Summary
In this final chapter, we saw some use cases with neural networks and deep learning. This should form the basis of your future work on neural networks. The usage is common in most cases, with changes in the dataset involved for the model during training and testing.
We saw the following examples in this chapter:
- Integrating TensorFlow and Keras with R, which opens up vast set of use cases to be built using R
- Building a digit recognizer through classification using H2O
- Understanding the LSTM function with MxNet
- PCA using H2O
- Building an autoencoder using H2O
- Usage of
darchfor classification problems
R is a very flexible and a major statistical programming language for data scientists across the world. A grasp of neural networks with R will help the community evolve further and increase the usage of R for deep learning and newer use cases.
