If I apply rotation of 5 different angles and randomly cropp 10 different images from each rotated image and than divided into training testing and validation. Will it be totally incorrect evaluation of dataset?
$\begingroup$
$\endgroup$
asked Jun 14, 2019 at 9:29
Add a comment
|
1 Answer
$\begingroup$
$\endgroup$
1
Yes, it will be incorrect.
When you augment before splitting, it is possible that one crop of the test set appears in the training set, and the other in the test/validation set. This would leak information about the test set to the model during training, resulting in invalid evaluation.
-
$\begingroup$ So what will be the best data arrangement. data augmentation on training data.. validation and testing remaining as original data. if its that so.. lets say training data will be after data augmentation will be 10000 images and validation and testing will be 500 images each? $\endgroup$ Jun 15, 2019 at 1:52