TensorFlow is a state-of-the-art machine learning framework that specializes in the ability to develop deep learning neural networks. And now, it's available in R! This course will walk you through the basics of using TensorFlow in R. From simple linear regressions to more complex deep learning neural networks (which perform extremely well with BIG datasets) , you'll be introduced to both the basics of TensorFlow and higher-level APIs such as Keras and TFEstimators. We'll put your new-found skills to the test by exploring whether there is a predictable relationship between beer consumption and weather, and find out if we can accurately build a deep neural network to help predict whether a banknote is forged or genuine based on image data.
Let's get you started in TensorFlow! To begin the course, you'll learn the history of the program and will become comfortable using TensorFlow syntax. You'll become versed in TensorFlow constants, placeholders, and Variables and we'll explore some dataflow diagrams using TensorBoard, the TensorFlow visualization tool! This chapter is a great start to get you comfortable with using TensorFlow in R.
Have you ever wondered if we can predict beer consumption in university towns based on weather factors, such as temperature, precipitation, or time of week? Well then do I have a chapter for you! In this chapter, we'll explore linear regression models using both the Core TensorFlow API, as well as the Estimators API (a high-level API with canned models set to speed up the user experience). We'll train and evaluate several models to get you familiar with all the APIs TensorFlow has to offer. And you'll finally be able to answer - do people drink less beer when it's rainy out?
Let’s dive into some deep learning with TensorFlow! In this chapter, you’ll create a complete end-to-end DNN Classifier with the Keras API, exploring if you can predict online customer buy/don’t buy behaviour. Want to see behind-the-scenes of your classifier? TensorBoard is your answer. You’ll explore scalars and graphs in TensorBoard and take a closer look at the visualizations directly available in R. Finish off the chapter with a glimpse into using a canned DNN Classifier in Estimators. Which is the better API for your model? You decide!
Now that you've successfully created your first DNN models using TensorFlow in R, it's time to branch out and look at some ways to increase the accuracy of your models. In this chapter, you'll explore a few regularization techniques, including incorporating a Ridge Regression into a Keras model and adding a Dropout technique to an Estimators canned DNN. Finally, we'll wrap up this course by summarizing all the concepts you've learned, and give you some research ideas for you to try on your own!
PrerequisitesMachine Learning with caret in R
“I've used other sites—Coursera, Udacity, things like that—but DataCamp's been the one that I've stuck with.”
Devon Edwards Joseph
Lloyds Banking Group
“DataCamp is the top resource I recommend for learning data science.”
Harvard Business School
“DataCamp is by far my favorite website to learn from.”
Decision Science Analytics, USAA