Ouranosinc / xscen / 16534224455 / 5
90%
main: 90%

DEFAULT BRANCH: main
Ran
Files 18
Run time 1s
Badge
Embed ▾
README BADGES
x

If you need to use a raster PNG badge, change the '.svg' to '.png' in the link

Markdown

Textile

RDoc

HTML

Rst
coverage: 84.887% (+0.07%) from 84.814%
16534224455.5

push

github

web-flow 
Iterative creeping - save and load weights (#594)

<!-- Please ensure the PR fulfills the following requirements! -->
<!-- If this is your first PR, make sure to add your details to the
AUTHORS.rst! -->
### Pull Request Checklist:
- [ ] This PR addresses an already opened issue (for bug fixes /
features)
    - This PR fixes #xyz
- [x] (If applicable) Documentation has been added / updated (for bug
fixes / features).
- [x] (If applicable) Tests have been added.
- [x] This PR does not seem to break the templates.
- [x] CHANGELOG.rst has been updated (with summary of main changes).
- [x] Link to issue (:issue:`number`) and pull request (:pull:`number`)
has been added.

### What kind of change does this PR introduce?

* New argument `steps` to `creep_weights` so that we creep iteratively
this number of time, so filling a boundary of thickness `steps * n`. The
`creep_fill` detects it and applies the weights in sequence too.
* New methods to save those weights to disk and then load them back.
Similar to what `xESMF` does, but for our own `creep_fill`. This syntax
is completely xscen-specific. In theory CF conventions have what it
needs to encode/decode sparse arrays, but the implementation is still to
be done.

### Does this PR introduce a breaking change?
No.


### Other information:
I already use that technique in PC for the sea masking. And I now want
to use that for SPS. The idea is that `creep_weights` is slow but
`creep_fill` is fast, so if the mask is constant, you don't want to
recompute it often.


Peut-être que ceci pourrait être utile à @DoMatte.

3898 of 4592 relevant lines covered (84.89%)

0.85 hits per line

Source Files on job run-Python3.10 - 16534224455.5