futurefantasy commited on
Commit
5d495c4
·
verified ·
1 Parent(s): 2db6e5a

Update README for full-data release layout

Browse files
Files changed (1) hide show
  1. README.md +8 -8
README.md CHANGED
@@ -28,12 +28,12 @@ scripts/
28
  extract_vlac2_release_frames.py
29
 
30
  data/
31
- vlac2_release_data_part01_arx_group_a.tar.zst
32
- vlac2_release_data_part02_arx_group_b.tar.zst
33
- vlac2_release_data_part03_arx_group_c.tar.zst
34
- vlac2_release_data_part04_droid_group_a.tar.zst
35
- vlac2_release_data_part05_droid_group_b.tar.zst
36
- vlac2_release_data_part06_droid_group_c.tar.zst
37
  vlac2_release_data_part07_other_sources.tar.zst
38
  ```
39
 
@@ -45,7 +45,7 @@ data/
45
  bash scripts/unpack_data.sh /path/to/data
46
  ```
47
 
48
- Some archives are distributed as multi-part files to satisfy Hub file-size limits. `unpack_data.sh` detects these parts automatically and streams them into a single extraction pass.
49
 
50
  ### 2. Extract frames for a selected benchmark JSON
51
 
@@ -73,4 +73,4 @@ Replace `__VLAC2_FRAMES_ROOT__` with the absolute path to the extracted-frame di
73
  * This release contains the complete raw-data archive set used by the VLAC-Cut benchmark release workflow.
74
  * The `benchmark_style_all/all` split is larger than the benchmark evaluation subsets and is rebuilt from the full annotation PKL source tree.
75
  * The benchmark JSON files are pre-generated and do not need to be rebuilt for standard use.
76
- * Large raw-data archives may appear as split `.part-*` files in `data/`; these are unpacked transparently by the provided script.
 
28
  extract_vlac2_release_frames.py
29
 
30
  data/
31
+ vlac2_release_data_part01_arx_group_a.tar.zst.part-*
32
+ vlac2_release_data_part02_arx_group_b.tar.zst.part-*
33
+ vlac2_release_data_part03_arx_group_c.tar.zst.part-*
34
+ vlac2_release_data_part04_droid_group_a.tar.zst.part-*
35
+ vlac2_release_data_part05_droid_group_b.tar.zst.part-*
36
+ vlac2_release_data_part06_droid_group_c.tar.zst.part-*
37
  vlac2_release_data_part07_other_sources.tar.zst
38
  ```
39
 
 
45
  bash scripts/unpack_data.sh /path/to/data
46
  ```
47
 
48
+ Several large archives are distributed as multi-part files to satisfy Hub file-size limits. `unpack_data.sh` detects these parts automatically and streams them into a single extraction pass.
49
 
50
  ### 2. Extract frames for a selected benchmark JSON
51
 
 
73
  * This release contains the complete raw-data archive set used by the VLAC-Cut benchmark release workflow.
74
  * The `benchmark_style_all/all` split is larger than the benchmark evaluation subsets and is rebuilt from the full annotation PKL source tree.
75
  * The benchmark JSON files are pre-generated and do not need to be rebuilt for standard use.
76
+ * The large raw-data archives are distributed as split `.part-*` files in `data/`; these are unpacked transparently by the provided script.