Datasets:
Update README.md
Browse files
README.md
CHANGED
|
@@ -1,6 +1,51 @@
|
|
| 1 |
---
|
| 2 |
license: apache-2.0
|
|
|
|
|
|
|
| 3 |
---
|
| 4 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 5 |
- We release all of the text description data `text.tar.gz`; the trajectory data `data.tar.gz` with robot states, actions, DoF position, global position and global orientation informations.
|
| 6 |
-
- We release the train set split `train.txt`
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
license: apache-2.0
|
| 3 |
+
task_categories:
|
| 4 |
+
- robotics
|
| 5 |
---
|
| 6 |
|
| 7 |
+
<div align="center">
|
| 8 |
+
<h1> ALMI-X </h1>
|
| 9 |
+
</div>
|
| 10 |
+
<h5 align="center">
|
| 11 |
+
<a href="#"> coming soon :) ...</a>
|
| 12 |
+
</h5>
|
| 13 |
+
|
| 14 |
+
# Introduction of Dataset
|
| 15 |
+
|
| 16 |
- We release all of the text description data `text.tar.gz`; the trajectory data `data.tar.gz` with robot states, actions, DoF position, global position and global orientation informations.
|
| 17 |
+
- We release the train set split `train.txt`
|
| 18 |
+
|
| 19 |
+
Here we offer a simple demo code to introduce the data formats in the dataset:
|
| 20 |
+
|
| 21 |
+
``` python
|
| 22 |
+
import numpy as np
|
| 23 |
+
|
| 24 |
+
data = np.load("data_path"+"/xxx.npy", allow_pickle=True)
|
| 25 |
+
data.item()['obs'] # [frame_nums, 71]
|
| 26 |
+
data.item()['actions'] # [frame_nums, 21]
|
| 27 |
+
data.item()['dof_pos'] # [frame_nums, 21]
|
| 28 |
+
data.item()['root_trans'] # [frame_nums, 3]
|
| 29 |
+
data.item()['root_rot'] # [frame_nums, 4]
|
| 30 |
+
|
| 31 |
+
```
|
| 32 |
+
|
| 33 |
+
# Dataset Statistics
|
| 34 |
+
Percentage of steps for different categories of motions before and after data augmentation.
|
| 35 |
+
<br>
|
| 36 |
+

|
| 37 |
+
The visualization of $x-y$ coordinates of the robot for each step in the dataset. We down-sample the data for visualization.
|
| 38 |
+
<br>
|
| 39 |
+

|
| 40 |
+
|
| 41 |
+
# Dataset Collection Pipeline
|
| 42 |
+
|
| 43 |
+
coming soon...
|
| 44 |
+
|
| 45 |
+
# Citation
|
| 46 |
+
|
| 47 |
+
If you find our work helpful, please cite us:
|
| 48 |
+
|
| 49 |
+
```bibtex
|
| 50 |
+
coming soon...
|
| 51 |
+
}
|