All datasets located in Titanic directory.
pip install -r requirements.txt
- Filled in missing data with the average value in specified columns
- Dropped cabin and ticket columns.
- For names, sorted them according to titles, eg.Mr. and Mrs. had 'low' survival rates and Major. and Col. had higher survial chances.
- Used Logistic Regression model and a KFold alogrithm to split the model for training.
- The score_model function trains the model and displays the scores.
- After training model is saved in titanic_models directory using the pickle library.
1 for Survived, 0 for did not Survive