question about learning rate and rotate strategy
CoinCheung opened this issue · 2 comments
Hi,
Thanks for providing this awesome work to us!!!
After reading the code, I am not sure whether I have fully understood it, so I feel I better open an issue to ask:
-
the original cifar-10 is trained with learning-rate of 0.1 when the batchsize is 128. With the rotnet method, the batchsize is amplified to 512 (128x4) but the learning rate is still kept 0.1, is that right ?
-
I see in the paper that the strategy of "simultaneously rotate the input image by 4 degrees and enlarge the batchsize 4 times" outperforms "randomly choose one degree to rotate and kept the batchsize not changed". Will the "randomly choose method" bring a significantly bad result, or it is only slightly outperformed by the proposed "4 rotates method" ?
I would be very happy to have your rely. Would you please show me your ideas on these details?
Hi CoinCheung,
- Yes.
- No, it is not significantly worse. You might need to increase however the number of training epochs.
Thanks for support !!