Mlflow logging

-Added Mlflow metrics logging and artifacts logging for logistic regression, decision tree and random forest models.

notebooks/AbTestDecisionTree.dot

@@ -12,11 +12,11 @@ edge [fontname="helvetica"] ;
 3 -> 4 ;
 5 [label=<gini = 0.499<br/>samples = 379<br/>value = [199, 180]<br/>class = Aware>, fillcolor="#fdf3ec"] ;
 3 -> 5 ;
-6 [label=<date &le; 6.5<br/>gini = 0.485<br/>samples = 82<br/>value = [34, 48]<br/>class = Not Aware>, fillcolor="#c5e2f7"] ;
+6 [label=<browser &le; 3.0<br/>gini = 0.485<br/>samples = 82<br/>value = [34, 48]<br/>class = Not Aware>, fillcolor="#c5e2f7"] ;
 2 -> 6 ;
-7 [label=<gini = 0.494<br/>samples = 76<br/>value = [34, 42]<br/>class = Not Aware>, fillcolor="#d9ecfa"] ;
+7 [label=<gini = 0.346<br/>samples = 27<br/>value = [6, 21]<br/>class = Not Aware>, fillcolor="#72b9ec"] ;
 6 -> 7 ;
-8 [label=<gini = 0.0<br/>samples = 6<br/>value = [0, 6]<br/>class = Not Aware>, fillcolor="#399de5"] ;
+8 [label=<gini = 0.5<br/>samples = 55<br/>value = [28, 27]<br/>class = Aware>, fillcolor="#fefaf8"] ;
 6 -> 8 ;
 9 [label=<hour &le; 13.5<br/>gini = 0.408<br/>samples = 42<br/>value = [30, 12]<br/>class = Aware>, fillcolor="#efb388"] ;
 1 -> 9 ;
@@ -36,7 +36,7 @@ edge [fontname="helvetica"] ;
 0 -> 16 [labeldistance=2.5, labelangle=-45, headlabel="False"] ;
 17 [label=<date &le; 4.5<br/>gini = 0.388<br/>samples = 19<br/>value = [5, 14]<br/>class = Not Aware>, fillcolor="#80c0ee"] ;
 16 -> 17 ;
-18 [label=<device_make &le; 33.5<br/>gini = 0.5<br/>samples = 6<br/>value = [3, 3]<br/>class = Aware>, fillcolor="#ffffff"] ;
+18 [label=<browser &le; 2.5<br/>gini = 0.5<br/>samples = 6<br/>value = [3, 3]<br/>class = Aware>, fillcolor="#ffffff"] ;
 17 -> 18 ;
 19 [label=<gini = 0.48<br/>samples = 5<br/>value = [2, 3]<br/>class = Not Aware>, fillcolor="#bddef6"] ;
 18 -> 19 ;
@@ -48,18 +48,18 @@ edge [fontname="helvetica"] ;
 21 -> 22 ;
 23 [label=<gini = 0.5<br/>samples = 2<br/>value = [1, 1]<br/>class = Aware>, fillcolor="#ffffff"] ;
 21 -> 23 ;
-24 [label=<date &le; 4.5<br/>gini = 0.497<br/>samples = 24<br/>value = [11, 13]<br/>class = Not Aware>, fillcolor="#e1f0fb"] ;
+24 [label=<browser &le; 3.0<br/>gini = 0.497<br/>samples = 24<br/>value = [11, 13]<br/>class = Not Aware>, fillcolor="#e1f0fb"] ;
 16 -> 24 ;
-25 [label=<device_make &le; 46.5<br/>gini = 0.408<br/>samples = 7<br/>value = [2, 5]<br/>class = Not Aware>, fillcolor="#88c4ef"] ;
+25 [label=<device_make &le; 52.0<br/>gini = 0.48<br/>samples = 20<br/>value = [8, 12]<br/>class = Not Aware>, fillcolor="#bddef6"] ;
 24 -> 25 ;
-26 [label=<gini = 0.278<br/>samples = 6<br/>value = [1, 5]<br/>class = Not Aware>, fillcolor="#61b1ea"] ;
+26 [label=<gini = 0.5<br/>samples = 16<br/>value = [8, 8]<br/>class = Aware>, fillcolor="#ffffff"] ;
 25 -> 26 ;
-27 [label=<gini = 0.0<br/>samples = 1<br/>value = [1, 0]<br/>class = Aware>, fillcolor="#e58139"] ;
+27 [label=<gini = 0.0<br/>samples = 4<br/>value = [0, 4]<br/>class = Not Aware>, fillcolor="#399de5"] ;
 25 -> 27 ;
-28 [label=<date &le; 5.5<br/>gini = 0.498<br/>samples = 17<br/>value = [9, 8]<br/>class = Aware>, fillcolor="#fcf1e9"] ;
+28 [label=<device_make &le; 54.0<br/>gini = 0.375<br/>samples = 4<br/>value = [3, 1]<br/>class = Aware>, fillcolor="#eeab7b"] ;
 24 -> 28 ;
-29 [label=<gini = 0.42<br/>samples = 10<br/>value = [7, 3]<br/>class = Aware>, fillcolor="#f0b78e"] ;
+29 [label=<gini = 0.0<br/>samples = 1<br/>value = [0, 1]<br/>class = Not Aware>, fillcolor="#399de5"] ;
 28 -> 29 ;
-30 [label=<gini = 0.408<br/>samples = 7<br/>value = [2, 5]<br/>class = Not Aware>, fillcolor="#88c4ef"] ;
+30 [label=<gini = 0.0<br/>samples = 3<br/>value = [3, 0]<br/>class = Aware>, fillcolor="#e58139"] ;
 28 -> 30 ;
 }

notebooks/Decision Tree Classifier reports/report.txt

@@ -0,0 +1,8 @@
+              precision    recall  f1-score   support
+
+           0       0.89      0.53      0.67       109
+           1       0.15      0.56      0.24        16
+
+    accuracy                           0.54       125
+   macro avg       0.52      0.55      0.45       125
+weighted avg       0.80      0.54      0.61       125

notebooks/Logistic Regression reports/report.txt

@@ -0,0 +1,8 @@
+              precision    recall  f1-score   support
+
+           0       0.91      0.54      0.68       109
+           1       0.17      0.62      0.26        16
+
+    accuracy                           0.55       125
+   macro avg       0.54      0.58      0.47       125
+weighted avg       0.81      0.55      0.63       125