This is an official implementation for "Swin Transformer: Hierarchical Vision Transformer using Shifted Windows".
active 2024-05-03 → 2026-08-04 (UTC)
Activity over time
Daily event counts in the loaded window
Line chart, 824 days from 2024-05-03 to 2026-08-04. Pushes: 0 total, peak 0 in a day. Pull requests: 3 total, peak 1 in a day. Issues: 17 total, peak 3 in a day. Comments: 15 total, peak 2 in a day. Stars: 1,506 total, peak 12 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 |
|---|---|---|---|---|
| jasonrichdarmawan | 4 | 0 | 0 | 1 |
| nextday123 | 2 | 0 | 0 | 1 |
| mixed-fish | 1 | 0 | 0 | 0 |
| mramezani64 | 1 | 0 | 0 | 1 |
| wanghaoyucn | 1 | 0 | 0 | 0 |
| smilingfish000 | 1 | 0 | 0 | 0 |
| Athon-cyber | 1 | 0 | 0 | 0 |
| shijunticc | 1 | 0 | 0 | 1 |
| GamblerZSY | 1 | 0 | 0 | 0 |
| Matagi1996 | 1 | 0 | 0 | 1 |
| VictorLeeICT | 1 | 0 | 0 | 0 |
| Elin-lin | 1 | 0 | 0 | 1 |
| semaphore-egg | 1 | 0 | 0 | 1 |
| woongjoonchoi | 1 | 0 | 1 | 0 |
| guotong1988 | 1 | 0 | 0 | 0 |
| shubhranshii | 1 | 0 | 1 | 0 |
| abhiagwl4262 | 1 | 0 | 0 | 1 |
| wjyshuai | 1 | 0 | 0 | 1 |
| noelEOS | 1 | 0 | 1 | 0 |
| youthong | 1 | 0 | 0 | 0 |
Recent activity
Latest issues, pull requests and releases
- Issue comment#379btryq2025-11-25 03:43KeyError: 'LOCAL_RANK'
- Pull request#386noelEOS2025-09-26 02:35
- Issue comment#380abhiagwl42622025-09-22 12:57How to replace the YOLOv11backbone with SwinTransformer?
- Issue comment#111ctxqlxs2025-08-13 00:34Inferior performance of EMA model
- Issue comment#360jasonrichdarmawan2025-07-01 08:11What is `patches_resolution` in class `PatchEmbed`?
- Issue comment#382nextday1232025-03-20 05:28How do I use swin-transformer to predict the category of images in my own folder and show size mismatch when I load the model
- Issue#382nextday1232025-03-20 04:45How do I use swin-transformer to predict the category of images in my own folder and show size mismatch when I load the model
- Pull request#381shubhranshii2025-03-08 14:22
- Issue#379Athon-cyber2025-02-28 07:50KeyError: 'LOCAL_RANK'
- Pull request#376woongjoonchoi2025-01-15 07:44
- Issue#375GamblerZSY2024-12-27 02:20Structure of SwinT.
- Issue comment#256dongdongtong2024-12-25 10:58Equivalence between Patch Merging and Conv.
- Issue#374Tismer2024-12-25 09:52How to input custom image size data, such as 64×64, when training my own dataset using the Swin-TransformerV2?
- Issue#372sIHURs2024-12-01 12:53Mismatch Between Pre-trained Weights and Model Structure in SwinV2-Tiny Encoder: relative_coords_table/relative_position_index/attn_mask
- Issue#370dql04262024-10-08 06:34document missing
- Issue comment#335fosteman2024-10-01 11:49About using swinv2 for transfer training at high resolution
- Issue#369wanghaoyucn2024-09-26 06:39no weight decay setting error in SimMIM pretraining
- Issue#368youthong2024-09-26 02:58Accuracy Error Accumulation
- Issue comment#360hewg20082024-09-25 01:35What is `patches_resolution` in class `PatchEmbed`?
- Issue#367smilingfish0002024-09-20 01:45relative_coords
- Issue comment#342GuoST92024-09-15 11:48Swin Transformer training and evaluation script: error: the following arguments are required: --local_rank
- Issue comment#342wjyshuai2024-09-15 08:55Swin Transformer training and evaluation script: error: the following arguments are required: --local_rank
- Issue#366ignitemylife2024-08-15 08:28DropPath: why scale_by_keep=True? eg: batch_id which not masked will multipy with 1./keep_prob, it may not make sense?
- Issue#363AdharshC77772024-07-18 02:08Interpolation of positional embeddings when fine-tuning (320x320) when the model is pre-trained on 512x512.
- Issue#360jasonrichdarmawan2024-07-03 21:36What is `patches_resolution` in class `PatchEmbed`?
Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 1,506 stars here means stars gained during the window, not the repo's star count.