Priya Nagda
Priya Nagda
Would like to give adding alpha distribution a shot.
Seems like an example already exists here - [Keras Examples - Text Extraction with BERT](https://keras.io/examples/nlp/text_extraction_with_bert/)
For the most part, it seems pretty simple, I believe the steps would be to divide the images into patches, use `knn_graph` function to get the graph representation, and after...
Could I give it a try? Seems like a good issue to start with.
Thank you @oke-aditya! A question - we are looking to implement multiclass dice loss, correct?
Ahh alright, shouldn't be too hard to add binary dice loss. In the discussion here - https://github.com/pytorch/pytorch/issues/1249#issuecomment-305088398 there were mentions of multiclass dice loss, also I think Kornia supports multiclass...
I could give it a shot if it's still on the roadmap.
> Also, I feel that using RawKernel and CIndexer can simplify the code a lot and make it much faster, but this is just a thought out of the wild....
I would like to work on this, however, I had a few questions as to how detailed it has to be? I assume it would be the title, cost, description,...
I believe implementing the first two features you mentioned wouldn't be difficult, however, I believe product variation is beyond my scope. > Shall I assign it to you? Yes, would...