This project is based on the Titanic - Machine Learning from Disaster competition hosted on Kaggle. The repository contains notebooks, models, and documentation for the project. It includes data preprocessing, model training, evaluation, and visualization steps to build a predictive model. The objective of the competition is to predict if a passenger is able to survive the sinking of the Titanic based on the features defined in the Titanic dataset.
Through this project, my goal is to learn certain aspects of Machine Learning projects and prepare a standard workflow for future projects and implementations. Thus, this project follows a standard workflow which can be summarised as follows:
- Downloading and loading the Titanic dataset into the Jupyter notebook.
- Exploring and visualising the dataset with custom plotting functions.
- Preprocessing the dataset with Scikit-Learn imputers and transformers.
- Training different models on the preprocessed dataset.
- Evaluating and comparing the performance of the trained models.
- Selecting the best model and making final predictions.
- Submitting the final predictions and saving the best model.