GitHub - wangzining/Housing-Price: Predict the price of a house based on its attributes

House Price

Group Project by Wenxuan Wang, Wenda Zheng, Zining Wang

This project investigate how different features can affect house prices.

Since we used Jupyter, we directly exported the files as a pdf file

To run: 0. install Python 2.7(may require environmental variable setups)

install annaconda that comes with python packages
Up to the version of annaconda you may need to update some packages, including sklearn(scikit-learn) and seaborn by doing: conda update scikit-learn
open jupyter inside of annaconda
open combined.ipynb
to run each block, select that block and do SHIFT + ENTER

Alternatively, we provide a regular file that has extension of .py. This could be run in terminal with python combined.py

We added xgboost+lasso(0.13013,baseline).ipynb file and generated test results in price_sol.csv. This final model gives us better prediction.

Name		Name	Last commit message	Last commit date
Latest commit History 32 Commits
.ipynb_checkpoints		.ipynb_checkpoints
data		data
old version		old version
proposal		proposal
submissions		submissions
FinalProjectReport.pdf		FinalProjectReport.pdf
Lasso new.ipynb		Lasso new.ipynb
Lasso.ipynb		Lasso.ipynb
Presentation.pdf		Presentation.pdf
Presentation.pptx		Presentation.pptx
README.md		README.md
TestSetSol.csv		TestSetSol.csv
combined.ipynb		combined.ipynb
combined.py		combined.py
combined_view.pdf		combined_view.pdf
heatmap.png		heatmap.png
lasso_sol.csv		lasso_sol.csv
lasso_sol2.csv		lasso_sol2.csv
pairwise top10.png		pairwise top10.png
pairwise top15.png		pairwise top15.png
pairwise top5.png		pairwise top5.png
price_sol.csv		price_sol.csv
primal.ipynb		primal.ipynb
view.ipynb		view.ipynb
xgboost+lasso(0.13013,baseline).ipynb		xgboost+lasso(0.13013,baseline).ipynb