Kyle Barbary
Kyle Barbary
Nearly all the parameters of the model and `CatalogEntry` are given in world coordinates, independent of image pixel scale or WCS. For example, the position is given in (RA, Dec),...
For large numbers of dimensions, `vol_prefactor` gives 0. due to underflow: ``` In [8]: nestle.vol_prefactor(450) Out[8]: 5.73e-322 In [9]: nestle.vol_prefactor(500) Out[9]: 0.0 ``` One should probably not be using nestle...
The test suite currently includes a few "integration tests" - full problems with known analytic evidence as a benchmark - but only in low dimensions. (Though a moderately high dimensional...
It can be expensive to determine bounding ellipsoids, particuarly for the multi-ellipsoidal method. However, this doesn't need to be done every iteration. The `update_interval` parameter lets the user set how...
In the multi-ellipsoidal algorithm, a cluster must have a minimum of ndim+1 member points. This means that two widely separated modes will not be split into separate clusters if one...
In the multi-ellipsoidal clustering, clusters are created by recursively splitting points into two clusters. In some (contrived) examples, this results in strange partitioning. Example:  In this case, it would...
In the multi-ellipsoidal algorithm, points are split into clusters recursively. At each split, K-means with K=2 is used to cluster the points. In Feroz, Hobson & Bridges (2009), the algorithm...
I've seen a lot of confusion over the years similar to https://github.com/sncosmo/sncosmo/discussions/342 about the meaning of "magnitude", "flux" and "magnitude system" and how they all relate. It would probably be...
~A lot of the documentation is a bit stale and should be updated before the next release.~ *Now done* It would also be nice to have: - ~example of creating...
Functions like `sncosmo.get_source` create instances of `sncosmo.Source` instantiated with some model data. A lot of this data is a single spectral time series. In some instances it would be nice...