Description
Add a new finite_jacobian_tensor<N> function which follows the convention that even-order tensors (e.g., matrices) are stored in column-blocks, while odd-order tensors are stored in row-blocks.
This is based on the tensor vectorization presented in "Dynamic Deformables" by Kim and Eberle [2022].
- Update to C++17
Examples
-
finite_jacobian_tensor<3>(odd tensor order → row-block layout)$f: \mathbb{R}^n \mapsto \mathbb{R}^{p \times q}$ -
$\nabla f: \mathbb{R}^n \mapsto \mathbb{R}^{pq \times n}$ where every$p$ rows represent one derivative w.r.t.$x_i$ for$0 \leq i < n$
-
finite_jacobian_tensor<4>(even tensor order → column-block layout)$f: \mathbb{R}^n \mapsto \mathbb{R}^{p \times q \times r}$ -
$\nabla f: \mathbb{R}^n \mapsto \mathbb{R}^{pq \times rn}$ where every$q$ columns represent one derivative w.r.t.$x_i$ for$0 \leq i < n$
Full Changelog: v1.0.3...v1.0.4