Transfer Learning for Image Classification

In Chapter 3, Multi-Label Image Classification using Convolutional Neural Networks, we saw how to develop an end-to-end project for handling multi-label image classification problems using CNN based on Java and the Deeplearning4J (DL4J) framework on real Yelp image datasets. For that purpose, we developed a CNN model from scratch.

Unfortunately, developing such a model from scratch is very time consuming and requires a significant amount of computational resources. Secondly, sometimes, we may not even have enough data to train such deep networks. For example, ImageNet is one of the largest image datasets at the moment and has millions of labeled images.

Therefore, we will develop an end-to-end project ...

Get Java Deep Learning Projects now with O’Reilly online learning.

O’Reilly members experience live online training, plus books, videos, and digital content from 200+ publishers.