0

I am new in Theano and Deep Learning, I am running my experiments in Theano but I would like to reduce the time I spend per epoch by doing data augmentation directly using the GPU.

Unfortunately I can not use PyCuda, so I would like to know if is possible to do basic Data Augmentation using Theano. For example Translation or Rotation in images, meanwhile I am using scipy functions in CPU using Numpy but it is quite slow.

DeepLearner
  • 1
  • 1
  • I would take a look at [this repo](https://github.com/benanne/kaggle-ndsb). It is code from a Kaggle competition written by the creator of the [Lasagne](https://github.com/Lasagne/Lasagne) project. In his solution, he does all the data augmentation using his CPU and puts each augmented batch in a queue, while the GPU grabs batches from the queue and trains. – o-90 Aug 18 '16 at 16:07

1 Answers1

0

If the data augmentation is part of your computation graph, and can be executed on GPU, it will naturally be executed on the GPU. So the question narrows down to "is it possible to do common data augmentation tasks using Theano tensor operations on the GPU".

If the transformations you want to apply are just translations, you can just use theano.tensor.roll followed by some masking. If you want the rotations as well, take a look at this implementation of spatial transformer network. In particular take a look at the _transform function, it takes as an input a matrix theta that has a 2x3 transformation (left 2x2 is rotation, and right 1x2 is translation) one per sample and the actual samples, and applies the rotation and translation to those samples. I didn't confirm that what it does is optimized for the GPU (i.e. it could be that the bottleneck of that function is executed on the CPU, which will make it not appropriate for your use case), but it's a good starting point.

Ishamael
  • 12,583
  • 4
  • 34
  • 52