ABSTRACT
Studies have pointed out that using the features computed with deep nets in machine learning algorithms like Support Vector Machines or Random Forests, offers state-of-the-art results. This chapter provides insight into classification using classic Machine Learning techniques on deep learning features. Deep learning is used merely as feature engineering rather than a sole classifier. In a TensorFlow model, once an operator has a name, it is possible to fetch its result during the inference. In Orfeo ToolBox, composite applications are multiple applications connected together. Typically, one application input can come from another application output, instead of being read from file. The first time: for the classifier training. The second time: during the classification map generation using Train Classifier From Deep Features application. Once the training has been finished, one should have a.yaml file that serializes the random forest classification rule, in a human-readable format.
