Recurrent-Attention-CNN Data augmentation?

Data augmentation?

Open JohnnieXDU opened this issue 6 years ago • 1 comments

Hello, can you tell us some details about data augmentation?

thanks.

Mar 18 '18 15:03 JohnnieXDU

@JohnnieXDU For example, given an 300500 image, first resize it to 448 746 (keep ratio), then random crop (in training) 448*448 patches to be the input of the classification CNN, with the "mirror" parameter being "true". By the way, I think a large "dropout ratio" is important for a promising accuracy.

Mar 25 '18 02:03 Heliang-Zheng

Recurrent-Attention-CNN Recurrent-Attention-CNN copied to clipboard

Data augmentation?

Recurrent-Attention-CNN
Recurrent-Attention-CNN copied to clipboard