2. Neural Networks

Activity 2.01: Build a Multilayer Neural Network to Classify Sonar Signals

Solution

Let's see how the solution looks. Remember—this is one solution, but there could be many variations:

1. Import all the required libraries:

   import tensorflow as tf
   import pandas as pd
   from sklearn.preprocessing import LabelEncoder
   # Import Keras libraries
   from tensorflow.keras.models import Sequential
   from tensorflow.keras.layers import Dense

2. Load and examine the data:

   df = pd.read_csv('sonar.csv')
   df.head()

   The output is:

   Figure 2.37: Contents of sonar.csv

   Observe that there are 60 features, and the target has two values—Rock and Mine.

   This means that this is a binary classification problem. Let's prepare the data before we build the neural network.

3. Separate the features and the labels:

   X_input = df.iloc[:, :-1]
   Y_label = df['Class'].values

   In this code, X_input is selecting all the rows of all the columns except the Class...