This is a machine learning project that predicts the salaries of employees based on various job-related factors. The dataset used contains 12 features, such as years of experience, education level, and job type. The project goes through the process of exploratory data analysis, feature engineering, model training, and hyperparameter tuning to achieve a low mean absolute error (MAE) of 5.72 thousand dollars.
To run this project locally, you will need Python 3 and the following libraries:
- pandas
- numpy
- matplotlib
- seaborn
- scikit-learn To install these libraries, you can use pip. For example: pip install pandas
The main script is salary-prediction.ipynb. You can run this script in a Jupyter Notebook environment, such as Anaconda, or in a Google Colab notebook.
The notebook contains detailed explanations of each step in the machine learning pipeline, including:
- Exploratory data analysis
- Data cleaning and preprocessing
- Feature engineering
- Model selection and training
- Hyperparameter tuning
- Prediction and evaluation The final model is saved in the file salary_model.pkl.