For the full coding pipeline and tutorial please refer to the GitHub repository: https://github.com/EETChallenge/3et_challenge_2025

## Training Pipeline Description
The default training pipeline is very handy to use. Simply run the following command with the code in the GitHub repository:
```
python train.py --config train_baseline.json
```
This should give a baseline result for a basic CNN+GRU model trained on voxel-grid event representation. The training progress, checkpoints and metrics are all logged with the MLFlow library. See the repo for more details.
![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F19454843%2Fab2c86486ac9bb86d9097e691a92174b%2Finbox_18771626_e78190165f02d967a7bd8bc992e5ca64_event-based-eye-movement.gif?generation=1737084986065375&alt=media)