Julius Busecke
Julius Busecke
The handling of errors during the "metric-matching" step is still not great: 1. The docstrings of `postprocessing.match_metrics` and `postprocessing._parse_metric` need to include all valid options 2. I would like another...
Opening this to follow up on #253. I have so far not really given that big of a priority to the performance of the preprocessing module, but I think it...
The tests running on the actual cloud data take very long and are resource intensive. We currently can deactivate them by adding `[skip-ci]` to the commit message, but I propose...
The CI is currently failing with xgcm v0.7.0. I suspect I am using the deprecated 'extrapolate' padding in here. For now I have pinned xgcm
Name change
The functionality of this package has grown quite a bit over time and I find myself laughing at lines like ``` from cmip6_preprocessing.postprocessing import ... ``` I also think the...
Related to #212, we should for now explicitly recommend to use the older version of natural earth, or even import it within this package. This could be part of a...
I have recently encountered a situation where a time resolved metric was somehow short in the output. I think it would be nice to have a function that can tell...
#201 made me think that it would be nice if the parts of cmip6_pp which need cftime explicitly, check the inputs. This should be mostly restricted to the drift_removal module...
I assume this is due to the fact that is actually needs to compute some of the data for the comparison of missing values, but I wanted to flag this...
This is just a vanity feature request, but it could be nice to have a progressbar for all the 'batch processing' functions that iterate in some form over (possibly large)...