keyu-tian / SparK

[ICLR'23 Spotlight🔥] The first successful BERT/MAE-style pretraining on any convolutional network; Pytorch impl. of "Designing BERT for Convolutional Networks: Sparse and Hierarchical Masked Modeling"
https://arxiv.org/abs/2301.03580
MIT License
1.41k stars 82 forks source link

recommend a great library designed for sparse tensors #76

Closed xubin04 closed 4 months ago

xubin04 commented 6 months ago

How about refactoring some code using torch sparse++? I saw that their paper referenced your work, and it seems like a great library for handling sparse tensors

keyu-tian commented 6 months ago

@xubin04 That's a good idea but currently I don't have time for this. If you're interested you can give it a try and considering submitting a PR to our codebase :D.