Linpeng Tang

Thanks @awaelchli @ananyahjha93 ! I think the parameter on lars indeed needs to be updated in the document: https://pytorch-lightning-bolts.readthedocs.io/en/latest/self_supervised_models.html#id21  I will try some different settings, but I do have...

@ananyahjha93 I tried to set the learning rate to 0.1, but still no luck yet. Can we still use ADAM with LARS? ``` PL_EXP_VERSION=swav_stl10_gpu1_lars01 CUDA_VISIBLE_DEVICES=1,2 python -m vault_embed.swav.swav_module \ --online_ft...

I support this request as well. Can add a parameter to `@debug` about which error to trap (or trap them all), and in the debugger you can just step or...

Hi, great work about the HF model! We are looking forward to the one-shot object detection function as well, hopefully with a Gradio demo :)

Some of our customers are requiring this feature, @baskaryan @hwchase17 please let us know how to make this feature get through.

supported in 1.5