Beginning Anomaly Detection Using

kernel_initializer: An initializer for the weight matrix. For more 

information, check out the 

Initializers section.

• 

bias_initializer: Similar to the kernel_initializer, but for the bias.

Appendix A   intro to KerAs

---

330

• 

kernel_regularizer: A regularizer function that’s been applied to the 

weight matrix. For more information, check out the 

Regularizers 

section.

• 

bias_regularizer: Regularizer function applied to the bias.

• 

activity_regularizer: Regularizer function applied to the output of 

the layer.

• 

kernel_constraint: A constraint function applied to the weights. For 

more information, check out the 

Constraints section.

• 

bias_constraint: A constraint function applied to the bias.

For a better idea of what a dense layer is, check out Figure 

A-12

.

Input

Data

Input Layer

Dense Layer 1

Dense Layer 2

Output Layer

(Dense)

Figure A-12.  Dense layers in an artificial neural network

Appendix A   intro to KerAs

---

331

 Activation

keras.layers.Activation()

This layer applies an activation function to the input. Here is the argument:

•