Sebastian Proost

1. yes, this one should be tackled. It is an easy fix, just the message that needs to be corrected. I've created a small PR correcting the message. 2. Thomas...

I can't take credit for the work on statannotations, but I'm happy to help where I can as this is a feature that would be used all over imho. Have...

Finally got around writing up a post about how to use the current version of statannotatons with scikit-posthocs. http://blog.4dcu.be/programming/2021/12/30/Posthoc-Statannotations.html if there are no short-term plans to support post-hoc tests, it...

The argument ```group_col``` in the function ```posthoc_dunn``` should be set to "Disease" . The variables DiseaseHZDMW and dataHZDMW are created for the Kruskal-Wallis test, you don't need them later on...

yes, that is the way to do this. Do mention that only significant results are shown, but all pairwise combinations were tested (e.g. in the figure caption) as the number...

This discussion is getting a little far off topic, could you send me an email? There are contact details on my blog.

> #Create a list of "non-useful" pairs and trim the list removing rows with unuseful pairs nonuseful=[here the row numbers with the pairs that are not neededseparated by coma] molten_df_ri_trim...

We were discussing this before ( https://github.com/trevismd/statannotations/issues/39 ) and it isn't as easy as one might think to support these without a significant re-write. You can however compute these before...

https://github.com/trevismd/statannotations/pull/95 is the PR with the improved implementation, based on our discussion here.

Thanks for these models! ControlNet results with my 1.5 models were awesome, but I have trained so many 2.1 embeddings I'd love to use with this.