CapDec: SOTA Zero Shot Image Captioning Using CLIP and GPT2, EMNLP 2022 (findings)
active 2023-08-16 → 2026-05-27 (UTC)
Activity over time
Daily event counts in the loaded window
Line chart, 1016 days from 2023-08-16 to 2026-05-27. Pushes: 3 total, peak 1 in a day. Pull requests: 5 total, peak 3 in a day. Issues: 9 total, peak 1 in a day. Comments: 17 total, peak 6 in a day. Stars: 63 total, peak 2 in a day.
- Pushes
- Pull requests
- Issues
- Comments
- Stars
Stars, PRs, issues and forks are under-captured in the later part of this window. GH Archive progressively stopped capturing non-push events during 2026 — −95% or worse by the end of the window. Every series here except Pushes fades for that reason, so a decline above reflects the archive, not this repository. Pushes stay reliable throughout, so read them, and the contributor counts derived from them, as the real signal. Data health has the measurements.
Top contributors
Pushes, PRs, issues, reviews and comments — stars and forks excluded, so this is contribution rather than popularity
| Contributor | Contributions | Pushes | PRs | Comments |
|---|---|---|---|---|
| DavidHuji | 11 | 3 | 2 | 6 |
| hwan6615 | 5 | 0 | 0 | 3 |
| qq123aa456 | 5 | 0 | 0 | 3 |
| Panlizhi | 3 | 0 | 0 | 1 |
| davenu | 3 | 0 | 3 | 0 |
| LiLiParty | 2 | 0 | 0 | 1 |
| wxpqq826615304 | 2 | 0 | 0 | 2 |
| sszbuu | 1 | 0 | 0 | 0 |
| Flaick | 1 | 0 | 0 | 0 |
| daniyaniazi | 1 | 0 | 0 | 1 |
Recent activity
Latest issues, pull requests and releases
- Issue comment#15daniyaniazi2025-02-28 12:02Error(s) in loading state_dict for ClipCaptionModel
- Issue#21hwan66152025-02-25 01:46Performance without noise
- Issue comment#20hwan66152024-12-16 08:46Training args with COCO dataset
- Issue comment#20DavidHuji2024-12-14 21:20Training args with COCO dataset
- Issue comment#20hwan66152024-12-14 21:08Training args with COCO dataset
- Issue comment#20DavidHuji2024-12-14 21:02Training args with COCO dataset
- Issue comment#20hwan66152024-12-14 20:49Training args with COCO dataset
- Issue comment#20DavidHuji2024-12-14 18:26Training args with COCO dataset
- Issue#20hwan66152024-12-10 10:17Training args with COCO dataset
- Issue comment#19Panlizhi2024-03-09 15:26Why noise is added to “prefix”（visual embeding by clip） instead of token（text embeding by tokenizer）in the code？
- Issue#19Panlizhi2024-03-09 15:25Why noise is added to “prefix”（visual embeding by clip） instead of token（text embeding by tokenizer）in the code？
- Issue#19Panlizhi2024-02-22 14:14Why noise is added to “prefix”（visual embeding by clip） instead of token（text embeding by tokenizer）in the code？
- Pull request#18DavidHuji2024-01-28 10:01
- Pull request#18davenu2024-01-28 10:01
- Pull request#17davenu2024-01-14 14:20
- Pull request#16DavidHuji2024-01-14 14:17
- Pull request#16davenu2024-01-14 14:17
- Issue#15sszbuu2023-12-30 07:48Error(s) in loading state_dict for ClipCaptionModel
- Issue comment#14DavidHuji2023-12-29 17:42modality_offset
- Issue#14Flaick2023-12-29 16:06modality_offset
- Issue comment#7qq123aa4562023-12-12 09:19how to download / create single_caption_per_sample_val.json file
- Issue comment#7qq123aa4562023-12-12 09:18how to download / create single_caption_per_sample_val.json file
- Issue#13qq123aa4562023-12-12 09:11how to get train.json?
- Issue comment#13DavidHuji2023-12-12 08:55how to get train.json?
- Issue comment#7DavidHuji2023-12-12 08:53how to download / create single_caption_per_sample_val.json file
Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 63 stars here means stars gained during the window, not the repo's star count.