MayaFlux 1.0.0
Digital-First Multimedia Processing Framework
Loading...
Searching...
No Matches

◆ to_vector()

template<DataVariantElement T>
std::optional< std::vector< T > > MayaFlux::Kakshya::TensorData::to_vector ( ) const
inlinenodiscard

Copy into a vector of T.

Scalar and complex T need a dense tensor of that dtype. glm vector and matrix T need a dense f32 tensor whose bytes divide into whole elements.

Definition at line 132 of file Tensor.hpp.

133 {
134 const auto view = access();
135 if (!view)
136 return std::nullopt;
137 const auto raw = view->bytes();
138 if constexpr (TensorElement<T>) {
139 if (!m_spec.dtype.is<T>())
140 return std::nullopt;
141 } else {
142 if (!m_spec.dtype.is<float>() || raw.size() % sizeof(T) != 0)
143 return std::nullopt;
144 }
145 if (raw.empty())
146 return view->element_count() == 0 ? std::optional<std::vector<T>>(std::vector<T> {}) : std::nullopt;
147 std::vector<T> result(raw.size() / sizeof(T));
148 std::memcpy(result.data(), raw.data(), result.size() * sizeof(T));
149 return result;
150 }
std::optional< TensorAccess > access() const
Borrow the tensor, or std::nullopt when it holds nothing valid.
Definition Tensor.cpp:150

References access(), and m_spec.

Referenced by to_variant().

Here is the call graph for this function:
Here is the caller graph for this function: