Atharva Phatak

@otaj If this is still open, I would like to help you develop bolts.

@ananyahjha93 Hi, I would like to compile the papers for readme.md if its okay ?

I would love if TorchData becomes standalone. It could be the go to place for PyTorch Ecosystem libraries, since it essentially is along the same concept as torch.Dataset. All in...

So what was the reason for using single score (s->h). Does BARTScore holistically measure quality of generated text ? For example can you report s->h variant of BARTScore and say...

what you can do is inside your model class define let's say you are using ArcFace. define it as follows class Model: def __init__(self , args): #Define your model here...

Adding metric learning models like ArcFace , CosFace etc ? If so I can create a pull request and implement them ?

Well the modules like [ArcFace](https://arxiv.org/abs/1801.07698) , [CosFace](https://arxiv.org/abs/1801.09414) have different logic(some calculations are different) I was thinking of implementing them as simple classes which inherit from **nn.Module** and have their own...

Hi, So the paper mentions it as loss function, but in implementation it is actually a series of steps(called as margin products) which help you obtain the logits and when...

Hi @awaelchli, I am interested in this, is this up for the taking? Also could provide some additional in terms of whats expected in the PR ?

@awaelchli Thanks for the update. I will work on it once #11617 has merged.