GRASPNET_FINAL_moreData_h100
openpi π₀.₅ policy checkpoint.
Train config:
pi05_Franka_GRASPNET_FINAL_moreDataExperiment:
GRASPNET_FINAL_moreData_h100Checkpoints in this repo:
step_50000/step_99999/
Each folder is one training step and is self-contained.
- Training dataset: saifahmad123/GRASPNET_FINAL_moreData
Contents
| Path | Purpose |
|---|---|
step_50000/params/ |
Model weights. Required to serve the policy. |
step_50000/assets/ |
Normalization statistics. Required to serve the policy. |
step_50000/train_state/ |
Optimizer state. Only present if uploaded with --include-train-state; needed to resume training. |
Usage
from huggingface_hub import snapshot_download
from openpi.policies import policy_config
from openpi.training import config as _config
ckpt = snapshot_download("saifahmad123/GRASPNET_FINAL_moreData_h100", allow_patterns="step_50000/*") + "/step_50000"
train_config = _config.get_config("pi05_Franka_GRASPNET_FINAL_moreData")
policy = policy_config.create_trained_policy(train_config, ckpt)
action_chunk = policy.infer(observation)["actions"]