This repo will be made public under https://github.com/LDSSA/hackathon-3-challenge/ on the day of the hackathon. Until then, the instructors can iterate on it here.
hackathon-3-challenge
In this repository you have the following:
- A requirements.txt (this will be removed for the students)
- A
datadirectory that contains directories numbered 0-9. Each of these directories contains features that can be used to (maybe) increase the predictive power of your model. - A notebook
challenge.ipynbthat you can use to do the challenge.
Note that since each of the directories contains a different subset of
features, you will will also need to subset the features in X_test when
making predictions.
Submission
You can upload your submissions here
Your submission file will need to have two columns
id- this is exactly what it sounds likeviolent_crimes_per_pop- the relative portion of crimes per population
The y_test is available here so you can test your r2 score. Of course this will be via the leaderboard on the day of the actual hackathon.