python lerobot/scripts/train.py \
  hydra.run.dir=outputs/train/diffusion_pusht \
  hydra.job.name=diffusion_pusht \
  policy=diffusion training.save_model=true \
  env=pusht \
  env.task=PushT-v0 \
  dataset_repo_id=lerobot/pusht \
  training.save_freq=25000 \
  training.eval_freq=10000 \
  wandb.enable=true \
  device=cuda

The training and eval curves may be found at https://api.wandb.ai/links/none7/rd2trav7

This took about 6 hours to train on an Nvida Tesla P100 GPU.

Evaluation

The model was evaluated on the PushT environment from gym-pusht and compared to a similar model trained with the original Diffusion Policy code. There are two evaluation metrics on a per-episode basis:

Maximum overlap with target (seen as eval/avg_max_reward in the charts above). This ranges in [0, 1].
Success: whether or not the maximum overlap is at least 95%.

Here are the metrics for 500 episodes worth of evaluation. The "Theirs" column is for an equivalent model from the experiments of original Diffusion Policy repository and evaluated on LeRobot (the model weights may be found in the original_dp_repo branch of this respository).

	Ours	Theirs
Average max. overlap ratio	0.799	0.605
Success rate for 500 episodes (%)	32.00	17.00

The results of each of the individual rollouts may be found in eval_info.json.

Amandee/diffusion_transformer_pusht

Model Card for Transformer based Diffusion Policy / PushT

How to Get Started with the Model

Training Details

Evaluation

Amandee/
diffusion_transformer_pusht