real-stanford / diffusion_policy

[RSS 2023] Diffusion Policy Visuomotor Policy Learning via Action Diffusion
https://diffusion-policy.cs.columbia.edu/
MIT License
1.09k stars 203 forks source link

about eval results of pushT, is this normal? #72

Open yanghu819 opened 2 months ago

yanghu819 commented 2 months ago

I run with this python eval.py --checkpoint ./epoch=0550-test_mean_score=0.969.ckpt --output_dir data/pusht_eval_output --device cuda:0 the result is: { "test/mean_score": 0.9061810892200681, "test/sim_max_reward_4300000": 1.0, "test/sim_max_reward_4300001": 0.9886837971072213, "test/sim_max_reward_4300002": 1.0, ... while the result should be: { "test/mean_score": 0.9150393806777066, "test/sim_max_reward_4300000": 1.0, "test/sim_max_reward_4300001": 0.9872969750774386, ... "train/sim_video_1": "data/pusht_eval_output//media/2fo4btlf.mp4" }