jmonas commited on
Commit
94b418d
·
verified ·
1 Parent(s): df59132

Update README.md (#4)

Browse files

- Update README.md (95853906a54556d028711a792f537c92aff405a1)

Files changed (1) hide show
  1. README.md +41 -2
README.md CHANGED
@@ -12,7 +12,46 @@ Download with:
12
  huggingface-cli download 1x-technologies/worldmodel --repo-type dataset --local-dir data
13
  ```
14
 
15
- Current version: v1.1
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
16
 
17
  - **magvit2.ckpt** - weights for [MAGVIT2](https://github.com/TencentARC/Open-MAGVIT2) image tokenizer we used. We provide the encoder (tokenizer) and decoder (de-tokenizer) weights.
18
 
@@ -25,7 +64,7 @@ Contents of train/val_v1.1:
25
  - **neck_desired** `(N, 1)`: Desired neck pitch.
26
  - **l_hand_closure** `(N, 1)`: Left hand closure state (0 = open, 1 = closed).
27
  - **r_hand_closure** `(N, 1)`: Right hand closure state (0 = open, 1 = closed).
28
- #### Index-to-Joint Mapping
29
  ```
30
  {
31
  0: HIP_YAW
 
12
  huggingface-cli download 1x-technologies/worldmodel --repo-type dataset --local-dir data
13
  ```
14
 
15
+ Current version: v2.0
16
+
17
+ Contents of train/val_v2.0
18
+ - **videos** -
19
+ - **segment_indicies** - For video `i` and frame `j`, `segment_idx_i[j]` uniquely points to the segment index that frame `j` came from. You may want to use this to separate non-contiguous frames from different videos (transitions).
20
+ - **robot_states** - States arrays defined in `Index-to-Joint Mapping` stored in `np.float32` format. For video `i` and frame `j`, the corresponding state is given by `states_i[j]`.
21
+ - **metadata** - The `metadata.json` file provides high-level information about the entire dataset, while `metadata_[i].json` files contain specific details for each individual video `i`.
22
+
23
+ #### Index-to-State Mapping (NEW)
24
+ ```
25
+ {
26
+ 0: HIP_YAW
27
+ 1: HIP_ROLL
28
+ 2: HIP_PITCH
29
+ 3: KNEE_PITCH
30
+ 4: ANKLE_ROLL
31
+ 5: ANKLE_PITCH
32
+ 6: LEFT_SHOULDER_PITCH
33
+ 7: LEFT_SHOULDER_ROLL
34
+ 8: LEFT_SHOULDER_YAW
35
+ 9: LEFT_ELBOW_PITCH
36
+ 10: LEFT_ELBOW_YAW
37
+ 11: LEFT_WRIST_PITCH
38
+ 12: LEFT_WRIST_ROLL
39
+ 13: RIGHT_SHOULDER_PITCH
40
+ 14: RIGHT_SHOULDER_ROLL
41
+ 15: RIGHT_SHOULDER_YAW
42
+ 16: RIGHT_ELBOW_PITCH
43
+ 17: RIGHT_ELBOW_YAW
44
+ 18: RIGHT_WRIST_PITCH
45
+ 19: RIGHT_WRIST_ROLL
46
+ 20: NECK_PITCH
47
+ 21: Left hand closure state (0 = open, 1 = closed)
48
+ 22: Right hand closure state (0 = open, 1 = closed)
49
+ 23: Linear Velocity
50
+ 24: Angular Velocity
51
+ }
52
+
53
+
54
+ Previous version: v1.1
55
 
56
  - **magvit2.ckpt** - weights for [MAGVIT2](https://github.com/TencentARC/Open-MAGVIT2) image tokenizer we used. We provide the encoder (tokenizer) and decoder (de-tokenizer) weights.
57
 
 
64
  - **neck_desired** `(N, 1)`: Desired neck pitch.
65
  - **l_hand_closure** `(N, 1)`: Left hand closure state (0 = open, 1 = closed).
66
  - **r_hand_closure** `(N, 1)`: Right hand closure state (0 = open, 1 = closed).
67
+ #### Index-to-Joint Mapping (OLD)
68
  ```
69
  {
70
  0: HIP_YAW