Track-to-Learn: A general framework for tractography with deep reinforcement learning

First release of Track-to-Learn: A general framework for tractography with deep reinforcement learning

See trained models and results: https://zenodo.org/record/5080541

Bugs in the implementation prevent the full reproduction of results. Please refer to "What matters in reinforcement learning for tractography" for updated, reproducible, results.