When a breaking change occurs, it can break some steps in Q2Pipe, so a preconfigured test environment (including image building, dataset, parameters, etc.) could be a good idea.
It would also include some databases to evaluate if a classifier should be retrained
When a breaking change occurs, it can break some steps in Q2Pipe, so a preconfigured test environment (including image building, dataset, parameters, etc.) could be a good idea.
It would also include some databases to evaluate if a classifier should be retrained