Hugging Face's logo

KingLear1606
/
dm_300

Robotics
LeRobot
Safetensors
act_orin
Model card Files
xet
 Community
KingLear1606 commited on
Commit
a82676f
·
verified ·
1 Parent(s): cb1a075

Upload policy weights, train config and readme

Browse files
Files changed (4) hide show
  1. README.md +62 -0
  2. config.json +79 -0
  3. model.safetensors +3 -0
  4. train_config.json +228 -0
README.md ADDED
@@ -0,0 +1,62 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ datasets: KingLear1606/dm
3
+ library_name: lerobot
4
+ license: apache-2.0
5
+ model_name: act_orin
6
+ pipeline_tag: robotics
7
+ tags:
8
+ - lerobot
9
+ - act_orin
10
+ - robotics
11
+ ---
12
+
13
+ # Model Card for act_orin
14
+
15
+ <!-- Provide a quick summary of what the model is/does. -->
16
+
17
+
18
+ _Model type not recognized — please update this template._
19
+
20
+
21
+ This policy has been trained and pushed to the Hub using [LeRobot](https://github.com/huggingface/lerobot).
22
+ See the full documentation at [LeRobot Docs](https://huggingface.co/docs/lerobot/index).
23
+
24
+ ---
25
+
26
+ ## How to Get Started with the Model
27
+
28
+ For a complete walkthrough, see the [training guide](https://huggingface.co/docs/lerobot/il_robots#train-a-policy).
29
+ Below is the short version on how to train and run inference/eval:
30
+
31
+ ### Train from scratch
32
+
33
+ ```bash
34
+ lerobot-train \
35
+ --dataset.repo_id=${HF_USER}/<dataset> \
36
+ --policy.type=act \
37
+ --output_dir=outputs/train/<desired_policy_repo_id> \
38
+ --job_name=lerobot_training \
39
+ --policy.device=cuda \
40
+ --policy.repo_id=${HF_USER}/<desired_policy_repo_id>
41
+ --wandb.enable=true
42
+ ```
43
+
44
+ _Writes checkpoints to `outputs/train/<desired_policy_repo_id>/checkpoints/`._
45
+
46
+ ### Evaluate the policy/run inference
47
+
48
+ ```bash
49
+ lerobot-record \
50
+ --robot.type=so100_follower \
51
+ --dataset.repo_id=<hf_user>/eval_<dataset> \
52
+ --policy.path=<hf_user>/<desired_policy_repo_id> \
53
+ --episodes=10
54
+ ```
55
+
56
+ Prefix the dataset repo with **eval\_** and supply `--policy.path` pointing to a local or hub checkpoint.
57
+
58
+ ---
59
+
60
+ ## Model Details
61
+
62
+ - **License:** apache-2.0
config.json ADDED
@@ -0,0 +1,79 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "type": "act_orin",
3
+ "n_obs_steps": 1,
4
+ "input_features": {
5
+ "observation.state": {
6
+ "type": "STATE",
7
+ "shape": [
8
+ 6
9
+ ]
10
+ },
11
+ "observation.images.wrist_cam": {
12
+ "type": "VISUAL",
13
+ "shape": [
14
+ 480,
15
+ 640,
16
+ 3
17
+ ]
18
+ },
19
+ "observation.images.top": {
20
+ "type": "VISUAL",
21
+ "shape": [
22
+ 480,
23
+ 640,
24
+ 3
25
+ ]
26
+ },
27
+ "observation.images.side": {
28
+ "type": "VISUAL",
29
+ "shape": [
30
+ 848,
31
+ 640,
32
+ 3
33
+ ]
34
+ }
35
+ },
36
+ "output_features": {
37
+ "action": {
38
+ "type": "ACTION",
39
+ "shape": [
40
+ 6
41
+ ]
42
+ }
43
+ },
44
+ "device": "cuda",
45
+ "use_amp": false,
46
+ "use_peft": false,
47
+ "push_to_hub": true,
48
+ "repo_id": "KingLear1606/dm_300",
49
+ "private": null,
50
+ "tags": null,
51
+ "license": null,
52
+ "pretrained_path": null,
53
+ "chunk_size": 100,
54
+ "n_action_steps": 100,
55
+ "normalization_mapping": {
56
+ "VISUAL": "MEAN_STD",
57
+ "STATE": "MEAN_STD",
58
+ "ACTION": "MEAN_STD"
59
+ },
60
+ "vision_backbone": "resnet18",
61
+ "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1",
62
+ "replace_final_stride_with_dilation": false,
63
+ "pre_norm": false,
64
+ "dim_model": 512,
65
+ "n_heads": 8,
66
+ "dim_feedforward": 3200,
67
+ "feedforward_activation": "relu",
68
+ "n_encoder_layers": 4,
69
+ "n_decoder_layers": 1,
70
+ "use_vae": true,
71
+ "latent_dim": 32,
72
+ "n_vae_encoder_layers": 4,
73
+ "temporal_ensemble_coeff": null,
74
+ "dropout": 0.1,
75
+ "kl_weight": 0.1,
76
+ "optimizer_lr": 1e-05,
77
+ "optimizer_weight_decay": 0.0001,
78
+ "optimizer_lr_backbone": 1e-05
79
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:abec2d9d0b6a3c9f02cf4728c49f83d1b84a39b6e40f4788a24120d4786b77dd
3
+ size 206699736
train_config.json ADDED
@@ -0,0 +1,228 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset": {
3
+ "repo_id": "KingLear1606/dm",
4
+ "root": "/home/dc/.cache/huggingface/lerobot/KingLear1606/dm_300",
5
+ "episodes": null,
6
+ "image_transforms": {
7
+ "enable": false,
8
+ "max_num_transforms": 3,
9
+ "random_order": false,
10
+ "tfs": {
11
+ "brightness": {
12
+ "weight": 1.0,
13
+ "type": "ColorJitter",
14
+ "kwargs": {
15
+ "brightness": [
16
+ 0.8,
17
+ 1.2
18
+ ]
19
+ }
20
+ },
21
+ "contrast": {
22
+ "weight": 1.0,
23
+ "type": "ColorJitter",
24
+ "kwargs": {
25
+ "contrast": [
26
+ 0.8,
27
+ 1.2
28
+ ]
29
+ }
30
+ },
31
+ "saturation": {
32
+ "weight": 1.0,
33
+ "type": "ColorJitter",
34
+ "kwargs": {
35
+ "saturation": [
36
+ 0.5,
37
+ 1.5
38
+ ]
39
+ }
40
+ },
41
+ "hue": {
42
+ "weight": 1.0,
43
+ "type": "ColorJitter",
44
+ "kwargs": {
45
+ "hue": [
46
+ -0.05,
47
+ 0.05
48
+ ]
49
+ }
50
+ },
51
+ "sharpness": {
52
+ "weight": 1.0,
53
+ "type": "SharpnessJitter",
54
+ "kwargs": {
55
+ "sharpness": [
56
+ 0.5,
57
+ 1.5
58
+ ]
59
+ }
60
+ },
61
+ "affine": {
62
+ "weight": 1.0,
63
+ "type": "RandomAffine",
64
+ "kwargs": {
65
+ "degrees": [
66
+ -5.0,
67
+ 5.0
68
+ ],
69
+ "translate": [
70
+ 0.05,
71
+ 0.05
72
+ ]
73
+ }
74
+ }
75
+ }
76
+ },
77
+ "revision": null,
78
+ "use_imagenet_stats": true,
79
+ "video_backend": "torchcodec",
80
+ "streaming": false,
81
+ "features_to_keep": [
82
+ "observation.state",
83
+ "observation.images.wrist_cam",
84
+ "observation.images.top",
85
+ "observation.images.side",
86
+ "action",
87
+ "episode_index",
88
+ "frame_index",
89
+ "index",
90
+ "timestamp",
91
+ "task_index"
92
+ ],
93
+ "auto_keep_from_policy": false
94
+ },
95
+ "env": null,
96
+ "policy": {
97
+ "type": "act_orin",
98
+ "n_obs_steps": 1,
99
+ "input_features": {
100
+ "observation.state": {
101
+ "type": "STATE",
102
+ "shape": [
103
+ 6
104
+ ]
105
+ },
106
+ "observation.images.wrist_cam": {
107
+ "type": "VISUAL",
108
+ "shape": [
109
+ 480,
110
+ 640,
111
+ 3
112
+ ]
113
+ },
114
+ "observation.images.top": {
115
+ "type": "VISUAL",
116
+ "shape": [
117
+ 480,
118
+ 640,
119
+ 3
120
+ ]
121
+ },
122
+ "observation.images.side": {
123
+ "type": "VISUAL",
124
+ "shape": [
125
+ 848,
126
+ 640,
127
+ 3
128
+ ]
129
+ }
130
+ },
131
+ "output_features": {
132
+ "action": {
133
+ "type": "ACTION",
134
+ "shape": [
135
+ 6
136
+ ]
137
+ }
138
+ },
139
+ "device": "cuda",
140
+ "use_amp": false,
141
+ "use_peft": false,
142
+ "push_to_hub": true,
143
+ "repo_id": "KingLear1606/dm_300",
144
+ "private": null,
145
+ "tags": null,
146
+ "license": null,
147
+ "pretrained_path": null,
148
+ "chunk_size": 100,
149
+ "n_action_steps": 100,
150
+ "normalization_mapping": {
151
+ "VISUAL": "MEAN_STD",
152
+ "STATE": "MEAN_STD",
153
+ "ACTION": "MEAN_STD"
154
+ },
155
+ "vision_backbone": "resnet18",
156
+ "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1",
157
+ "replace_final_stride_with_dilation": false,
158
+ "pre_norm": false,
159
+ "dim_model": 512,
160
+ "n_heads": 8,
161
+ "dim_feedforward": 3200,
162
+ "feedforward_activation": "relu",
163
+ "n_encoder_layers": 4,
164
+ "n_decoder_layers": 1,
165
+ "use_vae": true,
166
+ "latent_dim": 32,
167
+ "n_vae_encoder_layers": 4,
168
+ "temporal_ensemble_coeff": null,
169
+ "dropout": 0.1,
170
+ "kl_weight": 0.1,
171
+ "optimizer_lr": 1e-05,
172
+ "optimizer_weight_decay": 0.0001,
173
+ "optimizer_lr_backbone": 1e-05
174
+ },
175
+ "output_dir": "/home/dc/outputs/train/act_dm_300",
176
+ "job_name": "dm_300",
177
+ "resume": false,
178
+ "seed": 1000,
179
+ "num_workers": 16,
180
+ "batch_size": 8,
181
+ "steps": 100000,
182
+ "eval_freq": 20000,
183
+ "log_freq": 200,
184
+ "tolerance_s": 0.0001,
185
+ "save_checkpoint": true,
186
+ "save_freq": 10000,
187
+ "use_policy_training_preset": false,
188
+ "optimizer": {
189
+ "type": "adamw",
190
+ "lr": 1e-05,
191
+ "weight_decay": 0.0001,
192
+ "grad_clip_norm": 10.0,
193
+ "betas": [
194
+ 0.9,
195
+ 0.999
196
+ ],
197
+ "eps": 1e-08
198
+ },
199
+ "scheduler": {
200
+ "type": "cosine_decay_with_warmup",
201
+ "num_warmup_steps": 1000,
202
+ "num_decay_steps": 100000,
203
+ "peak_lr": 1e-05,
204
+ "decay_lr": 1e-06
205
+ },
206
+ "eval": {
207
+ "n_episodes": 50,
208
+ "batch_size": 50,
209
+ "use_async_envs": false
210
+ },
211
+ "wandb": {
212
+ "enable": true,
213
+ "disable_artifact": true,
214
+ "project": "dm_300",
215
+ "entity": null,
216
+ "notes": null,
217
+ "run_id": "116nz2qd",
218
+ "mode": "offline"
219
+ },
220
+ "peft": null,
221
+ "use_rabc": false,
222
+ "rabc_progress_path": null,
223
+ "rabc_kappa": 0.01,
224
+ "rabc_epsilon": 1e-06,
225
+ "rabc_head_mode": "sparse",
226
+ "rename_map": {},
227
+ "checkpoint_path": null
228
+ }