Exacteness tests and CI improvements (#73 and #74) #78
Conversation
|
What is the reason for the CI failure ? |
setup.py
Outdated
| @@ -17,13 +17,14 @@ | |||
| ), | |||
| package_dir={"": "src"}, | |||
| install_requires=[ | |||
| 'pandas>=1.2.1', | |||
| 'pandas==1.2.1', | |||
There was a problem hiding this comment.
In order to avoid the DataFrame has no attribute 'iteritems' error from simulation_based and mediation_g_estimator
There was a problem hiding this comment.
Yes, well, but we cannot be stuck with a Pandas version that seem a bit old. Let's take more recent versions, but without imposing one.
One of the reason is I think because of the fact that there's an infinitesimal difference between the obtained results and the expected results so I have to increase the tolerance threshold of the I am going to open a new PR with CI active for |
Codecov ReportAttention: Patch coverage is
❗ Your organization needs to install the Codecov GitHub app to enable full functionality. Additional details and impacted files@@ Coverage Diff @@
## main #78 +/- ##
=======================================
Coverage ? 86.53%
=======================================
Files ? 6
Lines ? 750
Branches ? 0
=======================================
Hits ? 649
Misses ? 101
Partials ? 0 ☔ View full report in Codecov by Sentry. |
|
I think we can merge this PR |
This PR is the combinaison of the following PRs :
This is at the same time a great test of the new CI pipeline as it is supposed to be launched every time a new PR is created on the
mainbranch.If any change has to be done, I suggest to :
develop