search

OpenSenseAction / OPENSENSE_sandbox

Collection of runable examples with software packages for processing opportunistic rainfall sensors
BSD 3-Clause "New" or "Revised" License
13 stars 16 forks source link

fix `sublink` dimension name in pynncml_basic_example.ipynb #70

Open cchwala opened 1 year ago

cchwala commented 1 year ago

There were changes in the data transformation code, which now only returns a xarray.Dataset with reshaped data and updated dimension (cml_id, sublink_id, time) if it is called with the option to reshape the data. Since reshaping the data takes too long to run (an might crash on binder due to limited RAM) the simplest fix is to not reshape and just use the sublink dimension as it is called in the existing OpenMRG NetCDF file.

At some point, the pynncml example could use the smaller example NetCDF from the OpenMRG dataset that is currently (since #62) part of the sandbox repository.

github-actions[bot] commented 1 year ago

Binder :point_left: Launch a binder notebook on branch _OpenSenseAction/OPENSENSE_sandbox/pynncml_sublinkkwarg

cchwala commented 1 year ago

Note: In the last commit it did what I suggested above:

At some point, the pynncml example could use the smaller example NetCDF from the OpenMRG dataset that is currently (since https://github.com/OpenSenseAction/OPENSENSE_sandbox/pull/62) part of the sandbox repository.

The problem is that the final rain rates estimations does not seem to work, maybe because the time series is too short now.

@haihabi Can you have a look at the notebook? Maybe you can fiddle with it directly in binder based on the link above to find out what is going wrong. If the reason is that the time series is too short, we will have to switch back to downloading the full OpenMRG dataset just using the fix from my first commit in this PR.

I tried to fix this now, because it would be great to have your notebook working when the sandbox is promoted during the training school in October.