DL4J implementation

The second implementation of object recognition we are going to do is in Scala and involves the DL4J framework. To train and evaluate the model, we are still going to use the CIFAR-10 dataset. The dependencies for this project are a DataVec data image, DL4J, NN, and ND4J, plus Guava 19.0 and Apache commons math 3.4.

If you look at the CIFAR-10 dataset download page (see the following screenshot), you can see that there are specific archives available for the Python, MatLab, and C programming languages, but not for Scala or Java:

Figure 13.11: The CIFAR-10 dataset download page

There's no need to separately download and ...

Get Hands-On Deep Learning with Apache Spark now with O’Reilly online learning.

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