This codebase contains an example submission for the Final phase of the STOIC2021 COVID-19 AI Challenge. It implements simple pipelines for training a baseline model and running inference with it.
The training pipeline trains an I3D model on the STOIC2021 training data, and produces trained model weights in a specified directory. The inference pipeline uses the trained model weights to predict COVID-19 presence and COVID-19 severity from a CT scan.
If something in this codebase does not work for you, please do not hesitate to contact us or add a post in the forum. If the problem is related to the code of this repository, please create a new issue on GitHub.
We recommend using this repository on Linux. If you are using Windows, we recommend installing Windows Subsystem for Linux (WSL). Please watch the official tutorial by Microsoft for installing WSL 2 with GPU support.
- Have Docker installed.
- Have the STOIC2021 public training set downloaded to your machine. Instructions for downloading it can be found here.
This codebase serves as a template for submissions to the Final phase of the STOIC2021 challenge. Submissions to this phase must include:
- code in the
training/directory for generating a Docker container that can train your Algorithm; - code in the
inference/directory for generating a Docker container that can run inference for your Algorithm; - a LICENSE file in the root directory containing a permissive open source license.
The README.md files in the training/ and inference/ directories provide a tutorial on how to use this template. Before submitting your solution, please make sure that the training/test.sh and inference/test.sh scripts (or test.bat scripts for Windows users) complete successfully.
Specifications of the hardware used in the Final phase will be released soon.
Details on how to submit will be shared with participants when invited to the Final phase.