Machine Learning project
A try to improve the ML model of Kevin Markham (DataSchool Teacher & Founder)
- Dataset : Pima Indian Diabetes Study
- Dataset info : https://goo.gl/p8ocBn
Tools :
- Python 3.5.1 | Anaconda 2.4.1
- Jupyter Notebook 4.0.6
- SciKitLearn 0.17
- Pandas 0.17.1
- Matplotlib 1.5.0
- Numpy 1.10.4
The model try to predict the risk of diabetes for an out of sample patient
note: Result = 85.3% on ROC/AUC evaluation score // Null Accuracy = 67.7%