Please enable JavaScript.
Coggle requires JavaScript to display documents.
Chapters 17 & 18 (17.1 Evaluate Model Performance Introduction…
Chapters 17 & 18
-
17.3 ROC Curve
ROC Curve screen is where several central measures of model success exist beyond the original optimization metric, LogLoss
the threshold. This is the probability value at which DataRobot changes a prediction from negative to positive
Traffic Light Example
the light is green, and the algorithm has determined that it is green. This is called a true positive (TP)
In the top-left, the light is red, and the algorithm has determined
that it is red. This is called a true negative (TN)
The top-right quadrant represents the case where the traffic light was red, but the algorithm classified it as green. This is a False Positive (FP)
In the opposite quadrant, the bottom-left, the traffic light was green, while the algorithm determined that the light is red. This is a False Negative
-
-
-
-