Functionality to save the trained model
ShayaNovelis opened this issue · 4 comments
ShayaNovelis commented
Is there a way to save the trained model or its weights when using MirroredStrategyRunner?
marcialf commented
Only chief node can save the model.
alexperry-shifu commented
Is there an example of saving the model when starting with your simple Spark example?
rahulnatarajan98 commented
Are there any updates on this?
tomersagi commented
Only chief node can save the model.
ok, but how do you tell the train function to save the model only on the chief node in the Spark API?