This a public phishing site dataset taken from UCI repository.
Download the dataset and save as dataset.arff. The preprocess.py loads the arff file and converts it to numpy array. Then dataset metadata is printed and then dataset is splited into training and testing set with 30% for testing.
Change working directory to /backend/dataset and Run the preprocessor with
python3 preprocess.py
Training and testing data *.npy files are created in the working directory.