Time Based ExponentialMovingStatistics and ExponentialCovariance#31
Time Based ExponentialMovingStatistics and ExponentialCovariance#31alanmazankiewicz wants to merge 90 commits intograntjenks:masterfrom
Conversation
…Added Docstrings in respective class
…pdef clear() (return value is an object)
…on of fromstate() -> changed Nonetype to double
…__main__.py, added to benchmark
…istics, implemented ExpoenentialCovariance (push and correlation TBD)
…class 'runstats._core.Statistics'>
|
Can you have a look at the pipeline failure of this PR: https://github.com/alanmazankiewicz/python-runstats/pull/1/checks?check_run_id=2978450007. Its a merge from my "rebase" branch onto my "master" branch. I created it to test the pipeline. The pipeline fails at "test (ubuntu-latests, 3.x)" with Have you encountered that issue before? When I locally checkout the remote/upstream/master branch (your current version) and run |
|
I'm not sure how it worked before but it should be fixed now on master. Please rebase. |
…gStatistics, added ExponentialMovingCovariance to tests
…ion instead of ExponentialStatistics
…t (from Optional[float]) by using float('nan') for cython compatibility
…n to ExponentialMovingCovariance make_exponential_covariance
…ased function definition renamed to test_pickle_exponential_statistics
…t_delay to delay setter for cython compatiability
…nction to ExponentialMovingStatistics
|
Pull Request Rebased. The Problems mentioned in the description are not present anymore. |
|
Did you had the chance to look at changes? |
Main features:
Smaller fixes/changes/features:
Problems:
I have done my best to resolve all issues that are raised by the CI pipeline. However, I could not resolve all of them. You've been just changing the structure of the project including the CI pipeline and the way Cython is compiled. Thus, I guess you should be able to identify the sources of these problems fairly quickly.
RunStats_ is used. I don't know why this is happening.When changing in the Readme
<class 'runstats.core.Statistics'>to<class 'runstats._core.Statistics'>this does not resolve the issue. Since currently in the main repositories master branch its<class 'runstats.core.Statistics'>I left it that way.tox -e pyit tells me that there is 98% test coverage and reports missing linesThese are the lines where Exceptions are raised in the new
freeze()andunfreeze()methods. However, these are being tested bytest_raise_if_not_time_exp_stats()