adding all work done so far (lessons 1 - 5)

python/Supervised Learning/Decision Trees/16. Calculating information gain on a dataset/calculate.py

@@ -0,0 +1,13 @@
+import numpy as np
+
+
+def two_group_ent(first, tot):
+    return -(first / tot * np.log2(first / tot) +
+             (tot - first) / tot * np.log2((tot - first) / tot))
+
+
+tot_ent = two_group_ent(10, 24)
+g17_ent = 15 / 24 * two_group_ent(11, 15) + 9 / 24 * two_group_ent(6, 9)
+
+answer = tot_ent - g17_ent
+print(answer)

python/Supervised Learning/Decision Trees/16. Calculating information gain on a dataset/ml-bugs.csv

@@ -0,0 +1,25 @@
+Species,Color,Length (mm)
+Mobug,Brown,11.6
+Mobug,Blue,16.3
+Lobug,Blue,15.1
+Lobug,Green,23.7
+Lobug,Blue,18.4
+Lobug,Brown,17.1
+Mobug,Brown,15.7
+Lobug,Green,18.6
+Lobug,Blue,22.9
+Lobug,Blue,21.0
+Lobug,Blue,20.5
+Mobug,Green,21.2
+Mobug,Brown,13.8
+Lobug,Blue,14.5
+Lobug,Green,24.8
+Mobug,Brown,18.2
+Lobug,Green,17.9
+Lobug,Green,22.7
+Mobug,Green,19.9
+Mobug,Blue,14.6
+Mobug,Blue,19.2
+Lobug,Brown,14.1
+Lobug,Green,18.8
+Mobug,Blue,13.1