This project is based on the famous "California Housing Data-set". The project involved preprocessing on the data, using pipeline for various algorithms and then performing a acomparative study of different algorithms. We have used standard functions for pipelines and adder functions.
We have applied algorithms on the data-set, Linear Regression, Ridge Regression, Neural Networks, Random Forrest and Gradient Boosting.
Gradient Boosting outperformed every other algorithm we used.
This file represents all the work, includes are trials as well: https://github.com/divyanshujhawar/MachineLearning/blob/master/Housing%20Value%20Prediction/Housing.ipynb
This file is systematic representation of results: https://github.com/divyanshujhawar/MachineLearning/blob/master/Housing%20Value%20Prediction/Small%20version.ipynb
Thank to Aurélien Geron for providing the guidance to get hands-on-experience with ML and the dataset. {https://github.com/ageron}