Hugging Face's logo

pravsels
/
molmoact2_block_stack_quantile_norm_fix_25k

Robotics
LeRobot
Safetensors
molmoact2
vla
Model card Files
xet
 Community
pravsels commited on
Commit
2cfc37f
·
verified ·
1 Parent(s): 479ae7c

Upload MolmoAct2 025000 to repo root

Browse files
Files changed (7) hide show
  1. config.json +139 -0
  2. model.safetensors +3 -0
  3. policy_postprocessor.json +36 -0
  4. policy_postprocessor_step_1_molmoact2_masked_unnormalizer.safetensors +3 -0
  5. policy_preprocessor.json +102 -0
  6. policy_preprocessor_step_2_molmoact2_masked_normalizer.safetensors +3 -0
  7. train_config.json +277 -0
config.json ADDED
@@ -0,0 +1,139 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "type": "molmoact2",
3
+ "n_obs_steps": 1,
4
+ "input_features": {
5
+ "observation.state": {
6
+ "type": "STATE",
7
+ "shape": [
8
+ 6
9
+ ]
10
+ },
11
+ "observation.images.front": {
12
+ "type": "VISUAL",
13
+ "shape": [
14
+ 3,
15
+ 576,
16
+ 1024
17
+ ]
18
+ },
19
+ "observation.images.top": {
20
+ "type": "VISUAL",
21
+ "shape": [
22
+ 3,
23
+ 576,
24
+ 1024
25
+ ]
26
+ },
27
+ "observation.images.wrist": {
28
+ "type": "VISUAL",
29
+ "shape": [
30
+ 3,
31
+ 720,
32
+ 1280
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": false,
48
+ "repo_id": null,
49
+ "private": null,
50
+ "tags": null,
51
+ "license": null,
52
+ "pretrained_path": null,
53
+ "checkpoint_path": "allenai/MolmoAct2",
54
+ "checkpoint_revision": null,
55
+ "checkpoint_force_download": false,
56
+ "chunk_size": 30,
57
+ "n_action_steps": 30,
58
+ "action_mode": "continuous",
59
+ "inference_action_mode": "continuous",
60
+ "discrete_action_tokenizer": "allenai/MolmoAct2-FAST-Tokenizer",
61
+ "discrete_generation_max_steps": null,
62
+ "norm_tag": null,
63
+ "setup_type": "single so100/so101 robotic arm in molmoact2",
64
+ "control_mode": "absolute joint pose",
65
+ "image_keys": [
66
+ "observation.images.top",
67
+ "observation.images.wrist",
68
+ "observation.images.front"
69
+ ],
70
+ "normalize_language": true,
71
+ "add_setup_tokens": true,
72
+ "add_control_tokens": true,
73
+ "normalize_gripper": true,
74
+ "num_state_tokens": 256,
75
+ "max_sequence_length": null,
76
+ "expected_max_action_dim": 32,
77
+ "num_flow_timesteps": 8,
78
+ "flow_matching_cutoff": 1.0,
79
+ "flow_matching_time_offset": 0.001,
80
+ "flow_matching_time_scale": 0.999,
81
+ "flow_matching_beta_alpha": 1.0,
82
+ "flow_matching_beta_beta": 1.5,
83
+ "num_inference_steps": null,
84
+ "mask_action_dim_padding": true,
85
+ "enable_inference_cuda_graph": true,
86
+ "per_episode_seed": false,
87
+ "eval_seed": null,
88
+ "rtc_config": null,
89
+ "train_mode_vlm": "freeze",
90
+ "lora_rank": 64,
91
+ "lora_alpha": 16,
92
+ "lora_dropout": 0.05,
93
+ "lora_bias": "none",
94
+ "enable_knowledge_insulation": false,
95
+ "freeze_embedding": true,
96
+ "gradient_checkpointing": false,
97
+ "model_dtype": "bfloat16",
98
+ "llm_residual_dropout": 0.1,
99
+ "softmax_auxiliary_loss": true,
100
+ "softmax_auxiliary_loss_scale": 0.0001,
101
+ "discrete_loss_token_weighting": "root_subsegments_root_tokens",
102
+ "optimizer_lr": 1e-05,
103
+ "optimizer_vit_lr": 5e-06,
104
+ "optimizer_connector_lr": 5e-06,
105
+ "optimizer_action_expert_lr": 5e-05,
106
+ "optimizer_betas": [
107
+ 0.9,
108
+ 0.95
109
+ ],
110
+ "optimizer_eps": 1e-06,
111
+ "optimizer_weight_decay": 0.0,
112
+ "optimizer_grad_clip_norm": 1.0,
113
+ "scheduler_warmup_steps": 200,
114
+ "scheduler_decay_steps": 30000,
115
+ "scheduler_decay_lr": 1e-06,
116
+ "normalization_mapping": {
117
+ "ACTION": "QUANTILES",
118
+ "STATE": "QUANTILES",
119
+ "VISUAL": "IDENTITY"
120
+ },
121
+ "dataset_feature_names": {
122
+ "action": [
123
+ "shoulder_pan.pos",
124
+ "shoulder_lift.pos",
125
+ "elbow_flex.pos",
126
+ "wrist_flex.pos",
127
+ "wrist_roll.pos",
128
+ "gripper.pos"
129
+ ],
130
+ "observation.state": [
131
+ "shoulder_pan.pos",
132
+ "shoulder_lift.pos",
133
+ "elbow_flex.pos",
134
+ "wrist_flex.pos",
135
+ "wrist_roll.pos",
136
+ "gripper.pos"
137
+ ]
138
+ }
139
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e4f71871567b158cecdc5762dc90f0fff98aa79087141754e7cb6d87bc928ea3
3
+ size 10884573720
policy_postprocessor.json ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "name": "policy_postprocessor",
3
+ "steps": [
4
+ {
5
+ "registry_name": "molmoact2_clamp_action",
6
+ "config": {}
7
+ },
8
+ {
9
+ "registry_name": "molmoact2_masked_unnormalizer",
10
+ "config": {
11
+ "eps": 1e-08,
12
+ "features": {
13
+ "action": {
14
+ "type": "ACTION",
15
+ "shape": [
16
+ 6
17
+ ]
18
+ }
19
+ },
20
+ "norm_map": {
21
+ "ACTION": "QUANTILES",
22
+ "STATE": "QUANTILES",
23
+ "VISUAL": "IDENTITY"
24
+ }
25
+ },
26
+ "state_file": "policy_postprocessor_step_1_molmoact2_masked_unnormalizer.safetensors"
27
+ },
28
+ {
29
+ "registry_name": "device_processor",
30
+ "config": {
31
+ "device": "cpu",
32
+ "float_dtype": null
33
+ }
34
+ }
35
+ ]
36
+ }
policy_postprocessor_step_1_molmoact2_masked_unnormalizer.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:318f69ba8b9a693fc37e49dc356e3d343c56ed25de9d0ebf944d4f61664eb36d
3
+ size 8728
policy_preprocessor.json ADDED
@@ -0,0 +1,102 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "name": "policy_preprocessor",
3
+ "steps": [
4
+ {
5
+ "registry_name": "rename_observations_processor",
6
+ "config": {
7
+ "rename_map": {}
8
+ }
9
+ },
10
+ {
11
+ "registry_name": "to_batch_processor",
12
+ "config": {}
13
+ },
14
+ {
15
+ "registry_name": "molmoact2_masked_normalizer",
16
+ "config": {
17
+ "eps": 1e-08,
18
+ "features": {
19
+ "observation.state": {
20
+ "type": "STATE",
21
+ "shape": [
22
+ 6
23
+ ]
24
+ },
25
+ "observation.images.front": {
26
+ "type": "VISUAL",
27
+ "shape": [
28
+ 3,
29
+ 576,
30
+ 1024
31
+ ]
32
+ },
33
+ "observation.images.top": {
34
+ "type": "VISUAL",
35
+ "shape": [
36
+ 3,
37
+ 576,
38
+ 1024
39
+ ]
40
+ },
41
+ "observation.images.wrist": {
42
+ "type": "VISUAL",
43
+ "shape": [
44
+ 3,
45
+ 720,
46
+ 1280
47
+ ]
48
+ },
49
+ "action": {
50
+ "type": "ACTION",
51
+ "shape": [
52
+ 6
53
+ ]
54
+ }
55
+ },
56
+ "norm_map": {
57
+ "ACTION": "QUANTILES",
58
+ "STATE": "QUANTILES",
59
+ "VISUAL": "IDENTITY"
60
+ }
61
+ },
62
+ "state_file": "policy_preprocessor_step_2_molmoact2_masked_normalizer.safetensors"
63
+ },
64
+ {
65
+ "registry_name": "molmoact2_clamp_normalized",
66
+ "config": {}
67
+ },
68
+ {
69
+ "registry_name": "molmoact2_pack_inputs",
70
+ "config": {
71
+ "checkpoint_path": "allenai/MolmoAct2",
72
+ "checkpoint_revision": null,
73
+ "checkpoint_force_download": false,
74
+ "action_mode": "continuous",
75
+ "discrete_action_tokenizer": "allenai/MolmoAct2-FAST-Tokenizer",
76
+ "image_keys": [
77
+ "observation.images.top",
78
+ "observation.images.wrist",
79
+ "observation.images.front"
80
+ ],
81
+ "allow_image_key_fallback": false,
82
+ "setup_type": "single so100/so101 robotic arm in molmoact2",
83
+ "control_mode": "absolute joint pose",
84
+ "normalize_language": true,
85
+ "add_setup_tokens": true,
86
+ "add_control_tokens": true,
87
+ "num_state_tokens": 256,
88
+ "max_sequence_length": null,
89
+ "chunk_size": 30,
90
+ "max_action_dim": 32,
91
+ "env_action_dim": 6
92
+ }
93
+ },
94
+ {
95
+ "registry_name": "device_processor",
96
+ "config": {
97
+ "device": "cuda",
98
+ "float_dtype": null
99
+ }
100
+ }
101
+ ]
102
+ }
policy_preprocessor_step_2_molmoact2_masked_normalizer.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d044cc364fc7a3df5db57679d44465b372830130feadbb69ca385fa749e5d0cb
3
+ size 8744
train_config.json ADDED
@@ -0,0 +1,277 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset": {
3
+ "repo_id": "villekuosmanen/armnetbench_block_stack",
4
+ "root": null,
5
+ "episodes": null,
6
+ "image_transforms": {
7
+ "enable": true,
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": "pyav",
80
+ "return_uint8": false,
81
+ "streaming": false
82
+ },
83
+ "env": null,
84
+ "policy": {
85
+ "type": "molmoact2",
86
+ "n_obs_steps": 1,
87
+ "input_features": {
88
+ "observation.state": {
89
+ "type": "STATE",
90
+ "shape": [
91
+ 6
92
+ ]
93
+ },
94
+ "observation.images.front": {
95
+ "type": "VISUAL",
96
+ "shape": [
97
+ 3,
98
+ 576,
99
+ 1024
100
+ ]
101
+ },
102
+ "observation.images.top": {
103
+ "type": "VISUAL",
104
+ "shape": [
105
+ 3,
106
+ 576,
107
+ 1024
108
+ ]
109
+ },
110
+ "observation.images.wrist": {
111
+ "type": "VISUAL",
112
+ "shape": [
113
+ 3,
114
+ 720,
115
+ 1280
116
+ ]
117
+ }
118
+ },
119
+ "output_features": {
120
+ "action": {
121
+ "type": "ACTION",
122
+ "shape": [
123
+ 6
124
+ ]
125
+ }
126
+ },
127
+ "device": "cuda",
128
+ "use_amp": false,
129
+ "use_peft": false,
130
+ "push_to_hub": false,
131
+ "repo_id": null,
132
+ "private": null,
133
+ "tags": null,
134
+ "license": null,
135
+ "pretrained_path": null,
136
+ "checkpoint_path": "allenai/MolmoAct2",
137
+ "checkpoint_revision": null,
138
+ "checkpoint_force_download": false,
139
+ "chunk_size": 30,
140
+ "n_action_steps": 30,
141
+ "action_mode": "continuous",
142
+ "inference_action_mode": "continuous",
143
+ "discrete_action_tokenizer": "allenai/MolmoAct2-FAST-Tokenizer",
144
+ "discrete_generation_max_steps": null,
145
+ "norm_tag": null,
146
+ "setup_type": "single so100/so101 robotic arm in molmoact2",
147
+ "control_mode": "absolute joint pose",
148
+ "image_keys": [
149
+ "observation.images.top",
150
+ "observation.images.wrist",
151
+ "observation.images.front"
152
+ ],
153
+ "normalize_language": true,
154
+ "add_setup_tokens": true,
155
+ "add_control_tokens": true,
156
+ "normalize_gripper": true,
157
+ "num_state_tokens": 256,
158
+ "max_sequence_length": null,
159
+ "expected_max_action_dim": 32,
160
+ "num_flow_timesteps": 8,
161
+ "flow_matching_cutoff": 1.0,
162
+ "flow_matching_time_offset": 0.001,
163
+ "flow_matching_time_scale": 0.999,
164
+ "flow_matching_beta_alpha": 1.0,
165
+ "flow_matching_beta_beta": 1.5,
166
+ "num_inference_steps": null,
167
+ "mask_action_dim_padding": true,
168
+ "enable_inference_cuda_graph": true,
169
+ "per_episode_seed": false,
170
+ "eval_seed": null,
171
+ "rtc_config": null,
172
+ "train_mode_vlm": "freeze",
173
+ "lora_rank": 64,
174
+ "lora_alpha": 16,
175
+ "lora_dropout": 0.05,
176
+ "lora_bias": "none",
177
+ "enable_knowledge_insulation": false,
178
+ "freeze_embedding": true,
179
+ "gradient_checkpointing": false,
180
+ "model_dtype": "bfloat16",
181
+ "llm_residual_dropout": 0.1,
182
+ "softmax_auxiliary_loss": true,
183
+ "softmax_auxiliary_loss_scale": 0.0001,
184
+ "discrete_loss_token_weighting": "root_subsegments_root_tokens",
185
+ "optimizer_lr": 1e-05,
186
+ "optimizer_vit_lr": 5e-06,
187
+ "optimizer_connector_lr": 5e-06,
188
+ "optimizer_action_expert_lr": 5e-05,
189
+ "optimizer_betas": [
190
+ 0.9,
191
+ 0.95
192
+ ],
193
+ "optimizer_eps": 1e-06,
194
+ "optimizer_weight_decay": 0.0,
195
+ "optimizer_grad_clip_norm": 1.0,
196
+ "scheduler_warmup_steps": 200,
197
+ "scheduler_decay_steps": 30000,
198
+ "scheduler_decay_lr": 1e-06,
199
+ "normalization_mapping": {
200
+ "ACTION": "QUANTILES",
201
+ "STATE": "QUANTILES",
202
+ "VISUAL": "IDENTITY"
203
+ },
204
+ "dataset_feature_names": {
205
+ "action": [
206
+ "shoulder_pan.pos",
207
+ "shoulder_lift.pos",
208
+ "elbow_flex.pos",
209
+ "wrist_flex.pos",
210
+ "wrist_roll.pos",
211
+ "gripper.pos"
212
+ ],
213
+ "observation.state": [
214
+ "shoulder_pan.pos",
215
+ "shoulder_lift.pos",
216
+ "elbow_flex.pos",
217
+ "wrist_flex.pos",
218
+ "wrist_roll.pos",
219
+ "gripper.pos"
220
+ ]
221
+ }
222
+ },
223
+ "reward_model": null,
224
+ "output_dir": "/scratch/u6kr/pravsels.u6kr/molmoact2/checkpoints/molmoact2_block_stack_base_quantile_norm_fix",
225
+ "job_name": "molmoact2_block_stack_base_quantile_norm_fix",
226
+ "resume": false,
227
+ "seed": 1000,
228
+ "cudnn_deterministic": false,
229
+ "num_workers": 16,
230
+ "batch_size": 8,
231
+ "prefetch_factor": 4,
232
+ "persistent_workers": true,
233
+ "steps": 30000,
234
+ "eval_freq": -1,
235
+ "log_freq": 20,
236
+ "tolerance_s": 0.0001,
237
+ "save_checkpoint": true,
238
+ "save_freq": 5000,
239
+ "use_policy_training_preset": true,
240
+ "optimizer": {
241
+ "type": "adamw",
242
+ "lr": 1e-05,
243
+ "weight_decay": 0.0,
244
+ "grad_clip_norm": 1.0,
245
+ "betas": [
246
+ 0.9,
247
+ 0.95
248
+ ],
249
+ "eps": 1e-06
250
+ },
251
+ "scheduler": {
252
+ "type": "molmoact2_cosine_decay_with_warmup",
253
+ "num_warmup_steps": 200,
254
+ "num_decay_steps": 30000,
255
+ "peak_lr": 1e-05,
256
+ "decay_lr": 1e-06
257
+ },
258
+ "eval": {
259
+ "n_episodes": 50,
260
+ "batch_size": 50,
261
+ "use_async_envs": true
262
+ },
263
+ "wandb": {
264
+ "enable": true,
265
+ "disable_artifact": false,
266
+ "project": "molmoact2_block_stack",
267
+ "entity": "pravsels",
268
+ "notes": null,
269
+ "run_id": "pir9oxln",
270
+ "mode": "offline",
271
+ "add_tags": true
272
+ },
273
+ "peft": null,
274
+ "sample_weighting": null,
275
+ "rename_map": {},
276
+ "checkpoint_path": null
277
+ }