I'm working on a fine-tuning task with a small dataset and could use advice on handling catastrophic forgetting. My scenario (just as an example): I have a DeimV2 model that works great at detecting a ...
一些您可能无法访问的结果已被隐去。
显示无法访问的结果一些您可能无法访问的结果已被隐去。
显示无法访问的结果