Feature-Engineering
The functions we created are included in a script. The necessary parts for pre-processing were taken. Analysis complete.
Business Problem
/Required for a machine learning pipeline data preprocessing and variable engineering script needs to be prepared.
/When the dataset is passed through this script, the modeling starts. Expected to be ready.
Dataset Story
*The data set is the data set of the people who were in the Titanic shipwreck.
*It consists of 768 observations and 12 variables.
The target variable is specified as “Survived”;
1: one’s survival,
0: indicates the person’s inability to survive.
Variables
-
Survived
-
0 Died, 1 Survived
-
Pclass – Ticket Class
-
1 = Grade 1, 2 = Grade 2, 3 = Grade 3
-
Age – Age
-
Sibsp – Number of siblings / spouses on the Titanic
-
Sex – Gender
-
Parch – Number of parents/children on Titanic
-
Embarked: – Passenger embarkation port
-
(C = Cherbourg, Q = Queenstown, S = Southampton
-
Fare – Ticket fare
-
Cabin: Cabin number
Project Tasks
1- Open a directory called helpers in the working directory and enter it.
Add a script named data_prep.py.
In the Feature Engineering section, all of our own
collect functions into this script.
Functions that should be here:
▪ outlier_thresholds
▪ replace_with_thresholds
▪ check_outlier
▪ grab_outliers
▪ remove_outlier
▪ missing_values_table
▪ missing_vs_target
▪ label_encoder
▪ one_hot_encoder
▪ rare_analyser
▪ rare_encoder
2- Write a function called titanic_data_prep.
Data preprocessing or EDA functions required for this function,
Get it from the eda.py and data_prep.py files in the helpers.
3- Save the data set you preprocessed to the disk with pickle.
