Salt2 fits were performed in Dataflow prior to #94. Port the code into a new Cloud Function so that we retain the option of running them.
I think you should keep SALT2 as a thing you can run. In general, we will want to be able to run various classifiers/fitters, SALT2 is one of them, SuperNNova, Upsilon, etc. These are all just functions that take as input the alert data and produce a classification, score, fit parameters as outputs. They should structurally fit in in the same way in the pipeline.
If what you're saying is that you should remove the Dataflow job that runs SALT2, then of course, yes. But we will want to be able to run SALT2 on the input data.
Salt2 fits were performed in Dataflow prior to #94. Port the code into a new Cloud Function so that we retain the option of running them.