search

JohannesBuchner / UltraNest

Fit and compare complex models reliably and rapidly. Advanced nested sampling.
https://johannesbuchner.github.io/UltraNest/
Other
142 stars 30 forks source link

Add scipy in requirements in setup.py #82

Closed ferrigno closed 1 year ago

ferrigno commented 1 year ago

I installed ultranest on a fresh environment and importing it requires scipy, so I add it the setup.py

JohannesBuchner commented 1 year ago

Hmm, I thought it wouldn't be required, only for some optional features.

I see that in ultranest.plot it is imported and then used here: https://github.com/JohannesBuchner/UltraNest/blob/4caddcb9d9e527fd3fc906c914142c3a88a80406/ultranest/plot.py#L115

Perhaps it would be best to replace this line with a call to numpy.quantile / percentile.

ferrigno commented 1 year ago

If such a heavy package is used just for this, better to exploit numpy, I agree. However, I might not be able to take care of this in the next days, so if you prefer to fix it by yourself, it would be better.

Il Sab 24 Dic 2022, 11:21 Johannes Buchner @.***> ha scritto:

Hmm, I thought it wouldn't be required, only for some optional features.

I see that in ultranest.plot it is imported and then used here:

https://github.com/JohannesBuchner/UltraNest/blob/4caddcb9d9e527fd3fc906c914142c3a88a80406/ultranest/plot.py#L115

Perhaps it would be best to replace this line with a call to numpy.quantile / percentile.

— Reply to this email directly, view it on GitHub https://github.com/JohannesBuchner/UltraNest/pull/82#issuecomment-1364504297, or unsubscribe https://github.com/notifications/unsubscribe-auth/AATI74ZTIDNGIIYOMFUL6TTWO3FELANCNFSM6AAAAAATGYKXJI . You are receiving this because you authored the thread.Message ID: @.***>

JohannesBuchner commented 1 year ago

Closing this, since it is not required unless a user chooses to work with the PredictionBand class.