class ov::intel_gpu::ocl::USMTensor¶
Overview¶
This class represents an abstraction for GPU plugin remote tensor which can be shared with user-supplied USM device pointer. The plugin object derived from this class can be obtained with ClContext::create_tensor() call. More…
#include <ocl.hpp>
class USMTensor: public ov::RemoteTensor
{
public:
// methods
static voidconst Tensor& type_check();
void \* get();
};Inherited Members¶
public:
// methods
Tensor&const Tensor& operator = ();
Tensor&Tensor&& operator = ();
voidconst ov::Shape& set_shape();
const element::Type& get_element_type() const;
const Shape& get_shape() const;
voidov::Tensor copy_to() const;
bool is_continuous() const;
size_t get_size() const;
size_t get_byte_size() const;
Strides get_strides() const;
void \*const element::Type& data() const;
template <, >
T \* data() const;
bool operator ! () const;
operator bool () const;
template <>
std::enable_if<std::is_base_of<Tensor, T>::value, bool>::type is() const;
template <>
const std::enable_if<std::is_base_of<Tensor, T>::value, T>::type as() const;
static voidconst Tensor& type_check();
static voidconst Tensor&const std::map<std::string, std::vector<std::string>>& type_check(
,
);
void \*const element::Type data();
template <>
T \* data();
voidov::Tensor& copy_to() const;
ov::AnyMap get_params() const;Detailed Documentation¶
This class represents an abstraction for GPU plugin remote tensor which can be shared with user-supplied USM device pointer. The plugin object derived from this class can be obtained with ClContext::create_tensor() call.
User can obtain USM pointer from this class.