Image classification using transfer learning with BigDL for Apache Spark

We will demonstrate a very straight-forward implementation of image transfer learning using deep learning methods.

We will apply an image classification algorithm to classify house images according to their architectural style.

The implementation will be done using BigDL for Apache Spark framework. We may also compare-and-contrast it with implementation in Tensorflow (tbd).

We will assume that the audience has a basic knowledge of deep learning and machine learning concepts, such as loss function, model training, model inference, and feed-forward conv. neural networks.

Speaker: Sergey Ermolin, Intel