Hugging Face's logo

jmercat
/
diffusion_sample_koch_feed_cat

Robotics
LeRobot
Safetensors
diffusion-policy
model_hub_mixin
pytorch_model_hub_mixin
Model card Files
xet
Community
jmercat commited on
Commit
6c09e08
·
verified ·
1 Parent(s): 8e1bb9d

Upload folder using huggingface_hub

Browse files
Files changed (4) hide show
  1. README.md +12 -0
  2. config.json +55 -0
  3. config.yaml +141 -0
  4. model.safetensors +3 -0
README.md ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: lerobot
3
+ tags:
4
+ - diffusion-policy
5
+ - model_hub_mixin
6
+ - pytorch_model_hub_mixin
7
+ - robotics
8
+ ---
9
+
10
+ This model has been pushed to the Hub using the [PytorchModelHubMixin](https://huggingface.co/docs/huggingface_hub/package_reference/mixins#huggingface_hub.PyTorchModelHubMixin) integration:
11
+ - Library: https://github.com/huggingface/lerobot
12
+ - Docs: [More Information Needed]
config.json ADDED
@@ -0,0 +1,55 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "beta_end": 0.02,
3
+ "beta_schedule": "squaredcos_cap_v2",
4
+ "beta_start": 0.0001,
5
+ "clip_sample": true,
6
+ "clip_sample_range": 1.0,
7
+ "crop_is_random": true,
8
+ "crop_shape": [
9
+ 432,
10
+ 576
11
+ ],
12
+ "diffusion_step_embed_dim": 128,
13
+ "do_mask_loss_for_padding": false,
14
+ "down_dims": [
15
+ 512,
16
+ 1024,
17
+ 2048
18
+ ],
19
+ "horizon": 16,
20
+ "input_normalization_modes": {
21
+ "observation.images.phone": "mean_std",
22
+ "observation.state": "mean_std"
23
+ },
24
+ "input_shapes": {
25
+ "observation.images.phone": [
26
+ 3,
27
+ 480,
28
+ 640
29
+ ],
30
+ "observation.state": [
31
+ 6
32
+ ]
33
+ },
34
+ "kernel_size": 5,
35
+ "n_action_steps": 8,
36
+ "n_groups": 8,
37
+ "n_obs_steps": 1,
38
+ "noise_scheduler_type": "DDPM",
39
+ "num_inference_steps": null,
40
+ "num_train_timesteps": 100,
41
+ "output_normalization_modes": {
42
+ "action": "mean_std"
43
+ },
44
+ "output_shapes": {
45
+ "action": [
46
+ 6
47
+ ]
48
+ },
49
+ "prediction_type": "sample",
50
+ "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1",
51
+ "spatial_softmax_num_keypoints": 32,
52
+ "use_film_scale_modulation": true,
53
+ "use_group_norm": false,
54
+ "vision_backbone": "resnet18"
55
+ }
config.yaml ADDED
@@ -0,0 +1,141 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ resume: false
2
+ device: cuda
3
+ use_amp: false
4
+ seed: 1000
5
+ dataset_repo_id: jmercat/koch_feed_cat
6
+ video_backend: pyav
7
+ training:
8
+ offline_steps: 16000
9
+ num_workers: 4
10
+ batch_size: 64
11
+ eval_freq: -1
12
+ log_freq: 200
13
+ save_checkpoint: true
14
+ save_freq: 1600
15
+ online_steps: 0
16
+ online_rollout_n_episodes: 1
17
+ online_rollout_batch_size: 1
18
+ online_steps_between_rollouts: 1
19
+ online_sampling_ratio: 0.5
20
+ online_env_seed: null
21
+ online_buffer_capacity: null
22
+ online_buffer_seed_size: 0
23
+ do_online_rollout_async: false
24
+ image_transforms:
25
+ enable: false
26
+ max_num_transforms: 3
27
+ random_order: false
28
+ brightness:
29
+ weight: 1
30
+ min_max:
31
+ - 0.8
32
+ - 1.2
33
+ contrast:
34
+ weight: 1
35
+ min_max:
36
+ - 0.8
37
+ - 1.2
38
+ saturation:
39
+ weight: 1
40
+ min_max:
41
+ - 0.5
42
+ - 1.5
43
+ hue:
44
+ weight: 1
45
+ min_max:
46
+ - -0.05
47
+ - 0.05
48
+ sharpness:
49
+ weight: 1
50
+ min_max:
51
+ - 0.8
52
+ - 1.2
53
+ grad_clip_norm: 10
54
+ lr: 0.0001
55
+ lr_scheduler: cosine
56
+ lr_warmup_steps: 500
57
+ adam_betas:
58
+ - 0.95
59
+ - 0.999
60
+ adam_eps: 1.0e-08
61
+ adam_weight_decay: 1.0e-06
62
+ delta_timestamps:
63
+ action:
64
+ - 0.0
65
+ - 0.03333333333333333
66
+ - 0.06666666666666667
67
+ - 0.1
68
+ - 0.13333333333333333
69
+ - 0.16666666666666666
70
+ - 0.2
71
+ - 0.23333333333333334
72
+ - 0.26666666666666666
73
+ - 0.3
74
+ - 0.3333333333333333
75
+ - 0.36666666666666664
76
+ - 0.4
77
+ - 0.43333333333333335
78
+ - 0.4666666666666667
79
+ - 0.5
80
+ eval:
81
+ n_episodes: 5
82
+ batch_size: 5
83
+ use_async_envs: false
84
+ wandb:
85
+ enable: true
86
+ disable_artifact: false
87
+ project: lerobot
88
+ notes: ''
89
+ fps: 30
90
+ env:
91
+ name: real_world
92
+ task: null
93
+ state_dim: 6
94
+ action_dim: 6
95
+ fps: ${fps}
96
+ policy:
97
+ name: diffusion
98
+ n_obs_steps: 1
99
+ horizon: 16
100
+ n_action_steps: 8
101
+ input_shapes:
102
+ observation.images.phone:
103
+ - 3
104
+ - 480
105
+ - 640
106
+ observation.state:
107
+ - ${env.state_dim}
108
+ output_shapes:
109
+ action:
110
+ - ${env.action_dim}
111
+ input_normalization_modes:
112
+ observation.images.phone: mean_std
113
+ observation.state: mean_std
114
+ output_normalization_modes:
115
+ action: mean_std
116
+ vision_backbone: resnet18
117
+ crop_shape:
118
+ - 432
119
+ - 576
120
+ crop_is_random: true
121
+ pretrained_backbone_weights: ResNet18_Weights.IMAGENET1K_V1
122
+ use_group_norm: false
123
+ spatial_softmax_num_keypoints: 32
124
+ down_dims:
125
+ - 512
126
+ - 1024
127
+ - 2048
128
+ kernel_size: 5
129
+ n_groups: 8
130
+ diffusion_step_embed_dim: 128
131
+ use_film_scale_modulation: true
132
+ noise_scheduler_type: DDPM
133
+ num_train_timesteps: 100
134
+ beta_schedule: squaredcos_cap_v2
135
+ beta_start: 0.0001
136
+ beta_end: 0.02
137
+ prediction_type: sample
138
+ clip_sample: true
139
+ clip_sample_range: 1.0
140
+ num_inference_steps: null
141
+ do_mask_loss_for_padding: false
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1f33fa50e147e32e1502131e94b7d170ff9c8474d25462e75a63c36a2eda0ed2
3
+ size 1043856648