localized_covariance: add xarray wrapper #191
Conversation
Codecov Report
Additional details and impacted files@@ Coverage Diff @@
## main #191 +/- ##
=======================================
Coverage ? 80.24%
=======================================
Files ? 33
Lines ? 1473
Branches ? 0
=======================================
Hits ? 1182
Misses ? 291
Partials ? 0
Flags with carried forward coverage won't be shown. Click here to find out more. Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here. ☔ View full report at Codecov. |
|
@znicholls do you have time to take a look at this? |
mesmer/core/localized_covariance.py
Outdated
| (sample_dim,) = set(all_dims) - {dim} | ||
| out_dims = create_equal_dim_names(sample_dim) | ||
|
|
||
| # TODO: not sure which one is nicer |
There was a problem hiding this comment.
I prefer apply_ufunc, just for consistency with stuff that is elsewhere and so people can see how to use that functionality when dimensions don't line up perfectly
mesmer/core/utils.py
Outdated
| @@ -4,6 +4,10 @@ | |||
| import numpy as np | |||
| import xarray as xr | |||
|
|
|||
| # xarray cannot represent two dims with the same name -> need to give them another name | |||
| # TODO: expose this via function argument or config? | |||
There was a problem hiding this comment.
I'd do function argument in this case
| ar_coefs = np.random.randn(3) | ||
| ar_coefs = xr.DataArray(ar_coefs, dims="cell") | ||
|
|
||
| cov = np.cov(random_data_5x3, rowvar=False) |
There was a problem hiding this comment.
It's super annoying that there's not a method in xarray to calculate a covariance matrix (I get that naming the output dimensions is hard but it's such a fundamental thing, it's weird it doesn't already exist)
Co-authored-by: Zeb Nicholls <zebedee.nicholls@climate-energy-college.org>
|
Thanks for your review! |
isort . && black . && flake8CHANGELOG.rstxarray wrapper for
find_localized_empirical_covarianceandadjust_covariance_ar1. Ugly as hell but there is no good way to have the same dim more than once for a DataArray.cc @znicholls