This repository contains code to train machine learning models for risk prediction.
Besides pytorch and the usual ML stack, you will need speedrun which you can install as:
pip install git+https://github.com/inferno-pytorch/speedrun.git@dev`.
For logging with wandb, you'll also need to pip install wandb.
Download the dataset from here, and decompress in to data.
To train a model, first mkdir exp and then:
python train.py exp/MY-CTT-EXPERIMENT-0 --inherit base_config/CTT-0
This will start training on a GPU, if available. If you want to use a CPU instead, append the --config.device cpu to the above command.