COMS4771-Spring-2022-Regression-Competition

Comptetition link - https://www.kaggle.com/competitions/coms4771-spring-2022-regression-competition/leaderboard

Username	`vt2353`
Rank	1/167

Large-scale regressor for predicting trip duration for an Uber-esque transportation service.

It uses a deep neural network to regress on features obtained post feature-selection for predicting trip duration in seconds.

The neural network minimizes L1 loss and uses Adam optimizer.

The training phase includes validating the model to find the best epoch based on validation loss.

Setup Instructions

Move into top-level directory

cd CS4771-Spring-2022-Regression-Competition

Install environment

conda env create -f environment.yml

Activate environment

conda activate dnn

Install package

pip install -e src/dnn

Including the optional -e flag will install package in "editable" mode, meaning that instead of copying the files into your virtual environment, a symlink will be created to the files where they are.

Run jupyter server

jupyter notebook notebooks/

You can now use the jupyter kernel dnn to run notebooks.