This is the Linux app named ToMe (Token Merging) whose latest release can be downloaded as ToMesourcecode.tar.gz. It can be run online in the free hosting provider OnWorks for workstations.
Download and run online this app named ToMe (Token Merging) with OnWorks for free.
请按照以下说明运行此应用程序：
- 1. 在您的 PC 中下载此应用程序。
- 2. 在我们的文件管理器 https://www.onworks.net/myfiles.php?username=XXXXX 中输入您想要的用户名。
- 3. 在这样的文件管理器中上传这个应用程序。
- 4. 从此网站启动OnWorks Linux online 或Windows online emulator 或MACOS online emulator。
- 5. 从您刚刚启动的 OnWorks Linux 操作系统，使用您想要的用户名转到我们的文件管理器 https://www.onworks.net/myfiles.php?username=XXXXX。
- 6. 下载应用程序，安装并运行。
SCREENSHOTS
Ad
ToMe (Token Merging)
商品描述
ToMe (Token Merging) is a PyTorch-based optimization framework designed to significantly accelerate Vision Transformer (ViT) architectures without retraining. Developed by researchers at Facebook (Meta AI), ToMe introduces an efficient technique that merges similar tokens within transformer layers, reducing redundant computation while preserving model accuracy. This approach differs from token pruning, which removes background tokens entirely; instead, ToMe merges tokens based on feature similarity, allowing it to compress both foreground and background information efficiently. ToMe integrates seamlessly into existing transformer models such as DeiT, MAE, SWAG, and timm ViTs, offering 2–3x speedups during inference and substantial efficiency gains during training. The method can be applied dynamically at inference time or incorporated into training for improved performance.
功能
- Supported for both ImageNet evaluation and research extensions
- Open source PyTorch patching tools for quick integration
- Offers pretrained checkpoints for DeiT, ViT-B/L/H, and MAE models
- Can be applied without retraining or integrated during training for better results
- Compatible with timm, SWAG, and MAE ViT implementations
- Provides 2–3× inference speedup with minimal accuracy loss
- Token Merging algorithm for accelerating Vision Transformers
程式语言
Python
分类
This is an application that can also be fetched from https://sourceforge.net/projects/tome-token-merging.mirror/. It has been hosted in OnWorks in order to be run online in an easiest way from one of our free Operative Systems.
