imitation.scripts

Command-line scripts.

Modules

imitation.scripts.analyze	Commands to analyze experimental results.
imitation.scripts.config	Configuration settings for scripts.
imitation.scripts.convert_trajs	Converts old-style pickle or npz trajectories to new-style HuggingFace datasets.
imitation.scripts.eval_policy	Evaluate policies: render policy interactively, save videos, log episode return.
imitation.scripts.ingredients	Ingredients for Sacred experiments.
imitation.scripts.parallel	Runs a Sacred experiment in parallel.
imitation.scripts.train_adversarial	Train GAIL or AIRL.
imitation.scripts.train_imitation	Trains DAgger on synthetic demonstrations generated from an expert policy.
imitation.scripts.train_preference_comparisons	Train a reward model using preference comparisons.
imitation.scripts.train_rl	Uses RL to train a policy from scratch, saving rollouts and policy.
imitation.scripts.tuning	Tunes the hyperparameters of the algorithms.