search

nitishsrivastava / deepnet

Implementation of some deep learning algorithms.
BSD 3-Clause "New" or "Revised" License
893 stars 438 forks source link

recursive approach #64

Open iBM88 opened 10 years ago

iBM88 commented 10 years ago

datahandler.py line 408: if skip: return self.Get(batchsize, get_last_piece=get_last_piece)

Getting batches of data is done in recursive way which is not efficient at all. with batchsize=100 for a new larger dataset, I get error: "maximum recursion depth exceeded while calling a Python object" it could have been easily implemented with a "for loop" as we know the size of dataset. I assume parallel for loops would have been available as well.

YutingWang commented 8 years ago

I met the same "maximum recursion depth exceeded while calling a Python object" error. How to solve this? Thanks.