Can you provide a detailed description about the paras after "--load"

[Golevka2001]

I'm confused about the parameters, like Text2Image_cuda:0.

And I hope a more detailed description can be provided. Then users can enter proper paras to match their own GPUs with different memory.

[Wang-Xiaodong1899]

Hi @Golevka2001 , we just updated the details in README.md. The parameter --load indicates which Visual Foundation Model to use and where it will be loaded to. The model and device are sperated by underline '_', the different models are seperated by comma ','. The available Visual Foundation Models can be found in the table. For example, if you want to load ImageCaptioning to cpu and Text2Image to cuda:0, use: --load ImageCaptioning_cpu,Text2Image_cuda:0

Thanks~