Downloads · 30 days
16
24% of all-time downloads
ZeyuLing/motius-evaluator-g1-38d-tmr
motius-evaluator-g1-38d-tmr is a feature extraction model from ZeyuLing. Use it when you need embeddings to search or compare text. It is set up for motius. The card lists the license as mit.
Motius reproduction of the TMR architecture trained from scratch for native Unitree G1 motion. The motion input is the canonicalized G1-38D representation at 30 fps: root XY velocity and height, root rotation 6D, and…
Downloads · 30 days
16
24% of all-time downloads
All-time downloads
68
Public
Parameters
16M
64.1 MB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors64.1 MB · 100%
From the Hugging Face model README
Motius reproduction of the TMR architecture trained from scratch for native Unitree G1 motion. The motion input is the canonicalized G1-38D representation at 30 fps: root XY velocity and height, root rotation 6D, and 29 joint angles.
TMRBundle / TMRG1Evaluatormodel.safetensors: motion encoder, text encoder, and reconstruction decoderstats/: G1-38D training-set normalization statisticsconfig.json and preprocessor_config.json: architecture and protocolartifact_manifest.json: file and source-checkpoint hashesThe public checkpoint is epoch 139, trained on the 344,966-motion training split of the 359,153-clip HYMotion Data G1 materialization.
from motius.evaluation import TMRG1Evaluator
evaluator = TMRG1Evaluator.from_pretrained(
"ZeyuLing/motius-evaluator-g1-38d-tmr"
)