dnn_conv() documentation fixup after comments.

doc/library/tensor/nnet/conv.txt

@@ -96,13 +96,15 @@ TODO: Give examples on how to use these things! They are pretty complicated.
           f = theano.function(..., mode=mode)
 
     - :func:`dnn_conv <theano.sandbox.cuda.dnn.dnn_conv>` GPU-only
-      convolution using NVIDIA's cuDNN library.  To enable it (and
-      other cudnn-accelerated ops), set
+      convolution using NVIDIA's cuDNN library.  To have conv2d()
+      automatically converted set
       ``THEANO_FLAGS=optimizer_including=cudnn`` in your environment.
-      This requires that you have cuDNN installed and available.  It
-      also requires a GPU with compute capability 3.0 or more.
+      This will also replace other operations by their a
+      cuDNN-accelerated equivalent.  This requires that you have cuDNN
+      installed and available.  It requires a GPU with compute
+      capability 3.0 or more.
 
-      Since it has a gradient defined it can be used manually.
+      Since it has a gradient defined it can also be used manually.
 
     - :func:`conv3D <theano.tensor.nnet.Conv3D.conv3D>`
       3D Convolution applying multi-channel 3D filters to batches of

theano/sandbox/cuda/dnn.py

@@ -382,6 +382,8 @@ def dnn_conv(img, kerns, border_mode='valid', subsample=(1, 1),
     """
     GPU convolution using cuDNN from NVIDIA.
 
+    The memory layout to use is 'bc01'.
+
     :param img: images to do the convolution over
     :param kerns: convolution filters
     :param border_mode: one of 'valid', 'full' (default: 'valid')