This is a toy reinforcement learning project. The goal is to automate the braking process of a train, without any prior knowledge of how a train should brake in order to reach a particular point in space (on a linear axis)

Training

TODO You can train your model using python train.py train path_to_model

TODO Once your model is trained, you may use it by calling python train.py run path_to_model

For now, you can run the model by calling : python train.py run -x objective_x_coordinate -s initial_speed