yiju-1213's picture
Publish single-step grasp refinement evaluation checkpoint
4e5de6c verified
|
Raw
History Blame Contribute Delete
341 Bytes

Normalization

No separate observation running mean/variance file was found for the selected experiment or current RL code path.

The released checkpoint stores the actor-critic weights and calibrator. Action scaling is defined in the environment config:

translation_bound: [0.01, 0.01, 0.01]
rotation_bound: [0.1, 0.1, 0.1]