Intel(R) Math Kernel Library for Deep Neural Networks (Intel(R) MKL-DNN)
0.10
Performance library for Deep Learning
|
A class for wrapping an Intel(R) MKL-DNN handle. It is used as the base class for primitive (mkldnn_primitive_t), engine (mkldnn_engine_t), and stream (mkldnn_stream_t) handles. An object of the mkldnn::handle class can be passed by value. This class enables wrapping: More...
#include <mkldnn.hpp>
Public Member Functions | |
handle (const handle &other) | |
handle & | operator= (const handle &other) |
void | reset (T t, bool weak=false) |
Resets the value of a C handle. More... | |
T | get () const |
Returns the value of the underlying C handle. More... | |
bool | operator== (const handle &other) const |
bool | operator!= (const handle &other) const |
Protected Member Functions | |
handle (T t=0, bool weak=false) | |
Constructs a C handle wrapper. More... | |
bool | operator== (const T other) const |
bool | operator!= (const T other) const |
A class for wrapping an Intel(R) MKL-DNN handle. It is used as the base class for primitive (mkldnn_primitive_t), engine (mkldnn_engine_t), and stream (mkldnn_stream_t) handles. An object of the mkldnn::handle class can be passed by value. This class enables wrapping:
std::shared_ptr
with a proper deleter function specified through the handle_traits
class.std::weak_ptr
).