Reshape.hpp File Reference

Copying reshape and flatten operations. More...

#include <algorithm>
#include <stratax/core/dtypes/Concepts.hpp>
#include <stratax/core/Shape.hpp>
#include <stratax/exceptions/Exceptions.hpp>
#include <stratax/containers/Tensor.hpp>
#include <stratax/containers/Vector.hpp>

Include dependency graph for Reshape.hpp:

This graph shows which files directly or indirectly include this file:

Go to the source code of this file.

Functions
template<Array A>
stratax::container::Tensor< typename A::value_type >	stratax::manipulation::reshape (const A &arr, const stratax::core::Shape &shape)
	Copies a Stratax array or view into a tensor with a new shape.
template<Array A>
stratax::container::Vector< typename A::value_type >	stratax::manipulation::flatten (const A &arr)
	Copies a Stratax array or view into a rank-one vector.

Detailed Description

Copying reshape and flatten operations.

Definition in file Reshape.hpp.

Function Documentation

flatten()

template<Array A>

stratax::container::Vector< typename A::value_type > stratax::manipulation::flatten

(

const A &

arr

)

Copies a Stratax array or view into a rank-one vector.

Values retain the source array's flat row-major iterator order. The returned vector has shape {arr.size()} and owns storage independent of arr.

Template Parameters

A	Array or view type satisfying Array.

Parameters

arr	Source array to flatten.

Returns

Owning Vector with A::value_type and copied values.

Exceptions

std::bad_alloc

If output allocation fails.

Complexity

O(arr.size()) for owning input; O(arr.size() * arr.rank()) for strided views.

Definition at line 69 of file Reshape.hpp.

reshape()

template<Array A>

stratax::container::Tensor< typename A::value_type > stratax::manipulation::reshape	(	const A &	arr,
		const stratax::core::Shape &	shape
	)

Copies a Stratax array or view into a tensor with a new shape.

The target shape must describe exactly the same number of elements as the source. Values are copied in flat iterator order, so only the logical shape and row-major strides change. The returned tensor owns independent storage.

Template Parameters

A	Array or view type satisfying Array.

Parameters

arr	Source array whose flat element order is preserved.
shape	Requested shape of the returned tensor.

Returns

Owning Tensor with A::value_type, shape, and copied values.

Exceptions

Exceptions::DimensionError	If shape.elements() or a target stride cannot be represented.
Exceptions::ShapeError	If shape.elements() != arr.size().
std::bad_alloc	If output storage or metadata allocation fails.

Complexity

O(arr.size() + shape.rank()) for owning input; O(arr.size() * arr.rank() + shape.rank()) for strided views.

Definition at line 39 of file Reshape.hpp.