Intro
This week will be mostly about planning and inspirations. We will go together through the schedule, evaluation criteria, talk briefly about the basis of ML and have a look at different examples of ML projects.
Resources
Weekly task
Watch the serie on RNN from NOC2018 by the Coding Train. You can skip the part about ‘Perceptron’ (10.2, 10.3) and ‘Matrix Math’ (10.6 … 10.11). Some helpful links are available at NOC2018 github page and the source code for the toy NN can be downloaded here (slightly different than the one written in the video)
Next week we will play with the XOR, Doodle and Color predictor examples. An online demo of the color predictor is available here together with source code.
The goal is to start understanding how a simple NN work. It’s ok if some parts (even a lot) are unclear for now, we will go through some code example and play around with the toy NN built by Daniel Shiffman together next week. The example below are written in JS, similar examples written in Python are included in ‘Going further’ below.
You can also have a look at the proposed tools and give them a try!