Please enable JavaScript.
Coggle requires JavaScript to display documents.
Chapter 17 & 18 (18 Comparing Model Pairs (18.2 Prioritizing Model…
Chapter 17 & 18
18 Comparing Model Pairs
18.1 Model Comparison
Model comparison
Change model
Lift
Dual lift
compute data
Cross Validation
Number of bins
18.2 Prioritizing Model Criteria
Predictive Accuracy
Prediction Speed
Speed to build model
Familiarity w/ model
Insights
Chapter 17 Evaluate Model Performance
17.1 Intro
AutoML:understandig and learnig
Priori
Sorted by cross validation
eXtreme Gradient Bootsed Trees (xGboost)
17.2 Sample Algorithm and Model
decision Tree Classifier
tree-based algorithms build on the logic of the decision-tree classifier
conceptual simplicity
go to the bottom of the model leaderboard
discharge_disposition_id, number_diagnoses, and number_inpatient
Find the most predictive feature (the one that best explains the target) and place it at the root of the tree
Split the feature into two groups at the point of the feature where the two groups are as homogenous as possible
Repeat step 2 for each new branch (box
17.3 ROC Curve
Click on the ROC Curve
look now to the “Prediction Distribution"
determined a threshold
quadrant and compare
Confusion Matrix
True Positive Rate (TPR)
Positive Predictive Value
True Positive Rate, False Positive Rate, True Negative Rate, Positive Predictive Value, and Negative Predictive Value
1 more item...
17.4 Using Lift Chart
Blueprint Lift Chart
A hospital may want to examine the dollar value
Enable drill down
Optional Feature
All data
Download