Rewrite inverse methods of model estimation #189

BjoernLudwigPTB · 2020-11-06T15:26:49Z

This will finally resolve #147.

Merge and rewrite

LSIIR, invLSIIR and invLSIIR_unc based on Monte Carlo,
LSFIR, invLSFIR, invLSFIR_unc, invLSFIR_uncMC.
After reviewing all this, we should merge Refactor and finally finish implementation #251 into this one here and then merge all that into prepare_major_release_2.0.0

Checkout the docs as well to be sure they look nicely: https://pydynamic.readthedocs.io/en/fix-147_rewrite_inverse_methods_of_model_estimation

codecov-io · 2020-11-06T21:36:11Z

Codecov Report

Merging #189 (8118b55) into prepare_major_release_2.0.0 (f4ea51c) will increase coverage by 2.62%.
The diff coverage is 52.17%.

@@                       Coverage Diff                       @@
##           prepare_major_release_2.0.0     #189      +/-   ##
===============================================================
+ Coverage                        49.72%   52.34%   +2.62%     
===============================================================
  Files                               24       24              
  Lines                             1798     1790       -8     
  Branches                           315      319       +4     
===============================================================
+ Hits                               894      937      +43     
+ Misses                             818      760      -58     
- Partials                            86       93       +7

Impacted Files	Coverage Δ
PyDynamic/__init__.py	`100.00% <ø> (ø)`
PyDynamic/misc/filterstuff.py	`60.65% <ø> (+26.22%)`	⬆️
PyDynamic/model_estimation/__init__.py	`100.00% <ø> (ø)`
PyDynamic/model_estimation/fit_filter.py	`31.96% <52.17%> (+13.02%)`	⬆️

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update f4ea51c...d58d702. Read the comment docs.

BjoernLudwigPTB · 2020-11-06T21:56:27Z

Before I start to work on LSFIR and the others it would be great if @mgrub @eichstaedtPTB we could take a peek together (do a line by line code review) and you could give some feedback on the approaches I have chosen. I will adress this on next tuesday.

PyDynamic/model_estimation/fit_filter.py

test/conftest.py

PyDynamic/model_estimation/fit_filter.py

mgrub

Great and extensive work!

Is there a reason, that the testing of the model_estimation-module is done in test_identification.py? Would a cosmetic change to test_model_estimation.py be suitable?

test/test_identification.py

PyDynamic/model_estimation/fit_filter.py

…accordingly

…sis test" This reverts commit 673b79f.

…ctions

…FIR_unc to filter coefficients

…_comparison Improve random covariance check for positive semi-definiteness

…r_fitting_H_directly): remove _

…ings

…ures

…nto refactor_and_finally_finish_implementation

…nto refactor_and_finally_finish_implementation # Conflicts: # test/test_LSFIR.py

… comparison

…ables

…reciprocal with uncertainties BREAKING CHANGE: Combine _deconvolution.fit_filter_ and _identification.fit_filter_ into _model_estimation.fit_filter_ and provide access to all functionality via according parameter sets for _model_estimation.fit_filter.LSFIR_ and _model_estimation.fit_filter.LSIIR_.

BjoernLudwigPTB self-assigned this Nov 6, 2020

BjoernLudwigPTB changed the title ~~Fix/#147 rewrite inverse methods of model estimation~~ Fix #147 rewrite inverse methods of model estimation Nov 6, 2020

BjoernLudwigPTB changed the title ~~Fix #147 rewrite inverse methods of model estimation~~ Rewrite inverse methods of model estimation Nov 6, 2020

BjoernLudwigPTB mentioned this pull request Nov 6, 2020

Prepare major release 2.0.0 #151

Merged

12 tasks

BjoernLudwigPTB requested a review from mgrub November 6, 2020 21:34

BjoernLudwigPTB removed the request for review from mgrub November 6, 2020 21:54

BjoernLudwigPTB commented Nov 9, 2020

View reviewed changes