search

keyu-tian / SparK

[ICLR'23 SpotlightšŸ”„] 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

Are there any plans to make a port to tensorflow and Keras? #68

Closed noahcjones closed 4 months ago

noahcjones commented 8 months ago

Hi, My existing codebase is using Tensorflow/Keras and I want to know if the authors knew of or planned to release an official version for pretraining and finetuning that works with Keras?

keyu-tian commented 8 months ago

Hi @noahcjones, we have less mastery of other frameworks and it would be very welcome if you could migrate our pytorch implementation to them ā¤. We can involve in code reviews if you want.