Beginning Anomaly Detection Using

Figure 4-73.  Showing the anomalies relative to the threshold

Figure 4-74.  Model graph shown in TensorBoard

Chapter 4   autoenCoders

---

176

Figure 

4-75

 shows the graph of the model as visualized by TensorBoard.

Figure 

4-76

 shows the plotting of the accuracy during the training process through 

the epochs of training.

Figure 4-75.  Model graph shown in TensorBoard

Figure 4-76.  Plotting of accuracy shown in TensorBoard

Figure 

4-77

 shows the plotting of the loss during the training process through the 

epochs of training.

Chapter 4   autoenCoders

---

177

Figure 

4-78

 shows the plotting of the accuracy of validation during the training 

process through the epochs of training.

Figure 4-77.  Plotting of loss shown in TensorBoard

Figure 

4-79

 shows the plotting of the loss of validation during the training process 

through the epochs of training.

Figure 4-78.  Plotting of validation accuracy shown in TensorBoard

Chapter 4   autoenCoders

---

178

 Summary

In this chapter, we discussed autoencoders, types of autoencoders, and how they can 

be used to build anomaly detection engines. We looked at implementing a simple 

autoencoder and sparse, deep, convolutional, and denoising autoencoders. We also 

explored the variational autoencoder as a means to detect anomalies.

In the next chapter, we will look at another method of anomaly detection,