To make sure if someone starts using Vantage6, and is not surprised by algorithms not working, we will need full integration tests.
These would comprise of:
1.: setting up a dummy infrastructure
2.: including dummy data
3.: running the algorithms and testing the output
If we would combine this with a list of currently running/active algorithms (including researcher code how to execute it). This would greatly benefit the general adoption.
To make sure if someone starts using Vantage6, and is not surprised by algorithms not working, we will need full integration tests.
These would comprise of: 1.: setting up a dummy infrastructure 2.: including dummy data 3.: running the algorithms and testing the output
If we would combine this with a list of currently running/active algorithms (including researcher code how to execute it). This would greatly benefit the general adoption.