear unit |
| Ccldnn_arg | Custom primitive kernel argument type |
| Ccldnn_batch_norm_desc | Batch normalization primitive |
| Ccldnn_build_option | Represents network build option |
| Ccldnn_concatenation_desc | |
| Ccldnn_convolution_desc | Performs forward spatial convolution with weight sharing. Also supports built-in Relu cldnn_activation_desc available by setting it in arguments |
| Ccldnn_crop_desc | Performs crop operation on input |
| Ccldnn_custom_gpu_primitive_desc | This primitive executes a custom kernel provided by the application |
| Ccldnn_data_desc | Provides input data to topology |
| Ccldnn_deconvolution_desc | Performs transposed convolution. Also supports built-in Relu cldnn_activation_desc available by setting it in arguments |
| Ccldnn_detection_output_desc | Generates a list of detections based on location and confidence predictions by doing non maximum suppression |
| Ccldnn_eltwise_desc | Performs elementwise operations (sum, subtract, max or product) on two input primitives Also supports built-in Relu cldnn_activation_desc available by setting it in arguments. |
| Ccldnn_engine_configuration | Configuration parameters for created engine |
| Ccldnn_engine_info | Information about the engine returned by cldnn_get_engine_info() |
| Ccldnn_float_arr | Represents reference to an array of floats |
| Ccldnn_fully_connected_desc | Performs forward fully connected layer (inner product). Also supports built-in Relu cldnn_activation_desc available by setting it in arguments |
| Ccldnn_input_layout_desc | Provides input layout for a data to be passed later to network |
| Ccldnn_layout | Memory layout description |
| Ccldnn_lrn_desc | Local response normalization |
| Ccldnn_network_output | Output information for executed cldnn_network |
| Ccldnn_normalize_desc | Normalizes the input using an L2 norm and multiplies the output with scale value. The scale can be equal for all channels or one scale per channel |
| Ccldnn_padding | Padding information |
| Ccldnn_permute_desc | Permutes data in the memory, with respect to provided order |
| Ccldnn_pooling_desc | Performs "pooling" operation which is a form of non-linear down-sampling |
| Ccldnn_primitive_desc | Basic primitive descriptor structure |
| Ccldnn_primitive_id_arr | Represents reference to an array of primitive ids |
| Ccldnn_prior_box_desc | Generates a set of default bounding boxes with different sizes and aspect ratios |
| Ccldnn_profiling_interval | Profiling information for an executed network primitive |
| Ccldnn_proposal_desc | |
| Ccldnn_region_yolo_desc | Region softmax specific for yolo2 topology |
| Ccldnn_reorder_desc | Changes how data is ordered in memory. Value type is not changed & all information is preserved |
| Ccldnn_reorg_yolo_desc | Yolo2 topology specific data reorganization primitive |
| Ccldnn_reshape_desc | Changes information about inputs's layout effectively creating new memory which share underlaying buffer but is interpreted in a different way (different shape) |
| Ccldnn_roi_pooling_desc | |
| Ccldnn_scale_desc | Performs elementwise product of input and scale_input |
| Ccldnn_softmax_desc | Normalizes results so they sum to 1. The scope of normalization is defined by a member dimension |
| Ccldnn_split_desc | Performs split operation on input |
| Ccldnn_tensor | N-dimensional vector. Mostly used to represent memory size |
| Ccldnn_tensor_arr | Represents reference to an array of tensor |
| Ccldnn_tuning_config | Tuning config |
| Ccldnn_uint16_t_arr | Represents reference to an array of uint16_t |
| Ccldnn_upsampling_desc | Performs nearest neighbor/bilinear upsampling Also supports built-in Relu activation available by setting it in arguments |
| Ccldnn_version | |
1.8.14