search

markovmodel / msmtools

Tools for estimating and analyzing Markov state models
GNU Lesser General Public License v3.0
40 stars 27 forks source link

[estimation/sparse/newton] New primal-dual interior-point solver for … #82

Closed trendelkampschroer closed 6 years ago

trendelkampschroer commented 8 years ago

…reversible MLE problem

franknoe commented 8 years ago

Thanks a lot! Do we have any significant tests for reversible P-matrix estimation? I have only found a single test example for this and for the fixed-point iteration each. I think we should give a few test examples, and we should tests for consistency of both estimators in the limit of many iterations.

marscher commented 6 years ago

@trendelkampschroer Do you mind checking the "enable edits from maintainers" box in this PR on the Github website, so I can apply my fixes for your PR? Sorry for taking such a long time - how are you doing by the way? :)

trendelkampschroer commented 6 years ago

Gladly, but can't find the button on the site

marscher commented 6 years ago

On 24.06.2018 14:19, Benjamin Trendelkamp-Schroer wrote:

Gladly, but can't find the button on the site

You should find it on the PR page on the bottom right side

trendelkampschroer commented 6 years ago

Ok, done!

marscher commented 6 years ago

Thank you very much!

codecov-io commented 6 years ago

Codecov Report

Merging #82 into devel will decrease coverage by 0.24%. The diff coverage is 59.16%.

Impacted file tree graph

@@            Coverage Diff             @@
##            devel      #82      +/-   ##
==========================================
- Coverage   89.73%   89.48%   -0.25%     
==========================================
  Files         110      116       +6     
  Lines        9125     9384     +259     
==========================================
+ Hits         8188     8397     +209     
- Misses        937      987      +50
Impacted Files Coverage Δ
msmtools/analysis/tests/test_decomposition.py 99.76% <100%> (ø) :arrow_up:
msmtools/analysis/sparse/stationary_vector_test.py 96.87% <100%> (ø) :arrow_up:
msmtools/analysis/dense/stationary_vector_test.py 96.66% <100%> (ø) :arrow_up:
msmtools/analysis/sparse/decomposition_test.py 99.58% <100%> (ø) :arrow_up:
msmtools/dtraj/tests/test_trajectory.py 97.87% <100%> (+0.04%) :arrow_up:
msmtools/analysis/dense/decomposition_test.py 99.49% <100%> (ø) :arrow_up:
msmtools/estimation/sparse/newton/__init__.py 100% <100%> (ø)
...mtools/estimation/sparse/newton/objective_dense.py 12.85% <12.85%> (ø)
msmtools/estimation/sparse/newton/linsolve.py 13.69% <13.69%> (ø)
msmtools/estimation/sparse/newton/mle_rev.py 84.76% <84.76%> (ø)
... and 15 more

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 621cc79...fd2dc9a. Read the comment docs.