Add partial resampling #202
Conversation
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## main #202 +/- ##
==========================================
+ Coverage 98.25% 98.32% +0.06%
==========================================
Files 12 12
Lines 1263 1313 +50
==========================================
+ Hits 1241 1291 +50
Misses 22 22 ☔ View full report in Codecov by Sentry. |
ph-kev
force-pushed
the
kp/partial-resampled-as
branch
3 times, most recently
from
4d83c41 to
7d400f3
Compare
| If the string or iterable `dim_names` is nothing, then resampling is done over all | ||
| dimensions. Otherwise, resampling is done over the dimensions in `dim_names`. If | ||
| resampling is done over all dimensions, then reordering dimensions is | ||
| automatically done. |
There was a problem hiding this comment.
What names are used for the dimensions?
Are the actual names or the conventional names?
There was a problem hiding this comment.
Let's also add a !!! compat note to specify which version introduced this feature
| to fit the dimensions in `dest_var`. Resampling only over selected dimensions is | ||
| possible with the keyword `dim_names`. Resampling is not possible when the |
There was a problem hiding this comment.
Let's add an entry in the FAQ about this
src/outvar_dimensions.jl
Outdated
| """ | ||
| find_corresponding_dim_name(dim_name::AbstractString, dim_names::Iterable) | ||
|
|
||
| Find the corresponding dimension name in `dim_names` that match `dim_name`. |
There was a problem hiding this comment.
Can you clarify that this is using the internal name bindings / going through the conventioanl names?
| x_long = 0.0:180.0 |> collect | ||
| x_lat = 0.0:90.0 |> collect | ||
| x_data = reshape(1.0:(181 * 91), (91, 181)) | ||
| x_dims = OrderedDict(["lat" => x_lat, "long" => x_long]) | ||
| x_attribs = Dict("long_name" => "hi") | ||
| x_dim_attribs = OrderedDict([ | ||
| "lat" => Dict("units" => "test_units2"), | ||
| "long" => Dict("units" => "test_units1"), | ||
| ]) | ||
| x_var = ClimaAnalysis.OutputVar(x_attribs, x_dims, x_dim_attribs, x_data) |
There was a problem hiding this comment.
Not related to this PR, but we might want to think about adding a Factory or something to create OutputVar and reduce all this boilerplate
The keyword argument `dim_names` allows the caller to only check for a certain dimension as opposed to all of the dimensions in `x` and `y`.
ph-kev
force-pushed
the
kp/partial-resampled-as
branch
from
7d400f3 to
0e63f61
Compare
This PR adds partial resampling by adding the keyword argument
dim_namestoVar.resampled_as.We also add
Var.find_corresponding_dim_namewhich allows one to find the name of the dimension in aOutputVarby using another dimension name.