Dropping features that are redundant

From the dataset seen previously, there are a few columns that are redundant to the machine learning process:

• nameOrig: This column is a unique identifier that belongs to each customer. Since each identifier is unique with every row of the dataset, the machine learning algorithm will not be able to discern any patterns from this feature. 
• nameDest: This column is also a unique identifier that belongs to each customer and as such provides no value to the machine learning algorithm. 
• isFlaggedFraud: This column flags a transaction as fraudulent if a person tries to transfer more than 200,000 in a single transaction. Since we already have a feature called isFraud that flags a transaction as fraud, this feature becomes redundant. 

We can drop these features from the dataset by using the following code: 

#Dropping the redundant features

df = df.drop(['nameOrig', 'nameDest', 'isFlaggedFraud'], axis = 1)