mic1104/housing-prices

Kaggle Competition - House Prices: Advanced Regression Techniques

House Prices: Advanced Regression Techniques

[ https://www.kaggle.com/c/house-prices-advanced-regression-techniques/leaderboard ].

In this project we signed up for the Kaggle competition. The aim was to complete the competition and submit the code for rank. Then improve the code for better analysis to improve the rank. The competition we entered was “House Prices: Advanced Regression Techniques”. The competition details can be found on site: https://www.kaggle.com/c/house-prices-advanced-regression-techniques. We implemented different advanced regression algorithms and compared the results for best prediction.

Following are the details for the project implementations: Dataset: Provided by Kaggle and in known as Ames Housing Dataset Data Mining Tool: Python scikit library. Analysis & Prediction: Prediction of the sale price of the houses Algorithms: The following algorithms were implemented in the project: Advanced Regression Techniques like LASSO, XgBoost, PCA etc.

The system requirement for Python SKLearn have no any minimal specification. Since data analysis is a computationally intensive task—the better your hardware, the better your experience. Also, the memory should be enough to handle big data sets.

The installation steps of SKLearn, XGBoost is given on the site: http://scikitlearn.org/stable/install.html and http://xgboost.readthedocs.io/en/latest/python/python_intro.html.

Other installation required to support the scikit learn library are: Python (>= 2.6 or >= 3.3) - Python version of project is version 2.6 NumPy (>= 1.6.1) SciPy (>= 0.9) Note: The Python version required is 2.6. If version is different and libraries installed are for different version, code will fail to execute. For correct installation please check online instructions.

The steps to run the code are as follows: Install Python 2.6 and sklearn library, sciPy, numPy, matplotlib and pandas Download the dataset form here and save it on local drive.

Run the file Housing.py with train.csv and test.csv as arguments. To observe the results of individual algorithms please comment out the functions individually defined in main() inside Housing.py.