Intel(R) Math Kernel Library for Deep Neural Networks (Intel(R) MKL-DNN)
0.13
Performance library for Deep Learning
|
Modules | |
Operation descriptors | |
Classes | |
struct | mkldnn_blocking_desc_t |
Generic description of blocked data layout for most memory formats. More... | |
struct | mkldnn_convolution_desc_t |
A descriptor of a convolution operation. More... | |
struct | mkldnn_eltwise_desc_t |
A descriptor of a element-wise operation. More... | |
struct | mkldnn_softmax_desc_t |
A descriptor of a Softmax operation. More... | |
struct | mkldnn_pooling_desc_t |
A descriptor of a pooling operation. More... | |
struct | mkldnn_lrn_desc_t |
A descriptor of a Local Response Normalization (LRN) operation. More... | |
struct | mkldnn_batch_normalization_desc_t |
A descriptor of a Batch Normalization operation. More... | |
struct | mkldnn_inner_product_desc_t |
A descriptor of an inner product operation. More... | |
struct | mkldnn_convolution_relu_desc_t |
A descriptor of a convolution followed by relu operation. More... | |
Macros | |
#define | TENSOR_MAX_DIMS 12 |
Maximum number of dimensions a tensor can have. More... | |
Typedefs | |
typedef int | mkldnn_dims_t[TENSOR_MAX_DIMS] |
A type to describe tensor dimensions. More... | |
typedef ptrdiff_t | mkldnn_strides_t[TENSOR_MAX_DIMS] |
A type to describe strides within a tensor. More... | |
typedef mkldnn_eltwise_desc_t | mkldnn_relu_desc_t |
#define TENSOR_MAX_DIMS 12 |
Maximum number of dimensions a tensor can have.
Only restricts the amount of space used for the tensor description. Individual computational primitives may support only tensors of certain dimensions.
typedef int mkldnn_dims_t[TENSOR_MAX_DIMS] |
A type to describe tensor dimensions.
typedef ptrdiff_t mkldnn_strides_t[TENSOR_MAX_DIMS] |
A type to describe strides within a tensor.