I have ran an AutoML forecasting job to predict sales. The training job modeled 69 models (from trial models) and 9 of them were selected to create an ensemble model as final model.
Q1: I want to evaluate and compare all trial models that autoML performed in the training, is it possible?
Q2: From the logs, I have all 69 models with their hyperparameters, how do I recreate all these model and get another predictions over it?
Thank you for your interest and please let me know if you have more questions.
