Parametrizing the model

Now that our data has been prepared for the SGD algorithm, we are ready to set the parameters of our experiment. In a way similar to scientific experimentation, we will want to try out several sets of parameters to test several models and pick up the best one. The next screenshot shows where we actually specify our model parameters:

  • Model memory
  • Data passes
  • Shuffling
  • Regularization

Get Effective Amazon Machine Learning now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.