Currently, when a user computes a binned mixing matrix (e.g., with shape (32, 1501)), there's no straightforward way to know that, on the unbinned side of the covariance matrix (length 1501), the first entry of each array corresponds to multipole ℓ = 0.0.
In contrast, the maximum ℓ can always be accessed via the .upper attribute of the Result class.
I don’t think it’s necessary to modify the Result class itself, but it is important to make the minimum ℓ value clear to the user. This could be addressed by:
- Adding it to the metadata printed in the header of the
.fits file
- Updating the Heracles documentation (this would be the easier and, in my opinion, more user-friendly solution)
Thanks for the great work, Heracles team!
Currently, when a user computes a binned mixing matrix (e.g., with shape
(32, 1501)), there's no straightforward way to know that, on the unbinned side of the covariance matrix (length 1501), the first entry of each array corresponds to multipole ℓ = 0.0.In contrast, the maximum ℓ can always be accessed via the
.upperattribute of theResultclass.I don’t think it’s necessary to modify the
Resultclass itself, but it is important to make the minimum ℓ value clear to the user. This could be addressed by:.fitsfileThanks for the great work, Heracles team!