Intel(R) Math Kernel Library for Deep Neural Networks (Intel(R) MKL-DNN)  1.0.4
Performance library for Deep Learning
Modules
Primitives

Modules

 Primitive descriptors
 
 Convolution
 Computes a forward propagation, backward propagation, or weight update for convolution operation with bias on a batch of multi-dimensional tensors.
 
 Deconvolution
 A primitive to compute deconvolution using different algorithms.
 
 LRN
 A primitive to perform local response normalization (LRN) across or within channels.
 
 Pooling
 A primitive to perform max or average pooling.
 
 Eltwise
 A primitive to compute element-wise operations such as rectified linear unit (ReLU).
 
 Softmax
 A primitive to perform softmax.
 
 Batch normalization
 A primitive to perform batch normalization.
 
 Inner Product
 A primitive to compute an inner product.
 
 RNN
 A primitive to compute common recurrent layer.
 
 Shuffle
 A primitive to shuffle data along the axis.
 

Detailed Description