Marko Toplak
Marko Toplak
Orange.statistics.utils.stats() returns number of unknown (NaN) values for dense tables and the number of nonzero values for sparse tables. This wrongly equates undefined values and unknown values. I'd change that....
CD diagrams crash headless tests on Linux. Because Orange does not use these functions we should just remove them.
##### Issue This code is likely obsolete. It was called just from a single function, so I refactored a single case that made sense there. ##### Includes - [X] Code...
**What's wrong?** - I created a fresh conda environment as per README and installed orange3 there - opened canvas - tried to install Text with add-on dialog - the add-on...
**What's wrong?** A user reported installation failing like this:  And then after a day followed up with: "I found out! If your installation path has spaces then this error...
`Orange.__init__.py` has some stuff that is Qt / openGL specific and handles some strange cases. This code unnecessary complicated installation for users that just want to use Orange for scripting...
`Table.copy` can sometimes change sparsity, which is unexpected. It is likely needlessly complex as it goes through the constructor (`Table.__new__`). I found this in one of our tests: ``` import...
- [ ] **What's wrong?** Normalization is really slow. I did normalization and PCA on the 100000x1000 data set, and PCA was much faster than normalization. Here is what it...
**What's your use case?** Sometimes a user wants to construct a numeric feature that is not used for modeling but only for display. I needed that feature quite a few...
As suggested in #309 by @kmilos - and I agree.