Hugging Face's logo

robotgeneralist
/
cs_basic_sitters

Model card Files
xet
 Community
cs_basic_sitters
File size: 2,468 Bytes
d398302
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
!!python/object:openpi.training.config.TrainConfig
name: cs_basic_sitters_continuous_shift15
project_name: openpi
exp_name: cs_basic_sitters_continuous_shift15
model: !!python/object:openpi.models.pi0_config.Pi0Config
    action_dim: 32
    action_horizon: 50
    max_token_len: 48
    dtype: bfloat16
    paligemma_variant: gemma_2b
    action_expert_variant: gemma_300m
    img_variant: So400m/14
    action_loss_horizon: null
    pi05: false
    discrete_state_input: false
weight_loader: !!python/object:openpi.training.weight_loaders.CheckpointWeightLoader
    local_dir: null
    repo_id: robotgeneralist/openpi_checkpoint_mirrors
    revision: main
    checkpoint_path: pi0_base
pytorch_weight_path: null
pytorch_training_precision: bfloat16
lr_schedule: !!python/object:openpi.training.optimizer.CosineDecaySchedule
    warmup_steps: 1000
    peak_lr: 2.5e-05
    decay_steps: 30000
    decay_lr: 2.5e-06
optimizer: !!python/object:openpi.training.optimizer.AdamW
    b1: 0.9
    b2: 0.95
    eps: 1.0e-08
    weight_decay: 1.0e-10
    clip_gradient_norm: 1.0
ema_decay: 0.99
freeze_filter: !!python/name:flax.nnx.filterlib.Nothing ''
data: !!python/object:openpi.training.data_configs.lerobot_nomagic_urx.NomagicURXDataConfig
    repo_ids: !!python/tuple
    - robotgeneralist/cs__basic_4a
    - robotgeneralist/cs__basic_4b
    - robotgeneralist/cs__basic_4c
    - robotgeneralist/cs__basic_4e
    - robotgeneralist/cs__basic_4f
    - robotgeneralist/cs__basic_4g
    - robotgeneralist/cs__basic_4h
    - robotgeneralist/cs__basic_4i
    assets: !!python/object:openpi.training.config.AssetsConfig
        assets_dir: null
        asset_id: ur10e
    norm_stats_sample_ratio: 0.05
    base_config: null
    repack_format: robirosV2
    tool_mode: continuous
    tool_shift: 15
    input_format: !!python/tuple
    - absolute
    - rotvec
    model_format: !!python/tuple
    - delta
    - euler
    output_format: null
    binary_action_keys: !!python/tuple []
    fourier_config: null
    per_action_index_norm_stats: false
    default_prompt: null
assets_base_dir: ./assets
checkpoint_base_dir: /storage_nvme_3/mwrochna/checkpoints
seed: 42
batch_size: 32
num_workers: 2
num_train_steps: 25001
log_interval: 100
save_interval: 1000
keep_period: 5000
overwrite: false
resume: true
wandb_enabled: true
policy_metadata: null
fsdp_devices: 1
test_split_ratio: 0.05
test_eval_interval: 1000
test_eval_use_train_mode: false
binary_action_keys: !!python/tuple []