What it does
Predicts the revenue produced by users in the following week when they install an app.
How we built it
Using python in VSCode, with pyTorch and other tools.
Challenges we ran into
Pre-processing, memory issues and a very sparse dataset.
Accomplishments that we're proud of
We got to first of the leaderboard after a long night of working in a long pipeline that led to our final model and preprocessing.
What we learned
Using dask as a data loader to deal with a way bigger datasets than we could otherwise.
Log in or sign up for Devpost to join the conversation.