House Prices

Inspiration

We were both interested in AI/ML and thought it was an interesting problem/dataset.

What it does

Makes a regression model given the data from train.csv, then uses CatBoost to predict the results for test.csv.

How we built it

We found the problem on kaggle and used the built-in Jupyter notebook IDE. We implemented the CatBoost library and used its regressor to create the model.

Challenges we ran into

How to deal with outliers.

Accomplishments that we're proud of

Neither of us had much experience in regression, but we still managed to make it to the top third of the leaderboard on kaggle.

What we learned

We learned to do research for libraries that may be useful to us (in this case, CatBoost).

What's next for House Prices

We would find a way to quantify the categorical (non-numerical) data, so that the model can be improved.

Built With

• python