Downloads · 30 days
0
coeva_cycle1_acopd_80
THU-SIGS-EILAB/coeva_cycle1_acopd_80
coeva_cycle1_acopd_80 is a machine learning model from THU-SIGS-EILAB. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
Inference-ready export of the no-ZV student trained with the Z/V-prior teacher.
Access
Public
Updated Sep 14, 2026
Repo size
8.6 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.pt8.6 GB · 100%
At a glance
- Access
- Public
- Created
- Sep 14, 2026
- Updated
- Sep 14, 2026
- SHA
- 0499ffc7
Model card
From the Hugging Face model README
AC-OPD LIBERO-10 student, step 80
Inference-ready export of the no-ZV student trained with the Z/V-prior teacher.
Contents:
actor/model_state_dict/full_weights.pt: step-80 full model weights.physical-intelligence/libero/norm_stats.json: required LIBERO action normalization statistics.rlinf_model_config.yaml: Pi0.5 model configuration.training_config.yaml: frozen configuration used for this AC-OPD run.
Source checkpoint:
logs/20260913-23:06:51-libero_10_ac_opd_pi05_zv_teacher_nozv_student_4gpu/cycle_00_acopd_zv_teacher_nozv_student/checkpoints/global_step_80.
This export intentionally excludes actor/dcp_checkpoint/, which is only
needed to resume distributed training; full_weights.pt is the inference and
standalone-evaluation checkpoint.
- Created
- Sep 14, 2026
- Updated
- Sep 14, 2026