Hugging Face's logo

cadene
/
2024_05_28_diffusion_real_world_aloha_v2_static_dora_test_wrist_gripper_005000

Transformers
Safetensors
pytorch_model_hub_mixin
model_hub_mixin
Inference Endpoints
Model card Files Community
2024_05_28_diffusion_real_world_aloha_v2_static_dora_test_wrist_gripper_005000
File size: 3,408 Bytes 
d9964f1
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
 
device: cuda
seed: 100000
dataset_repo_id: cadene/wrist_gripper
training:
  offline_steps: 200000
  online_steps: 0
  online_steps_between_rollouts: 1
  online_sampling_ratio: 0.5
  online_env_seed: ???
  eval_freq: 99999999999999
  save_freq: 1000
  log_freq: 100
  save_model: true
  num_workers: 4
  batch_size: 64
  grad_clip_norm: 10
  lr: 0.0001
  lr_scheduler: cosine
  lr_warmup_steps: 500
  adam_betas:
  - 0.95
  - 0.999
  adam_eps: 1.0e-08
  adam_weight_decay: 1.0e-06
  delta_timestamps:
    observation.images.cam_right_wrist:
    - -0.03333333333333333
    - 0.0
    observation.images.cam_left_wrist:
    - -0.03333333333333333
    - 0.0
    observation.images.cam_high:
    - -0.03333333333333333
    - 0.0
    observation.images.cam_low:
    - -0.03333333333333333
    - 0.0
    observation.state:
    - -0.03333333333333333
    - 0.0
    action:
    - -0.03333333333333333
    - 0.0
    - 0.03333333333333333
    - 0.06666666666666667
    - 0.1
    - 0.13333333333333333
    - 0.16666666666666666
    - 0.2
    - 0.23333333333333334
    - 0.26666666666666666
    - 0.3
    - 0.3333333333333333
    - 0.36666666666666664
    - 0.4
    - 0.43333333333333335
    - 0.4666666666666667
eval:
  n_episodes: 50
  batch_size: 50
  use_async_envs: false
wandb:
  enable: true
  disable_artifact: true
  project: lerobot
  notes: ''
fps: 30
env:
  name: dora
  task: DoraAloha-v0
  episode_length: 400
  fps: ${fps}
  state_dim: 14
  action_dim: 14
override_dataset_stats:
  observation.images.cam_right_wrist:
    mean:
    - - - 0.485
    - - - 0.456
    - - - 0.406
    std:
    - - - 0.229
    - - - 0.224
    - - - 0.225
  observation.images.cam_left_wrist:
    mean:
    - - - 0.485
    - - - 0.456
    - - - 0.406
    std:
    - - - 0.229
    - - - 0.224
    - - - 0.225
  observation.images.cam_high:
    mean:
    - - - 0.485
    - - - 0.456
    - - - 0.406
    std:
    - - - 0.229
    - - - 0.224
    - - - 0.225
  observation.images.cam_low:
    mean:
    - - - 0.485
    - - - 0.456
    - - - 0.406
    std:
    - - - 0.229
    - - - 0.224
    - - - 0.225
policy:
  name: diffusion
  n_obs_steps: 2
  horizon: 16
  n_action_steps: 8
  input_shapes:
    observation.images.cam_right_wrist:
    - 3
    - 480
    - 640
    observation.images.cam_left_wrist:
    - 3
    - 480
    - 640
    observation.images.cam_high:
    - 3
    - 480
    - 640
    observation.images.cam_low:
    - 3
    - 480
    - 640
    observation.state:
    - ${env.state_dim}
  output_shapes:
    action:
    - ${env.action_dim}
  input_normalization_modes:
    observation.images.cam_right_wrist: mean_std
    observation.images.cam_left_wrist: mean_std
    observation.images.cam_high: mean_std
    observation.images.cam_low: mean_std
    observation.state: mean_std
  output_normalization_modes:
    action: mean_std
  vision_backbone: resnet18
  crop_shape: null
  crop_is_random: false
  pretrained_backbone_weights: ResNet18_Weights.IMAGENET1K_V1
  use_group_norm: false
  spatial_softmax_num_keypoints: 32
  down_dims:
  - 512
  - 1024
  - 2048
  kernel_size: 5
  n_groups: 8
  diffusion_step_embed_dim: 128
  use_film_scale_modulation: true
  noise_scheduler_type: DDPM
  num_train_timesteps: 100
  beta_schedule: squaredcos_cap_v2
  beta_start: 0.0001
  beta_end: 0.02
  prediction_type: epsilon
  clip_sample: true
  clip_sample_range: 1.0
  num_inference_steps: 100
  do_mask_loss_for_padding: false