Beginning Anomaly Detection Using

Figure 4-69.  Code to evaluate the model

Figure 4-70.  Code to predict the anomalies based on the threshold

Figure 4-71.  Code to calculate AUC

Compute the AUC (Area Under the Curve 0.0 to 1.0); it comes up as 0.93, which is 

very high. Figure 

4-71

 shows the code to calculate the AUC.

You can now visualize the confusion matrix to see how well you did with the model. 

Figure 

4-72

 shows the code to show the confusion matrix.

Chapter 4   autoenCoders

---

174

Using the predictions of the labels (normal or anomaly) you can plot the anomalies 

in comparison to the normal data points. Figure 

4-73

 shows the anomalies relative to the 

threshold.

Figure 4-72.  Code to show the confusion matrix

Chapter 4   autoenCoders

---

175

Figure 

4-74

 shows the graph of the model as visualized by TensorBoard.