The base models (π₀, π₀-FAST, π₀.₅) use a 32-dimensional action space, but the documentation in docs/norm_stats.md only explains dimensions 0-15:
|
Out of the box, both the `pi0_base` and `pi0_fast_base` use the following action space definitions (left and right are defined looking from behind the robot towards the workspace): |
|
``` |
|
"dim_0:dim_5": "left arm joint angles", |
|
"dim_6": "left arm gripper position", |
|
"dim_7:dim_12": "right arm joint angles (for bi-manual only)", |
|
"dim_13": "right arm gripper position (for bi-manual only)", |
|
|
|
# For mobile robots: |
|
"dim_14:dim_15": "x-y base velocity (for mobile robots only)", |
|
``` |
What do dimensions 16-31 represent?
This makes it difficult to interpret model outputs and understand the complete action space structure.
The base models (π₀, π₀-FAST, π₀.₅) use a 32-dimensional action space, but the documentation in docs/norm_stats.md only explains dimensions 0-15:
openpi/docs/norm_stats.md
Lines 49 to 58 in 175f89c
What do dimensions 16-31 represent?
This makes it difficult to interpret model outputs and understand the complete action space structure.