Ian Ozsvald

I've upgraded to 1.3.2. I note that on each run I get a different result now - perhaps this is a consequence of using a sampling profiler? Notably: ``` 15...

Sorry for the delay, I'm testing in prep for my next Higher Performance Python class (frustratingly on Windows for a pension fund, so I only get a slide demo of...

You're very welcome! It was the Confusion Probabilities part that got me the feedback about ROC curves and, given that article, I think my chart (whilst being pretty) is less...

@bbengfort just for completeness - if there was a desire to include my histograms-of-probabilities confusion matrix variant, the code is online at In[27] in the Notebook I used for my...

Hey @bbengfort. I have no strong view on adding specificity/sensitivity, as you note this gets less useful in the multi-class case (but then, with many classes these visualisations are probably...

Hello Ajay. I'd be happy to review your contribution, thanks for picking up on the idea. Ian On Sat, 16 Feb 2019, 11:12 Benjamin Bengfort Hi @ajaysub110 you're more than...

Can you give some background on how you've setup your environment please? What package manager so you have (Windows? System python? Venv? Other?) Have you taken a fresh checkout of...

Let me confirm where you are. You note that you've followed https://www.scikit-yb.org/en/latest/contributing.html and I think you're saying that you've followed https://www.scikit-yb.org/en/latest/contributing.html#forking-the-repository with steps 1-4 (finishing with a `develop` branch) -...

I'll note that it took me a while to get the tests running on my Linux machine a year back (issues fixed by Ben), so take it slow to make...

I don't know what an "xfailed" result means, perhaps they're marked as being expected to fail (due to other changes occurring in the code base)? I think you're in good...