dataset_division.py

# Read in data
spam_collection = pd.read_csv('SMSSpamCollection', sep='\t', header=None, names=['Label', 'SMS'])

# Randomize the data set
randomized_collection = spam_collection.sample(frac=1, random_state=3)

# Calculate index for the split-up
training_test_index = round(len(randomized_collection) * 0.8)

# Training/Test split-up
training_set = randomized_collection[:training_test_index].reset_index(drop=True)
test_set = randomized_collection[training_test_index:].reset_index(drop=True)