Difference between chunks and batches?

[haroonchoudery]

Hello - I was wondering what the difference between the two hyperparameters (chunks and batches)? I've been trying to wrap my head around this but can't seem to grasp it.

[wadhwasahil]

There are same. Batch/chunk is nothing but a subset of your main dataset. TF/Theano all work in batches. SGD and parameters updates happen in batches. Batch training is much faster and produces converges to almost same minima.

[haroonchoudery]

Thanks @wadhwasahil! But if they are both the same then why are they set as separate hyperparameters?