Downloads · 30 days
0
panda-sort-three-stack-cup-mixed-30hz
alanoob/panda-sort-three-stack-cup-mixed-30hz
panda-sort-three-stack-cup-mixed-30hz is a machine learning model from alanoob. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
TrainConfig( name="sortthreestackcupmixed30hzinference", model=pi0config.Pi0Config( pi05=True, actionhorizon=20, discretestateinput=False, ), data=FrankaDataConfig( repoid="kitalr/pandasortthreestacktwocupmixed30hz",…
Access
Public
Updated Sep 3, 2026
Repo size
12.4 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
Other12.4 GB · 100%
At a glance
- Access
- Public
- Created
- Sep 3, 2026
- Updated
- Sep 3, 2026
- SHA
- 1754de21
Model card
From the Hugging Face model README
TrainConfig( name="sort_three_stack_cup_mixed_30hz_inference", model=pi0_config.Pi0Config( pi05=True, action_horizon=20, discrete_state_input=False, ), data=FrankaDataConfig( # 必须和 checkpoint 内 assets 的目录名称一致 repo_id="kitalr/panda_sort_three_stack_two_cup_mixed_30hz", base_config=DataConfig( prompt_from_task=True, ), use_ee_control=True, include_gripper_proprio=True, include_third_cam=True, ), batch_size=16, video_backend="pyav",
# 以下训练参数在推理时不会使用，但保留完整结构
lr_schedule=_optimizer.CosineDecaySchedule(
warmup_steps=1_000,
peak_lr=5e-5,
decay_steps=20_000,
decay_lr=5e-6,
),
optimizer=_optimizer.AdamW(
clip_gradient_norm=1.0,
),
ema_decay=0.999,
weight_loader=weight_loaders.NoOpWeightLoader(),
num_train_steps=20_001,
save_interval=2_000,
keep_period=10_000,
num_workers=10,
fsdp_devices=1,
)
- Created
- Sep 3, 2026
- Updated
- Sep 3, 2026