First of all, thank you very much for open sourcing your code, it has been very helpful and I will definitely make sure to cite it in my work.
I was wondering if it would be possible for you to release the 1000x evaluations obtained during the search process for CIFAR10? I am particularly curious about the performance of NAOnet during the search process to have an idea about the gap between performance during search vs. final training performance.
I didn't realize it but you already provided this info. There are 2 Google Drive files but they are both named data.tgz and I thought they were 2 links to the same file.
Hi,
First of all, thank you very much for open sourcing your code, it has been very helpful and I will definitely make sure to cite it in my work. I was wondering if it would be possible for you to release the 1000x evaluations obtained during the search process for CIFAR10? I am particularly curious about the performance of NAOnet during the search process to have an idea about the gap between performance during search vs. final training performance.
Thanks, Felipe