This repository contains the code that has been written for the Kaggle competition "Titanic - Machine Learning from Disaster". The goal is to use machine learning to create a model that predicts which passengers survived the Titanic shipwreck. The focus lies on two important concepts: Correct training in connection to Data Leakage and correct Evaluation in connection to Train Set, Validation Set and Test Set. For more information see the notebook itself.
The following techniques were used in the project:
- Data analysis and visualizations using pandas, matplotlib and seaborn
- Feature engineering using pandas
- Use of pipelines and transformers to prevent data leakage using sklearn
- Random Forest and Gradient Boosting models using sklearn
- Hyperparametertuning using Cross-Fold-Validation using sklearn
Kaggle Profile: https://www.kaggle.com/fynnlinusklling