Jie Li
Jie Li
I find the repo [coc-jedi](https://github.com/neoclide/coc-jedi), does it conflict with the integration of Microsoft/vscode-python? BTW, what's the difference between the sources (vim plugins?) and extensions (node package?). I'm new to coc.nvim...
I run into similary result as you. Any solution have you found? Thanks.
Hey, I wonder is there any progress on it? I believe this feature does make sense.
Maybe you should set `pgd_alpha` to 1 according to the log in this repo.
Hi, thank you for your interest. The NDCG implemented in our code is a simplified version while we treat the top half of the rank as positive pairs (label 1)...
Thanks for your interest. Please kindly refer to https://github.com/theFool32/PPBA/blob/9db2e7db3c86944d155a23e8b1494e530ca6c42c/gcv.py#L88 which is an example of performing the linf norm attack with 16. You should also remove the code for l2 norm...
Actually, no. Multiplying with the projection matrix does not change the l2 norm of `uz`. So it is the same to restrict on `uz` or the final noise. But it...
$$\|\delta\|_2 = \|A^Tz\|_2 = tr((A^T z)^TA^T z) = tr(z^TAA^T z) = tr(z^Tz) = \|z\|_2 $$
Hi, Kitzzaaa. Let us assume the probabilities of three situations should be: 0.2:0.3:0.5. We can sample a value r from [0,1]. If r is in [0, 0.2], it should be...