Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We鈥檒l occasionally send you account related emails.

Already on GitHub? Sign in to your account

VAMPyR-1.0 (or future) wishlist #49

Open
8 of 17 tasks
stigrj opened this issue Sep 6, 2021 · 3 comments
Open
8 of 17 tasks

VAMPyR-1.0 (or future) wishlist #49

stigrj opened this issue Sep 6, 2021 · 3 comments

Comments

@stigrj
Copy link
Contributor

stigrj commented Sep 6, 2021

Dear 馃巺

Please edit and add new features at will (not saying everything should be implemented for v1):

  • More user friendly MRA initialization, e.g. world_size = [-20, 20]**3
  • Function plotting (line, surface, cube)
  • Function printing, what should be displayed with print(tree)?
  • Basis visualization
    • scaling and wavelet functions
    • two-scale relations
    • dilation and translation of functions
  • Features to build your own wavelet code
    • extract filter matrices
    • Legendre polynomials
  • Tree/grid visualization
    • any ideas?
    • ASCII art for 1D trees?
    • existing Python library?
  • Difference plots
    • functions/differences at different resolution levels
    • plot wavelet information content as function of scale
  • New operators
    • convolution operator from arbitrary 1D Gaussian expansion (examples?)
    • other 1D operators (e.g. exponential operator for solving the heat equation and Burger's equation, see ABGV)
    • brute-force projection of any kind of non-separable operator?
  • Use scikit-build to build: https://scikit-build.readthedocs.io/en/latest/
  • Documented test-cases
    • H-atom
    • ZORA
    • generalized poisson
  • Interface documentation
  • Deep copy
  • Generic kernel G(x-y) which takes vector of functions g_i(x) such that G(x) = \sum_i g_i(x)
  • Resolve fast vs accurate dilemma for FunctionTree evaluation
  • Handling of complex functions (separate trees for Real and Imaginary part)
  • Add call function to operators, g_tree = P(f_tree)
  • New MWProjector class which enables f_tree = Q(func);
@robertodr
Copy link
Contributor

robertodr commented Sep 6, 2021

- [ ] Use scikit-build to build: https://scikit-build.readthedocs.io/en/latest/

@robertodr
Copy link
Contributor

robertodr commented Sep 8, 2021

  • Operator overloading for most common form of operations. For example:/
     .def("__add__", [](FunctionTree<D>& tree_a, FunctionTree<D>& tree_b) { /* code here */ }, py::is_operator());

@robertodr
Copy link
Contributor

robertodr commented Aug 11, 2022

  • KAIN and DIIS implementation.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

2 participants