It would be nice if the template codes would use the mocking-test versions of the Data and Simulations, as these are much faster to download than the whole Databank and full simulations.
This mocking-version could be the default, so the user would do their basic code development in this "sandbox" setting, but then once the code works, could easily switch to using the actual Databank.
So something like this:
!git clone https://github.com/NMRlipids/Databank
%cd Databank
#!git submodule update --init #-- Would download the data full, uncomment when switch to actual
!sed -i '/numpy/s/^/# /' Scripts/DatabankLib/requirements.txt
!pip3 install .
os.environ["NMLDB_ROOT_PATH"] = "/content/Databank"
os.environ["NMLDB_DATA_PATH"] = "/content/Databank/Scripts/tests/Data" #-- Comment when switch to actual
os.environ["NMLDB_SIMU_PATH"] = "/content/Databank/Scripts/tests/Data/Simulations.2" #-- With precomputed Data, comment when switch to actual
It would be nice if the template codes would use the mocking-test versions of the Data and Simulations, as these are much faster to download than the whole Databank and full simulations.
This mocking-version could be the default, so the user would do their basic code development in this "sandbox" setting, but then once the code works, could easily switch to using the actual Databank.
So something like this: