train model using decision tree using platform_os dataframe

notebooks/AbTestDecisionTree.dot

@@ -0,0 +1,57 @@
+digraph Tree {
+node [shape=box, style="filled, rounded", color="black", fontname="helvetica"] ;
+edge [fontname="helvetica"] ;
+0 [label=<hour &le; 20.5<br/>gini = 0.496<br/>samples = 895<br/>value = [486, 409]<br/>class = Aware>, fillcolor="#fbebe0"] ;
+1 [label=<device_make &le; 91.0<br/>gini = 0.495<br/>samples = 855<br/>value = [470, 385]<br/>class = Aware>, fillcolor="#fae8db"] ;
+0 -> 1 [labeldistance=2.5, labelangle=45, headlabel="True"] ;
+2 [label=<device_make &le; 68.5<br/>gini = 0.497<br/>samples = 814<br/>value = [441, 373]<br/>class = Aware>, fillcolor="#fbece0"] ;
+1 -> 2 ;
+3 [label=<hour &le; 12.5<br/>gini = 0.494<br/>samples = 738<br/>value = [410, 328]<br/>class = Aware>, fillcolor="#fae6d7"] ;
+2 -> 3 ;
+4 [label=<gini = 0.485<br/>samples = 374<br/>value = [219, 155]<br/>class = Aware>, fillcolor="#f7dac5"] ;
+3 -> 4 ;
+5 [label=<gini = 0.499<br/>samples = 364<br/>value = [191, 173]<br/>class = Aware>, fillcolor="#fdf3ec"] ;
+3 -> 5 ;
+6 [label=<browser &le; 3.0<br/>gini = 0.483<br/>samples = 76<br/>value = [31, 45]<br/>class = Not Aware>, fillcolor="#c1e1f7"] ;
+2 -> 6 ;
+7 [label=<gini = 0.355<br/>samples = 26<br/>value = [6, 20]<br/>class = Not Aware>, fillcolor="#74baed"] ;
+6 -> 7 ;
+8 [label=<gini = 0.5<br/>samples = 50<br/>value = [25, 25]<br/>class = Aware>, fillcolor="#ffffff"] ;
+6 -> 8 ;
+9 [label=<hour &le; 13.5<br/>gini = 0.414<br/>samples = 41<br/>value = [29, 12]<br/>class = Aware>, fillcolor="#f0b58b"] ;
+1 -> 9 ;
+10 [label=<hour &le; 12.5<br/>gini = 0.488<br/>samples = 26<br/>value = [15, 11]<br/>class = Aware>, fillcolor="#f8ddca"] ;
+9 -> 10 ;
+11 [label=<gini = 0.469<br/>samples = 24<br/>value = [15, 9]<br/>class = Aware>, fillcolor="#f5cdb0"] ;
+10 -> 11 ;
+12 [label=<gini = 0.0<br/>samples = 2<br/>value = [0, 2]<br/>class = Not Aware>, fillcolor="#399de5"] ;
+10 -> 12 ;
+13 [label=<device_make &le; 96.0<br/>gini = 0.124<br/>samples = 15<br/>value = [14, 1]<br/>class = Aware>, fillcolor="#e78a47"] ;
+9 -> 13 ;
+14 [label=<gini = 0.375<br/>samples = 4<br/>value = [3, 1]<br/>class = Aware>, fillcolor="#eeab7b"] ;
+13 -> 14 ;
+15 [label=<gini = 0.0<br/>samples = 11<br/>value = [11, 0]<br/>class = Aware>, fillcolor="#e58139"] ;
+13 -> 15 ;
+16 [label=<date &le; 5.5<br/>gini = 0.48<br/>samples = 40<br/>value = [16, 24]<br/>class = Not Aware>, fillcolor="#bddef6"] ;
+0 -> 16 [labeldistance=2.5, labelangle=-45, headlabel="False"] ;
+17 [label=<hour &le; 22.5<br/>gini = 0.498<br/>samples = 30<br/>value = [14, 16]<br/>class = Not Aware>, fillcolor="#e6f3fc"] ;
+16 -> 17 ;
+18 [label=<device_make &le; 67.0<br/>gini = 0.5<br/>samples = 28<br/>value = [14, 14]<br/>class = Aware>, fillcolor="#ffffff"] ;
+17 -> 18 ;
+19 [label=<gini = 0.497<br/>samples = 26<br/>value = [12, 14]<br/>class = Not Aware>, fillcolor="#e3f1fb"] ;
+18 -> 19 ;
+20 [label=<gini = 0.0<br/>samples = 2<br/>value = [2, 0]<br/>class = Aware>, fillcolor="#e58139"] ;
+18 -> 20 ;
+21 [label=<gini = 0.0<br/>samples = 2<br/>value = [0, 2]<br/>class = Not Aware>, fillcolor="#399de5"] ;
+17 -> 21 ;
+22 [label=<browser &le; 4.5<br/>gini = 0.32<br/>samples = 10<br/>value = [2, 8]<br/>class = Not Aware>, fillcolor="#6ab6ec"] ;
+16 -> 22 ;
+23 [label=<browser &le; 1.5<br/>gini = 0.198<br/>samples = 9<br/>value = [1, 8]<br/>class = Not Aware>, fillcolor="#52a9e8"] ;
+22 -> 23 ;
+24 [label=<gini = 0.32<br/>samples = 5<br/>value = [1, 4]<br/>class = Not Aware>, fillcolor="#6ab6ec"] ;
+23 -> 24 ;
+25 [label=<gini = 0.0<br/>samples = 4<br/>value = [0, 4]<br/>class = Not Aware>, fillcolor="#399de5"] ;
+23 -> 25 ;
+26 [label=<gini = 0.0<br/>samples = 1<br/>value = [1, 0]<br/>class = Aware>, fillcolor="#e58139"] ;
+22 -> 26 ;
+}