Maksim Terpilovskii

I have successfully used this wheel under Win10: https://github.com/fo40225/tensorflow-windows-wheel/tree/master/2.2.0/py38/CPU%2BGPU/cuda110cudnn8sse2

@sshkhr Could be you please add the school organized by our lab? **Summer School in Experimental Methods & Simulations in Social and Cognitive Sciences (EMS-SCS 2021)** *Format*: mixed (mostly online...

Hey! I think this may be due to insufficient statistical power of Dunn's test. You can try Conover's test instead. I will look for the relevant research on this. And...

Hello! Thank you for the suggestion. I'll try to implement it

Hey! I am using `PMCMRplus` as a reference package and two datasets (seaborn's `exercise`, a custom one for block design data, and one more for Tukey HSD) that are initialized...

[PMCMRplus](https://cran.r-project.org/web/packages/PMCMRplus/PMCMRplus.pdf) contains references to scientific articles as well as sample datasets from literature (such as Sachs, 1997). It would be nice to move `scikit-posthocs` tests to these data. I do...

We can look through the articles and take the results provided by researchers as references for tests. It is a huge work, though

Thank you for your contribution

Thank you. Actually, the problem we should deal with first is correctness of implementation. Your code tries to check a test by playing with type I errors (false positives), which...

Hello! Thank you for reporting this. I cannot find this type of error in the latest codebase of `scipy`. What version of `scipy` are you using?