Default of Credit Card Clients
project employs MRIs to predict and categorize brain tumor types, utilizing various machine learning models, image preprocessing techniques, and data augmentation
Overview
- Performed EDA to visualize/investigate the dataset, and built classification models using Scikit-learn, PyTorch, and Matplotlib.
- Improved the F1 score of a simple Logistic Regression from 0% to 54.62% with an accuracy of 80.33% by tuning the decision threshold and other hyperparameters of the Extreme Gradient Boosting (XGBoost) model.
- Please visit the project with Google Colab Notebook using this link to my notebook.