Pamphile Roy comments

Results 452 comments of


                                            Pamphile Roy

JOSS paper

Alright, the paper is approved 🎉 Getting this in and cutting a release now for the process.

Order in `MultiChoice`

Need to follow up with https://github.com/bokeh/bokeh/pull/11663#issuecomment-1755884702

Order in `MultiChoice`

This is tracked also on panel's side here https://github.com/holoviz/panel/issues/4454

Interactive figures

Another solution https://panel.holoviz.org/reference/panes/ECharts.html

Add `base_sequence` to `sample.sobol`

I see what you mean. I am not sure, do you have a use case yourself? This param is already questionable 😅 I don't mind adding though if you want....

Divided by zero in the Sobol's confidence level calculation

Thank you for the report. I am not sure about the check that you propose. This is a machine precision issue and so I would lean on updating the check...

Divided by zero in the Sobol's confidence level calculation

I would actually prefer if we worked toward using the new API we added to SciPy: https://github.com/scipy/scipy/blob/main/scipy/stats/_sensitivity_analysis.py Here we handle this issue by letting NaN propagate and then replace all...

Divided by zero in the Sobol's confidence level calculation

Thanks for the links. Out of curiosity, did you check the result that SciPy gives? Looks like we might have to also fix that in SciPy then. We need to...

Native MathML doesn't work on the numpy and scipy documentation websites

@ragdollfun can you check against old versions of SciPy's documentation? See https://docs.scipy.org/doc/ for links

ENH: biteopt global optimization algorithm

Thank you @dschmitz89 for reviving this. As said on the previous issue, I am +1 on adding new global optimisers. The move to Cython is not mandatory at all for...