This workshop is introductory and open to everyone assuming basic R/Data Science skills. Please note, the workshop is very hands-on oriented, so expect to get your fingers dirty! The aim will be an introduction to ANNs in R. ANNs form the basic unit of deep learning and are immensely powerful in predictive modeling, but not without pitfalls. In this workshop, we will be working with conceptually understanding what an ANN is, how we train an ANN and how predictions are subsequently made. We will also touch upon parameters, hyper-parameters and how to handle data all in context of model over-fitting. All of the aforementioned will be done using TensorFlow via Keras for R.
Your workshop host for the day will be Leon Eyrich Jessen
Rule number one for the exercises - Do read every line written, it is there for a reason and also have fun!
-
09.00 - 09.15 / 15.00 - 15.15 (15min) Brief talk: Introduction to Artificial Neural Networks
-
09.15 - 09.40 / 15.15 - 15.40 (25min) Exercise: Prototyping an ANN in R
-
09.40 - 09.45 / 15.40 - 15.45 (5min) Brief talk: Introduction to TensorFlow/Keras in R 1
-
09.45 - 10.05 / 15.45 - 16.05 (20min) Exercise: TensorFlow Playground
-
10.05 - 10.20 / 16.05 - 16.20 (15min) Brief talk: Introduction to TensorFlow/Keras in R 2
-
10.20 - 10.30 / 16.20 - 16.30 (10min) Brief talk: Session 1 Summary and Q&A
-
10.30 - 10.50 / 16.30 - 16.50 (20min) Coffee/bio Break
-
10.50 - 11.20 / 16.50 - 17.20 (30min) Exercise: Hello Keras (Classification)
-
11.20 - 11.35 / 17.20 - 17.35 (15min) Brief talk: A bit more on Keras
-
11.35 - 12.05 / 17.35 - 18.05 (30min) Exercise: Now, you must choose between:
-
12.05 - 12.35 / 18.05 - 18.35 (30min) A Case Story
-
12.35 - 12.50 / 18.35 - 18.50 (15min) Brief talk: Session 2 Summary
-
12.50 - 13.00 / 18.50 - 19.00 (10min) Workshop Q&A
(Please note the html preview currently seem not be functioning. All
slides are available under ‘Talks’ in your RStudio session Files pane.
Left-click on the .html file and choose ‘View in Web Browser’)