In this repository, I have been building an end-to-end machine learning project for stroke prediction with Streamlit, FastAPI.
The dataset of stroke prediction on Kaggle with 11 features and this project includes steps as below:
- Data loading
- Exploratory data analysis
- Data processing including filling null values, outliers handling, categorical data handling, checking correlations, handling imbalanced data and data scaling.
- Modelling and evaluation with Random Forrest, Grid Search and Cross Validation.
- Write and save model by using pickle
- Create an API with FastAPI.
- Create an UI with Streamlit.