Designing a neural network means creating the right architecture to achieve optimum results. This optimum, more than often, is 'vague' as this depends on the balance of model performance and computational expenses required to train the model and predict. However, even with this loosely defined term 'optimum', to begin with ...

To demonstrate the automated feature selection methods in Python we would use the diabetes dataset. Import the diabetes .csv file into a data-frame with Pandas as below:  

import pandas as pd
names = ['preg', 'plas', 'pres', 'skin', 'test', 'mass', 'pedi', 'age', 'class']
df_diabetes = pd.read_csv('diabetes.csv', names = names)
df_diabetes ...