图像训练 个人电脑可用VILA - a multi-image visual language model with training, inference and evaluation recipe, deployable from cloud to edge (Jetson Orin and laptops)
active 2024-02-23 → 2024-07-03 (UTC)
Activity over time
Daily event counts in the loaded window
Line chart, 132 days from 2024-02-23 to 2024-07-03. Pushes: 36 total, peak 12 in a day. Pull requests: 20 total, peak 6 in a day. Issues: 102 total, peak 6 in a day. Comments: 157 total, peak 20 in a day. Stars: 825 total, peak 53 in a day.
- Pushes
- Pull requests
- Issues
- Comments
- Stars
Top contributors
Pushes, PRs, issues, reviews and comments — stars and forks excluded, so this is contribution rather than popularity
| Contributor | Contributions | Pushes | PRs | Comments |
|---|---|---|---|---|
| Lyken17 | 43 | 5 | 0 | 27 |
| Efficient-Large-Language-Model | 40 | 10 | 6 | 18 |
| yaolug | 33 | 11 | 1 | 16 |
| hzhang57 | 14 | 0 | 0 | 7 |
| Li-Qingyun | 10 | 0 | 0 | 9 |
| kentang-mit | 7 | 4 | 1 | 2 |
| hkunzhe | 7 | 0 | 0 | 7 |
| meenchen | 7 | 2 | 2 | 2 |
| oroojlooy | 7 | 0 | 0 | 3 |
| SeanCraven314 | 6 | 0 | 3 | 3 |
| ys-2020 | 6 | 0 | 0 | 6 |
| lucasjinreal | 6 | 0 | 0 | 5 |
| HiHiAllen | 5 | 0 | 0 | 3 |
| hubenjm | 5 | 0 | 0 | 1 |
| joebradly | 4 | 0 | 0 | 3 |
| DwanZhang-AI | 4 | 0 | 0 | 2 |
| bfshi | 4 | 0 | 0 | 4 |
| rahulthakur319 | 4 | 0 | 0 | 2 |
| Lopa07 | 4 | 0 | 0 | 2 |
| Griffintaur | 3 | 0 | 0 | 3 |
Recent activity
Latest issues, pull requests and releases
- Issue#45rahulthakur3192024-07-03 10:25Add support for GPUs with compute capability lower than 8.0 for awq/kernels installation
- Issue comment#53Griffintaur2024-07-03 07:14working with VLLM
- Issue comment#69Griffintaur2024-07-02 23:45About VILADistributedSampler and gradient_accumulation_steps
- Issue comment#22Lyken172024-07-02 21:23More data leading to lower indicators?
- Issue comment#22Griffintaur2024-07-01 07:43More data leading to lower indicators?
- Issue#79oroojlooy2024-06-28 02:29Multi image inference quality
- Issue comment#78oroojlooy2024-06-27 19:56The inference video reports an error： ValueError: Unable to create tensor, you should probably activate padding with 'padding=True' to have batched tensors with the same length.
- Issue comment#78oroojlooy2024-06-27 19:51The inference video reports an error： ValueError: Unable to create tensor, you should probably activate padding with 'padding=True' to have batched tensors with the same length.
- Issue comment#68UelisonSantos2024-06-27 13:33Access to pretrained model weights
- Issue#78changqinyao2024-06-27 07:20The inference video reports an error： ValueError: Unable to create tensor, you should probably activate padding with 'padding=True' to have batched tensors with the same length.
- Issue comment#76yaolug2024-06-25 05:20What is the --conv-mode of VILA1.5-13b?
- Issue comment#76Lyken172024-06-25 02:15What is the --conv-mode of VILA1.5-13b?
- Issue comment#77Lyken172024-06-25 02:14Question about the output
- Issue comment#77DwanZhang-AI2024-06-24 00:35Question about the output
- Issue#77DwanZhang-AI2024-06-23 23:54Question about the output
- Issue#76DwanZhang-AI2024-06-23 23:25What is the --conv-mode of VILA1.5-13b?
- Issue comment#13DwanZhang-AI2024-06-23 22:31FlashAttention Bug
- Pull request#75poorfrombabylon2024-06-21 20:56
- Pull request#75poorfrombabylon2024-06-21 20:55
- Issue#74oroojlooy2024-06-21 18:54OpenVLM leaderboard
- Issue#68Lyken172024-06-21 02:58Access to pretrained model weights
- Issue comment#28Lyken172024-06-21 02:25Model checkpoints before supervised fine-tuning
- Issue#28Lyken172024-06-21 02:25Model checkpoints before supervised fine-tuning
- Issue comment#45Lyken172024-06-21 02:24Add support for GPUs with compute capability lower than 8.0 for awq/kernels installation
- Issue#65Lyken172024-06-21 02:23Does S2 able to unfreeze vit to train?
Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 825 stars here means stars gained during the window, not the repo's star count.