Open torronen opened 1 year ago
@torronen great feedbacks !!! @luisquintanilla for visibility.
Let's surface the information of what epoch the training is on inside of Model Builder.
@michaelgsharp Is there anything else the framework surfaces that we can use to give better timing estimates?
I was running Text Classification on my main desktop machine, just to test. It took a bit longer than expected, and the computer crashed after 3 weeks.
I think, just number of epochs completed and remaining to train would be enough. Estimated time to completion is probably not needed.
Perhaps the number of max epochs could also be user adjustable in some future version. Likewise, if it could autosave checkpoint and knew how to continue next time (or even guide for advanced cases to recover the file) would be nice.