[docs] update dataset card
Browse files
README.md
CHANGED
@@ -14,7 +14,7 @@ task_categories:
|
|
14 |
|
15 |
## Description
|
16 |
|
17 |
-
This is the raw dataset used in the paper
|
18 |
[Reactive Diffusion Policy: Slow-Fast Visual-Tactile Policy Learning for Contact-Rich Manipulation](https://reactive-diffusion-policy.github.io).
|
19 |
|
20 |
- [Paper](https://arxiv.org/pdf/2503.02881)
|
@@ -26,19 +26,46 @@ This is the raw dataset used in the paper
|
|
26 |
We offer two versions of the dataset:
|
27 |
one is the [full dataset](https://huggingface.co/datasets/WendiChen/reactive_diffusion_policy_dataset/tree/main/dataset_full) used to train the models in our paper,
|
28 |
and the other is a [mini dataset](https://huggingface.co/datasets/WendiChen/reactive_diffusion_policy_dataset/tree/main/dataset_mini) for easier examination.
|
29 |
-
Both versions include subsets of peeling, wiping and lifting.
|
30 |
-
Each subset is structured as follows:
|
31 |
|
|
|
32 |
```
|
33 |
subset_name
|
34 |
βββ seq_01.pkl
|
35 |
βββ seq_02.pkl
|
36 |
βββ ...
|
37 |
```
|
38 |
-
Note that we split the full lifting subset into 2 parts due to file size restrictions.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
39 |
|
40 |
## Usage
|
41 |
|
42 |
Follow the [README in our GitHub repo](https://github.com/xiaoxiaoxh/reactive_diffusion_policy)
|
43 |
-
to [postprocess the data](https://github.com/xiaoxiaoxh/reactive_diffusion_policy#data-postprocessing)
|
44 |
and [train the model](https://github.com/xiaoxiaoxh/reactive_diffusion_policy#-training).
|
|
|
14 |
|
15 |
## Description
|
16 |
|
17 |
+
This is the raw and postprocessed dataset used in the paper
|
18 |
[Reactive Diffusion Policy: Slow-Fast Visual-Tactile Policy Learning for Contact-Rich Manipulation](https://reactive-diffusion-policy.github.io).
|
19 |
|
20 |
- [Paper](https://arxiv.org/pdf/2503.02881)
|
|
|
26 |
We offer two versions of the dataset:
|
27 |
one is the [full dataset](https://huggingface.co/datasets/WendiChen/reactive_diffusion_policy_dataset/tree/main/dataset_full) used to train the models in our paper,
|
28 |
and the other is a [mini dataset](https://huggingface.co/datasets/WendiChen/reactive_diffusion_policy_dataset/tree/main/dataset_mini) for easier examination.
|
29 |
+
Both versions include raw and postprocessed subsets of peeling, wiping and lifting.
|
|
|
30 |
|
31 |
+
Each raw subset is structured as follows:
|
32 |
```
|
33 |
subset_name
|
34 |
βββ seq_01.pkl
|
35 |
βββ seq_02.pkl
|
36 |
βββ ...
|
37 |
```
|
38 |
+
> Note that we split the full raw lifting subset into 2 parts due to file size restrictions.
|
39 |
+
|
40 |
+
Each postprocessed subset is stored in [Zarr](https://zarr.dev) format, which is structured as follows:
|
41 |
+
```
|
42 |
+
βββ action (25710, 4) float32
|
43 |
+
βββ external_img (25710, 240, 320, 3) uint8
|
44 |
+
βββ left_gripper1_img (25710, 240, 320, 3) uint8
|
45 |
+
βββ left_gripper1_initial_marker (25710, 63, 2) float32
|
46 |
+
βββ left_gripper1_marker_offset (25710, 63, 2) float32
|
47 |
+
βββ left_gripper1_marker_offset_emb (25710, 15) float32
|
48 |
+
βββ left_gripper2_img (25710, 240, 320, 3) uint8
|
49 |
+
βββ left_gripper2_initial_marker (25710, 25, 2) float32
|
50 |
+
βββ left_gripper2_marker_offset (25710, 25, 2) float32
|
51 |
+
βββ left_gripper2_marker_offset_emb (25710, 15) float32
|
52 |
+
βββ left_robot_gripper_force (25710, 1) float32
|
53 |
+
βββ left_robot_gripper_width (25710, 1) float32
|
54 |
+
βββ left_robot_tcp_pose (25710, 9) float32
|
55 |
+
βββ left_robot_tcp_vel (25710, 6) float32
|
56 |
+
βββ left_robot_tcp_wrench (25710, 6) float32
|
57 |
+
βββ left_wrist_img (25710, 240, 320, 3) uint8
|
58 |
+
βββ right_robot_gripper_force (25710, 1) float32
|
59 |
+
βββ right_robot_gripper_width (25710, 1) float32
|
60 |
+
βββ right_robot_tcp_pose (25710, 9) float32
|
61 |
+
βββ right_robot_tcp_vel (25710, 6) float32
|
62 |
+
βββ right_robot_tcp_wrench (25710, 6) float32
|
63 |
+
βββ target (25710, 4) float32
|
64 |
+
βββ timestamp (25710,) float32
|
65 |
+
```
|
66 |
|
67 |
## Usage
|
68 |
|
69 |
Follow the [README in our GitHub repo](https://github.com/xiaoxiaoxh/reactive_diffusion_policy)
|
70 |
+
to [postprocess the raw data](https://github.com/xiaoxiaoxh/reactive_diffusion_policy#data-postprocessing)
|
71 |
and [train the model](https://github.com/xiaoxiaoxh/reactive_diffusion_policy#-training).
|