reverse-communication-iterative-solver 0.0.1

Description:

reversecommunicationiterativesolver 0.0.1

This package aims to provide a common template for building iterative solvers,
those solver that approximate problem solutions with incremental steps (called “update” in this project).
This template will over a minimal base for developing, tuning and using iterative solvers from the earliest
protypting phase to the delivery of block box algorithm.
The interface between an (advanced) user and the iterative solver is decomposed into the following 4 main phases

set the solver controls and problem inputs before the next update;
reset the solver controls reset problem inputs after a update failure;
study any you want combaning the information from problem, unknows, and solver;
set when the solver has finished its job.

This decomposition is done via a Reverse Communication (RC)
approach, an old style of programming used for
example in the ARPACK library, combined with Python classes.
We adopt this approach because it let the user work “manually” on the structures
describing its problem instead of having to define in advance
the procedure to handle the steps 1,2,3,4. This is particularly convient in early step of
the algorithm development. A nicer and clenear function hiding the RC cycle to an user who
wants a block-box solver can be easily build.

Free software: GNU Lesser General Public License v3 or later (LGPLv3+)


Installation
pip install reverse-communication-iterative-solver
You can also install the in-development version with:
pip install https://github.com/enricofacca/python-reverse_communication_iterative_solver/archive/master.zip


Documentation
https://python-reverse_communication_iterative_solver.readthedocs.io/


Development
To run all the tests run:
tox
Note, to combine the coverage data from all the tox environments run:






Windows
set PYTEST_ADDOPTS=--cov-append
tox


Other
PYTEST_ADDOPTS=--cov-append tox






Changelog

0.0.1 (2021-12-30)

Updating the README to real package aim



0.0.0 (2021-12-29)

First release on PyPI.