Kai: /* External Resources */

2016-10-03T20:22:28Z

External Resources

New page

=== Introduction ===
Deep learning is an approach based on multilayer [[Neural Networks | neural networks]] for learning a hierarchical set of features from the training data before the model optimization step.
 
 
=== Training of Video Data ===
A hybrid CPU/GPU approach has proven more efficient here than a GPU-heavy approach: At first, training was done by simultaneously feeding multiple frames from the video into the network. This requires a large (expensive) network. It was then seen that similar results can be achieved by extracting the optical flow data from the frames in advance (with a CPU-bound process) and feed the frames plus the flow data sequentially to the network ([http://cs.stanford.edu/people/karpathy/deepvideo/ Large-scale Video Classification with Convolutional Neural Networks] and the [https://code.google.com/archive/p/sports-1m-dataset/ Sports 1M data set]).
 
 
Consider converting the color space of the input frames from RGB to YUV in order to condition the network to a color perception model that is closer to human vision.
 
 
=== Software Tools ===
* [http://torch.ch Torch] by Facebook, Google, Twitter
* [[Caffe]] by UC Berkeley
* [http://deeplearning.net/software/theano Theano] with [https://github.com/lisa-lab/pylearn2 Pylearn2], by the University of Montréal
* [https://github.com/Microsoft/CNTK/wiki Computational Network Toolkit] by Microsoft
* [https://www.tensorflow.org TensorFlow] by Google
* [https://github.com/nervanasystems/neon Neon] by Nervana Systems
* [https://developer.nvidia.com/cudnn Nvidia cuDNN]
 

=== External Resources ===
* '''Site''' http://deeplearning.net
* '''Videos''' [http://videolectures.net/site/search/?q=deep+learning Deep Learning @ videolectures.net]

Deep Learning - Revision history

Kai: /* External Resources */