Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Dask-friendly nan check in xr.corr() and xr.cov() #5284

Merged
merged 10 commits into from
May 27, 2021
Merged

Dask-friendly nan check in xr.corr() and xr.cov() #5284

Changes from 2 commits
Commits
Show all changes
10 commits
Select commit Hold shift + click to select a range
b8d6daf
initial changes
AndrewILWilliams May 9, 2021
227b622
Using map_blocks to lazily mask input arrays, following #4559
AndrewILWilliams May 16, 2021
6619a6c
Adding lazy corr cov test with `raise_if_dask_computes`
AndrewILWilliams May 26, 2021
637ab80
adding test for one da without nans
AndrewILWilliams May 26, 2021
fff4556
checking ordering of arrays doesnt matter
AndrewILWilliams May 26, 2021
2756488
adjust inputs to test
AndrewILWilliams May 26, 2021
58c441d
add test for array with no missing values
AndrewILWilliams May 26, 2021
a9bd723
added whatsnew
AndrewILWilliams May 27, 2021
8f50b09
Merge branch 'master' into dask_corr_cov
AndrewILWilliams May 27, 2021
ccb5f67
fixing format issues
AndrewILWilliams May 27, 2021
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
19 changes: 15 additions & 4 deletions xarray/core/computation.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1326,12 +1326,23 @@ def _cov_corr(da_a, da_b, dim=None, ddof=0, method=None):

# 2. Ignore the nans
valid_values = da_a.notnull() & da_b.notnull()
valid_count = valid_values.sum(dim) - ddof

if not valid_values.all():
da_a = da_a.where(valid_values)
da_b = da_b.where(valid_values)
def _get_valid_values(da, other):
"""
Function to lazily mask da_a and da_b
following a similar approach to
https://github.com/pydata/xarray/pull/4559
"""
missing_vals = np.logical_or(da.isnull(), other.isnull())
if missing_vals.any():
da = da.where(~missing_vals)
return da
else:
return da

valid_count = valid_values.sum(dim) - ddof
da_a = da_a.map_blocks(_get_valid_values, args=[da_b])
da_b = da_b.map_blocks(_get_valid_values, args=[da_a])

# 3. Detrend along the given dim
demeaned_da_a = da_a - da_a.mean(dim=dim)
Expand Down