0.12.0
New version of sidpy has been released. The major change is that previous operations that would have resulted in return of numpy or dask arrays are now programmed to return sidpy dataset objects. This is particularly useful for instance when wishing to crop, slice, and perform standard arithmetic operations on sidpy datasets. The sidpy.fitter class has been updated to allow for complex datasets. An example notebook is provided here
In addition to this major change, there are minor bug fixes throughout to deal with changes/deprecations to numpy, matplotlib, etc. If you run into issues with the new version, please let us know. We expect that given this substantial change, not every workflow will operate as normal without any modification. If this occurs, most likely you simply have to convert the sidpy dataset to a numpy array when it fails in your codebase.