That is my first ML project!
It is supposed to predict the salary that people are making so the ml charity could ask those people for donations using some given features.
I used some python libires for that project: Numpy, Pandas, Sklearn.
The stages of that project were:
1-Exploring the given data.
2-Transforming Skewed Continuous Features.
3-Data Preprocessing: I implemented one-hot encoding on some features.
4-shuffling and splitting data.
5-Choosing the metrics and a Naive Predictor.
6-I chose three of the supervised learning models available on Sklearn - Decision Tree, SVC and Gradient Boosting - and tested them and Gradient Boosting was the best one.
7-Finally I Extract the feature importances using .feature_importances_ after training the model using the whole data not only the training set
Aml-Hassan-Abd-El-hamid/Finding-Donors-for-CharityML
Predicting salary of the people based on various data about them
Jupyter Notebook