I think the data was shuffled before creating the dataset file. This remove the need to shuffle it in each job.
Where did you took the code? If it is from the deep learning tutorial, it is already shuffled and partitioned. Le sam. 11 mars 2017 16:17, Zhanhong Jiang <[email protected]> a écrit : > Hi, all: > > I have been trying to use CNN code of Deep Learning tutorial maintained by > LISA lab. I am wondering why they didn't do random shuffle for training for > each epoch. But I have figured it out. However, when I wanted to partition > the training data set into several data subsets and use data subsets to > train the model, I experienced a hard time to partition the data and still > could't solve the problem. > > Anyone has some suggestions on that or did the similar thing before? I > will really appreciate any suggestions or help! > > Thanks!!!!!! > > -- > > --- > You received this message because you are subscribed to the Google Groups > "theano-users" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > For more options, visit https://groups.google.com/d/optout. > -- --- You received this message because you are subscribed to the Google Groups "theano-users" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/d/optout.
