Lax3n commited on
Commit
2ebaad8
·
verified ·
1 Parent(s): 733a28d

Upload folder using huggingface_hub

Browse files
Files changed (2) hide show
  1. README.md +4 -4
  2. metadata.jsonl +0 -0
README.md CHANGED
@@ -19,11 +19,11 @@ configs:
19
 
20
  # avqa-processed — YouTube source videos
21
 
22
- **3,093 videos** (185.3 GB) with 899 subtitle files, downloaded at source
23
  quality and hosted for direct download.
24
 
25
- > **Coverage (2026-11-15):** 3,093 complete YouTube source videos are published here. They come from a prepared selection of 9,513 YouTube IDs read from [`gijs/avqa-processed`](https://huggingface.co/datasets/gijs/avqa-processed) (revision `e2aa3fc443a4`); the selection stopped because all matching upstream metadata files were read. It is a bounded subset, not the complete upstream list.
26
- > Of the prepared IDs: 3,093 published, 459 unavailable at download time, 7 failed downloads, 0 not added because of the 264 GB cap, 11 in progress and 5,943 not attempted.
27
  > Every row of the `videos` configuration points to a file present in our storage when this card was generated, and the annotations are restricted to those videos.
28
 
29
 
@@ -34,7 +34,7 @@ quality and hosted for direct download.
34
  ## What's inside
35
 
36
  - **`metadata.jsonl`** — one row per video: `video_id`, `video_url` (direct HTTPS download, CDN-backed), `subtitle_url`, `youtube_url`, `size_bytes`
37
- - The 3,093 MP4s (185.3 GB) and 899 subtitle files are hosted on Hugging Face storage and served via the `video_url` column
38
 
39
  ## Quick start
40
 
 
19
 
20
  # avqa-processed — YouTube source videos
21
 
22
+ **3,555 videos** (212.5 GB) with 1,043 subtitle files, downloaded at source
23
  quality and hosted for direct download.
24
 
25
+ > **Coverage (2026-11-15):** 3,555 complete YouTube source videos are published here. They come from a prepared selection of 9,513 YouTube IDs read from [`gijs/avqa-processed`](https://huggingface.co/datasets/gijs/avqa-processed) (revision `e2aa3fc443a4`); the selection stopped because all matching upstream metadata files were read. It is a bounded subset, not the complete upstream list.
26
+ > Of the prepared IDs: 3,555 published, 530 unavailable at download time, 7 failed downloads, 0 not added because of the 264 GB cap, 14 in progress and 5,407 not attempted.
27
  > Every row of the `videos` configuration points to a file present in our storage when this card was generated, and the annotations are restricted to those videos.
28
 
29
 
 
34
  ## What's inside
35
 
36
  - **`metadata.jsonl`** — one row per video: `video_id`, `video_url` (direct HTTPS download, CDN-backed), `subtitle_url`, `youtube_url`, `size_bytes`
37
+ - The 3,555 MP4s (212.5 GB) and 1,043 subtitle files are hosted on Hugging Face storage and served via the `video_url` column
38
 
39
  ## Quick start
40
 
metadata.jsonl CHANGED
The diff for this file is too large to render. See raw diff