Python Classifier Submission for PhysioNet/CinC Challenge 2020

Source code for the PhysioNet/CinC Challenge 2020. Extends the Python Classifier Template.

Team Details

Team name: CVC

Alexander William Wong <alex.wong@ualberta.ca>
- University of Alberta
Weijie Sun <weijie2@ualberta.ca>
- Canadian Vigour Centre
Sunil Vasu Kalmady <kalmady@ualberta.ca>
- Canadian Vigour Centre
Padma Kaul <pkaul@ualberta.ca>
- Canadian Vigour Centre
Abram Hindle <abram.hindle@ualberta.ca>
- University of Alberta

Getting Started

All of the raw training data is available from this discussion post.

Retrieve the evaluation code
```
git submodule update --init --recursive
```
- This is required to resolve the symbolic links for util/evaluate_12ECG_score.py and weights.csv

Create a python3 virtual environment and install the required dependencies.

python3 -m venv env
source env/bin/activate
pip install --upgrade pip
pip install wheel
pip install -r requirements.txt

Download and extract the training data.
```
cd data
./quickstart.sh
```

Run the unittests.

python3 -m unittest discover -v -s ./tests -p test_*.py

Example Classifier

Docker must have GPU support! See NVIDIA/nvidia-docker

The following code is run from a fresh clone of this repository, and meant to demonstrate that the code works according to the challenge specifications.

Tested on a local server, NVIDIA-SMI 450.57 | Driver Version: 450.57 | CUDA Version: 11.0.

# Build the docker image
$ docker build -t image .
# Open up an interactive shell
$ docker run --gpus all -it image bash

# Within Docker, train a model on the unittest data and store it in the model folder
root@HASH$ python3 train_model.py tests/data model
# loky warnings may be ignored, alternatively to silence warnings:
root@HASH$ python3 -W ignore train_model.py tests/data model

# after model is trained, run the driver code
root@HASH$ python3 driver.py model tests/data output

# after output is generated, run the evaluation code
root@HASH$ python3 evaluation-2020/evaluate_12ECG_score.py tests/data output

Example classifier code for Python for the PhysioNet/CinC Challenge 2020

This code uses two main scripts to train the model and classify the data:

train_model.py Train your model. Add your model code to the train_12ECG_model function. It also performs all file input and output. Do not edit this script or we will be unable to evaluate your submission.
driver.py is the classifier which calls the output from your train_model script. It also performs all file input and output. Do not edit this script or we will be unable to evaluate your submission.

Check the code in these files for the input and output formats for the train_model and driver scripts.

To create and save your model, you should edit train_12ECG_classifier.py script. Note that you should not change the input arguments of the train_12ECG_classifier function or add output arguments. The needed models and parameters should be saved in a separated file. In the sample code, an additional script, get_12ECG_features.py, is used to extract hand-crafted features.

To run your classifier, you should edit the run_12ECG_classifier.py script, which takes a single recording as input and outputs the predicted classes and probabilities. Please, keep the formats of both outputs as they are shown in the example. You should not change the inputs and outputs of the run_12ECG_classifier function.

Use

You can run this classifier code by installing the requirements and running

python train_model.py training_data model
python driver.py model test_data test_outputs

where training_data is a directory of training data files, model is a directory of files for the model, test_data is the directory of test data files, and test_outputs is a directory of classifier outputs. The PhysioNet/CinC 2020 webpage provides a training database with data files and a description of the contents and structure of these files.

Submission

The driver.py, get_12ECG_score.py, and get_12ECG_features.py scripts must be in the root path of your repository. If they are inside a folder, then the submission will be unsuccessful.

Details

See the PhysioNet/CinC 2020 webpage for more details, including instructions for the other files in this repository.

awwong1/physionet-challenge-2020