Merge branch 'master' into data-parallelism

examples/mnist/mnist.jl

@@ -6,7 +6,7 @@ ENV["MOCHA_USE_CUDA"] = "true"
 using Mocha
 srand(12345678)
 
-data_layer  = HDF5DataLayer(name="train-data", source="data/train.txt", batch_size=64)
+data_layer  = HDF5DataLayer(name="train-data", source="data/train.txt", batch_size=64, shuffle=true)
 conv_layer  = ConvolutionLayer(name="conv1", n_filter=20, kernel=(5,5), bottoms=[:data], tops=[:conv])
 pool_layer  = PoolingLayer(name="pool1", kernel=(2,2), stride=(2,2), bottoms=[:conv], tops=[:pool])
 conv2_layer = ConvolutionLayer(name="conv2", n_filter=50, kernel=(5,5), bottoms=[:pool], tops=[:conv2])