Hugging Face's logo

continuallearning
/
dit_flow_cl_fft_real_pick_red_block_64_20000

Safetensors
Model card Files
xet
 Community
Ralf Römer commited on
Commit
233a8c1
·
1 Parent(s): 98b7cd4

Initial commit of pretrained policy

Browse files
Files changed (3) hide show
  1. config.json +82 -0
  2. model.safetensors +3 -0
  3. train_config.json +193 -0
config.json ADDED
@@ -0,0 +1,82 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "type": "ditflow_mt",
3
+ "n_obs_steps": 2,
4
+ "normalization_mapping": {
5
+ "VISUAL": "MEAN_STD",
6
+ "STATE": "MIN_MAX",
7
+ "ACTION": "MIN_MAX"
8
+ },
9
+ "input_features": {
10
+ "observation.images.primary": {
11
+ "type": "VISUAL",
12
+ "shape": [
13
+ 3,
14
+ 180,
15
+ 320
16
+ ]
17
+ },
18
+ "observation.images.wrist": {
19
+ "type": "VISUAL",
20
+ "shape": [
21
+ 3,
22
+ 180,
23
+ 320
24
+ ]
25
+ },
26
+ "observation.state.joint": {
27
+ "type": "STATE",
28
+ "shape": [
29
+ 7
30
+ ]
31
+ },
32
+ "observation.state": {
33
+ "type": "STATE",
34
+ "shape": [
35
+ 7
36
+ ]
37
+ }
38
+ },
39
+ "output_features": {
40
+ "action": {
41
+ "type": "ACTION",
42
+ "shape": [
43
+ 7
44
+ ]
45
+ }
46
+ },
47
+ "device": "cuda",
48
+ "use_amp": false,
49
+ "push_to_hub": false,
50
+ "repo_id": null,
51
+ "private": null,
52
+ "tags": null,
53
+ "license": null,
54
+ "horizon": 16,
55
+ "n_action_steps": 8,
56
+ "drop_n_last_frames": 7,
57
+ "tokenizer_max_length": 48,
58
+ "language_model_name": "openai/clip-vit-base-patch32",
59
+ "freeze_language_pretrained": true,
60
+ "vit_name": "facebook/dinov2-base",
61
+ "frequency_embedding_dim": 256,
62
+ "hidden_dim": 512,
63
+ "num_blocks": 6,
64
+ "num_heads": 16,
65
+ "dropout": 0.1,
66
+ "dim_feedforward": 4096,
67
+ "activation": "gelu",
68
+ "training_noise_sampling": "uniform",
69
+ "clip_sample": true,
70
+ "clip_sample_range": 1.0,
71
+ "num_inference_steps": 100,
72
+ "do_mask_loss_for_padding": false,
73
+ "optimizer_lr": 0.0001,
74
+ "optimizer_betas": [
75
+ 0.95,
76
+ 0.999
77
+ ],
78
+ "optimizer_eps": 1e-08,
79
+ "optimizer_weight_decay": 1e-06,
80
+ "scheduler_name": "cosine",
81
+ "scheduler_warmup_steps": 500
82
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b70dfc6bc317d9026a29a47d85298ad9e9dfe544132eb64cf3f87823f10af656
3
+ size 774573860
train_config.json ADDED
@@ -0,0 +1,193 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset": {
3
+ "repo_id": "continuallearning/pick-red-block",
4
+ "root": null,
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
+ }
62
+ },
63
+ "revision": null,
64
+ "use_imagenet_stats": true,
65
+ "video_backend": "torchcodec"
66
+ },
67
+ "env": null,
68
+ "policy": {
69
+ "type": "ditflow_mt",
70
+ "n_obs_steps": 2,
71
+ "normalization_mapping": {
72
+ "VISUAL": "MEAN_STD",
73
+ "STATE": "MIN_MAX",
74
+ "ACTION": "MIN_MAX"
75
+ },
76
+ "input_features": {
77
+ "observation.images.primary": {
78
+ "type": "VISUAL",
79
+ "shape": [
80
+ 3,
81
+ 180,
82
+ 320
83
+ ]
84
+ },
85
+ "observation.images.wrist": {
86
+ "type": "VISUAL",
87
+ "shape": [
88
+ 3,
89
+ 180,
90
+ 320
91
+ ]
92
+ },
93
+ "observation.state.joint": {
94
+ "type": "STATE",
95
+ "shape": [
96
+ 7
97
+ ]
98
+ },
99
+ "observation.state": {
100
+ "type": "STATE",
101
+ "shape": [
102
+ 7
103
+ ]
104
+ }
105
+ },
106
+ "output_features": {
107
+ "action": {
108
+ "type": "ACTION",
109
+ "shape": [
110
+ 7
111
+ ]
112
+ }
113
+ },
114
+ "device": "cuda",
115
+ "use_amp": false,
116
+ "push_to_hub": false,
117
+ "repo_id": null,
118
+ "private": null,
119
+ "tags": null,
120
+ "license": null,
121
+ "horizon": 16,
122
+ "n_action_steps": 8,
123
+ "drop_n_last_frames": 7,
124
+ "tokenizer_max_length": 48,
125
+ "language_model_name": "openai/clip-vit-base-patch32",
126
+ "freeze_language_pretrained": true,
127
+ "vit_name": "facebook/dinov2-base",
128
+ "frequency_embedding_dim": 256,
129
+ "hidden_dim": 512,
130
+ "num_blocks": 6,
131
+ "num_heads": 16,
132
+ "dropout": 0.1,
133
+ "dim_feedforward": 4096,
134
+ "activation": "gelu",
135
+ "training_noise_sampling": "uniform",
136
+ "clip_sample": true,
137
+ "clip_sample_range": 1.0,
138
+ "num_inference_steps": 100,
139
+ "do_mask_loss_for_padding": false,
140
+ "optimizer_lr": 0.0001,
141
+ "optimizer_betas": [
142
+ 0.95,
143
+ 0.999
144
+ ],
145
+ "optimizer_eps": 1e-08,
146
+ "optimizer_weight_decay": 1e-06,
147
+ "scheduler_name": "cosine",
148
+ "scheduler_warmup_steps": 500
149
+ },
150
+ "output_dir": "outputs/dit_flow_cl_fft_real_pick_red_block_64_50000",
151
+ "job_name": "dit_flow_cl_fft_real_pick_red_block_64_50000",
152
+ "resume": false,
153
+ "seed": 1000,
154
+ "num_workers": 16,
155
+ "batch_size": 64,
156
+ "steps": 50000,
157
+ "eval_freq": 0,
158
+ "log_freq": 100,
159
+ "save_checkpoint": true,
160
+ "save_freq": 10000,
161
+ "use_policy_training_preset": true,
162
+ "optimizer": {
163
+ "type": "adam",
164
+ "lr": 0.0001,
165
+ "weight_decay": 1e-06,
166
+ "grad_clip_norm": 10.0,
167
+ "betas": [
168
+ 0.95,
169
+ 0.999
170
+ ],
171
+ "eps": 1e-08
172
+ },
173
+ "scheduler": {
174
+ "type": "diffuser",
175
+ "num_warmup_steps": 500,
176
+ "name": "cosine"
177
+ },
178
+ "eval": {
179
+ "n_episodes": 50,
180
+ "batch_size": 50,
181
+ "use_async_envs": false,
182
+ "max_episodes_rendered": 4
183
+ },
184
+ "wandb": {
185
+ "enable": true,
186
+ "disable_artifact": true,
187
+ "project": "paper",
188
+ "entity": "470620104-technical-university-of-munich",
189
+ "notes": null,
190
+ "run_id": "h4bkpycs",
191
+ "mode": null
192
+ }
193
+ }