[GitHub] MennoK opened a new issue #8600: How many new data points do data augmentation with mx.io.ImageRecordIter produce?

[GitBox <gi...@apache.org>]

MennoK opened a new issue #8600: How many new data points do data augmentation with mx.io.ImageRecordIter produce?
URL: https://github.com/apache/incubator-mxnet/issues/8600
 
 
   I think I have an understanding of the concept of data augmentation (basically you increase the dataset by transforming the given data images). I understand you can call data augmentation with mx.io.ImageRecordIter by specifying certain parameters, e.g. cropping or rotation.
   
   However, how many 'new' data images do ImageRecordIter generate and where do you specify it? Does it do for example just one random rotation for the training given image, or does it multiple rotations of the same image, for example 10 rotations between +20 and -20 degrees and use the 10 'new' images to train a model?
   
   
   
   
   

----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
users@infra.apache.org


With regards,
Apache Git Services