Datasets:

Formats:
parquet
ArXiv:
Libraries:
Datasets
Dask
License:
aliberts HF staff commited on
Commit
d2fedb2
1 Parent(s): 946129f

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +123 -31
README.md CHANGED
@@ -1,16 +1,30 @@
1
  ---
 
2
  task_categories:
3
  - robotics
4
  tags:
5
  - LeRobot
 
 
 
6
  ---
 
7
  This dataset was created using [LeRobot](https://github.com/huggingface/lerobot).
8
 
9
- [meta/info.json](meta/info.json)
 
 
 
 
 
 
 
 
 
 
10
  ```json
11
  {
12
  "codebase_version": "v2.0",
13
- "data_path": "data/chunk-{episode_chunk:03d}/episode_{episode_index:06d}.parquet",
14
  "robot_type": "unknown",
15
  "total_episodes": 800,
16
  "total_frames": 20000,
@@ -22,36 +36,114 @@ This dataset was created using [LeRobot](https://github.com/huggingface/lerobot)
22
  "splits": {
23
  "train": "0:800"
24
  },
25
- "keys": [
26
- "observation.state",
27
- "action"
28
- ],
29
- "video_keys": [],
30
- "image_keys": [
31
- "observation.image"
32
- ],
33
- "shapes": {
34
- "observation.state": 4,
35
- "action": 3,
36
  "observation.image": {
37
- "width": 84,
38
- "height": 84,
39
- "channels": 3
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
40
  }
41
- },
42
- "names": {
43
- "observation.state": [
44
- "motor_0",
45
- "motor_1",
46
- "motor_2",
47
- "motor_3"
48
- ],
49
- "action": [
50
- "motor_0",
51
- "motor_1",
52
- "motor_2"
53
- ]
54
- },
55
- "videos": null
 
56
  }
 
57
  ```
 
1
  ---
2
+ license: mit
3
  task_categories:
4
  - robotics
5
  tags:
6
  - LeRobot
7
+ configs:
8
+ - config_name: default
9
+ data_files: data/*/*.parquet
10
  ---
11
+
12
  This dataset was created using [LeRobot](https://github.com/huggingface/lerobot).
13
 
14
+ ## Dataset Description
15
+
16
+
17
+
18
+ - **Homepage:** https://www.nicklashansen.com/td-mpc/
19
+ - **Paper:** https://arxiv.org/abs/2203.04955
20
+ - **License:** mit
21
+
22
+ ## Dataset Structure
23
+
24
+ [meta/info.json](meta/info.json):
25
  ```json
26
  {
27
  "codebase_version": "v2.0",
 
28
  "robot_type": "unknown",
29
  "total_episodes": 800,
30
  "total_frames": 20000,
 
36
  "splits": {
37
  "train": "0:800"
38
  },
39
+ "data_path": "data/chunk-{episode_chunk:03d}/episode_{episode_index:06d}.parquet",
40
+ "video_path": null,
41
+ "features": {
 
 
 
 
 
 
 
 
42
  "observation.image": {
43
+ "dtype": "image",
44
+ "shape": [
45
+ 84,
46
+ 84,
47
+ 3
48
+ ],
49
+ "names": [
50
+ "height",
51
+ "width",
52
+ "channel"
53
+ ]
54
+ },
55
+ "observation.state": {
56
+ "dtype": "float32",
57
+ "shape": [
58
+ 4
59
+ ],
60
+ "names": {
61
+ "motors": [
62
+ "motor_0",
63
+ "motor_1",
64
+ "motor_2",
65
+ "motor_3"
66
+ ]
67
+ }
68
+ },
69
+ "action": {
70
+ "dtype": "float32",
71
+ "shape": [
72
+ 3
73
+ ],
74
+ "names": {
75
+ "motors": [
76
+ "motor_0",
77
+ "motor_1",
78
+ "motor_2"
79
+ ]
80
+ }
81
+ },
82
+ "episode_index": {
83
+ "dtype": "int64",
84
+ "shape": [
85
+ 1
86
+ ],
87
+ "names": null
88
+ },
89
+ "frame_index": {
90
+ "dtype": "int64",
91
+ "shape": [
92
+ 1
93
+ ],
94
+ "names": null
95
+ },
96
+ "timestamp": {
97
+ "dtype": "float32",
98
+ "shape": [
99
+ 1
100
+ ],
101
+ "names": null
102
+ },
103
+ "next.reward": {
104
+ "dtype": "float32",
105
+ "shape": [
106
+ 1
107
+ ],
108
+ "names": null
109
+ },
110
+ "next.done": {
111
+ "dtype": "bool",
112
+ "shape": [
113
+ 1
114
+ ],
115
+ "names": null
116
+ },
117
+ "index": {
118
+ "dtype": "int64",
119
+ "shape": [
120
+ 1
121
+ ],
122
+ "names": null
123
+ },
124
+ "task_index": {
125
+ "dtype": "int64",
126
+ "shape": [
127
+ 1
128
+ ],
129
+ "names": null
130
  }
131
+ }
132
+ }
133
+ ```
134
+
135
+
136
+ ## Citation
137
+
138
+ **BibTeX:**
139
+
140
+ ```bibtex
141
+
142
+ @inproceedings{Hansen2022tdmpc,
143
+ title={Temporal Difference Learning for Model Predictive Control},
144
+ author={Nicklas Hansen and Xiaolong Wang and Hao Su},
145
+ booktitle={ICML},
146
+ year={2022}
147
  }
148
+
149
  ```