search

HanxunH / Unlearnable-Examples

[ICLR2021] Unlearnable Examples: Making Personal Data Unexploitable
https://hanxunh.github.io/Unlearnable-Examples/
MIT License
150 stars 17 forks source link

Can you share your experience with fast-autoaugment #18

Closed saigontrade88 closed 11 months ago

saigontrade88 commented 11 months ago

Hello,

Thanks you for sharing your good work. You showed in the Appendix, Table 3 that your class-wise and sample-wise noise is unstable when the model is trained with Fast AutoAugment. In the following question under the official implementation of the fast-autoaugment, there is some complaint about the incompatibility between RayX and Slurm. Can you help share how you run the fast-auto augment for your unlearnable example? Many thanks.

https://github.com/kakaobrain/fast-autoaugment/issues/27

HanxunH commented 11 months ago

Hi,

It seems the issue is due to an augmentation search rather than applying the augmentation in training. I'm not aware of any fix.

For running fast auto augment on unlearnable examples, have a look at the following:

https://github.com/HanxunH/Unlearnable-Examples/blob/28e1e417259ce5baf0d3e68d3408d9f8ecc3024f/dataset.py#L17

https://github.com/HanxunH/Unlearnable-Examples/blob/28e1e417259ce5baf0d3e68d3408d9f8ecc3024f/dataset.py#L117

saigontrade88 commented 11 months ago

Many thanks for the prompt reply and useful information.