Apache Spark Deep Learning Cookbook

How it works...

This section explains the methodology of transfer learning:

1. In early chapters, we discuss how machine learning models, and more importantly deep learning models, work best with larger samples for training purposes. In fact, the general motto with deep learning is the more the merrier.
2. However, there are situations when a high volume of data or images is just not available to train a model. It is in these circumstances where we wish to transfer the learning of one field to predict the outcome of a different field. The heavy lifting of extracting features and filtering through layers and layers within a convolutional neural network have already been performed by institutions that have developed many pre-trained models such ...