ComeçarComece de graça

Train and testing transformations (II)

Similar to applying the same scaler to both your training and test sets, if you have removed outliers from the train set, you probably want to do the same on the test set as well. Once again you should ensure that you use the thresholds calculated only from the train set to remove outliers from the test set.

Similar to the last exercise, we split the so_numeric_df DataFrame into train (so_train_numeric) and test (so_test_numeric) sets.

Este exercício faz parte do curso

Feature Engineering for Machine Learning in Python

Ver curso

Instruções do exercício

  • Calculate the standard deviation and mean of the ConvertedSalary column.
  • Calculate the upper and lower bounds as three standard deviations away from the mean in both the directions.
  • Trim the so_test_numeric DataFrame to retain all rows where ConvertedSalary is within the lower and upper bounds.

Exercício interativo prático

Experimente este exercício completando este código de exemplo.

train_std = so_train_numeric['ConvertedSalary'].____
train_mean = so_train_numeric['ConvertedSalary'].____

cut_off = train_std * 3
train_lower, train_upper = ____, train_mean + cut_off

# Trim the test DataFrame
trimmed_df = so_test_numeric[(so_test_numeric['ConvertedSalary'] < ____) \
                             & (so_test_numeric['ConvertedSalary'] > ____)]
Editar e executar o código