Programming Machine Learning

Tracing a Boundary

If we hope to understand classification intuitively, then we need a dataset that we can visualize easily. MNIST, with its mind-boggling hundreds of dimensions, is way too complex for that. Instead, we’ll use a simpler, brain-friendly dataset:

12_classifiers/linearly_separable.txt

Input_A          Input_B          Label
-0.470680718301  -1.905835436960  1
0.9952553595720  1.4019246363100  0
-0.903484238413  -1.233058043620  1
-1.775876322450  -0.436802254656  1

Those are just the first few lines. The file contains 300 examples in total, each with two input variables and a binary label. I wrote a program to plot these data, that you can find in the book’s source code as usual. (It’s called plot_data.py.) It uses the two input variables as coordinates, ...