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A newer version of this document is available.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.
Visible to Intel only — GUID: GUID-8899171F-9514-49F8-A1C1-776B8AB77D88
Visible to Intel only — GUID: GUID-8899171F-9514-49F8-A1C1-776B8AB77D88
struct dnnl::matmul
Overview
Matrix multiplication (matmul) primitive. More…
#include <dnnl.hpp> struct matmul: public dnnl::primitive { // structs struct primitive_desc; // construction matmul(); matmul(const primitive_desc& pd); matmul(const primitive_desc& pd, const std::vector<uint8_t>& cache_blob); };
Inherited Members
public: // enums enum kind; // methods handle<T, traits>& operator = (const handle<T, traits>&); handle<T, traits>& operator = (handle<T, traits>&&); void reset(T t, bool weak = false); T get(bool allow_empty = false) const; operator T () const; operator bool () const; bool operator == (const handle<T, traits>& other) const; bool operator != (const handle& other) const; const_dnnl_primitive_desc_t get_primitive_desc() const; kind get_kind() const; std::vector<uint8_t> get_cache_blob() const; void execute(const stream& astream, const std::unordered_map<int, memory>& args) const; handle(); handle(); handle(); handle();
Detailed Documentation
Matrix multiplication (matmul) primitive.
Construction
matmul()
Default constructor. Produces an empty object.
matmul(const primitive_desc& pd)
Constructs a matmul primitive.
Parameters:
pd |
Primitive descriptor for a matmul primitive. |
matmul(const primitive_desc& pd, const std::vector<uint8_t>& cache_blob)
Constructs a matmul primitive from a cache blob.
Parameters:
pd |
Primitive descriptor for a matmul primitive. |
cache_blob |
Cache blob. |