ViT-Slim
ViT-Slim copied to clipboard
Official code for our CVPR'22 paper “Vision Transformer Slimming: Multi-Dimension Searching in Continuous Optimization Space”
Dear repository maintainer, I am very interested in the ShareGPT-GLoRA-7B model that you have shared on GitHub. I would like to learn more about the specific training and evaluation code...
As shown in the figure, there are errors reported about the attribute "is_searched" not found. After tracing back this error, I found the problem is because in the class ModuleInjection,...
According to Equation 10, should the module D in Figure 1 be followed by a multiplication sign instead of a plus sign to indicate that Db0?
Hi, great work! Could you please provide me with the code to count the FOLPs and Latency in Table 3, so I can use it to follow your work better....
Hi, authors! Thanks for your great work! But I have a question about the hyper-parameter setting of fine-tuning on small datasets, which is not mentioned in the paper. I'd appreciate...
Hello! Thank you for the great work. I have some confusion about the patch selection process. In section 3.3 of the paper, it is stated that it is important to...