Make the default for workmen a theano flag.

theano/sandbox/cuda/dnn.py

 import os
 
 import theano
-from theano import Apply, gof, tensor
+from theano import Apply, gof, tensor, config
 from theano.scalar import as_scalar
 from theano.gradient import DisconnectedType
 from theano.gof import Optimizer, local_optimizer, COp
 from theano.gof.type import CDataType, Generic
 from theano.compat import PY3
 from theano.compile.ops import shape_i
+from theano.configparser import AddConfigVar, EnumStr
 from theano.tensor.nnet import SoftmaxGrad
 from theano.tensor.basic import ShapeError
 from theano.sandbox.cuda.type import CudaNdarrayType
@@ -328,6 +329,11 @@ class GpuDnnConvDesc(GpuOp):
         return (2, version())
 
 
+AddConfigVar('dnn.conv.workmem',
+             "Default value for the workmem attribute of cudnn convolutions.",
+             EnumStr('small', 'none', 'large'),
+             in_c_key=False)
+
 class GpuDnnConv(DnnBase, COp):
     """
     The forward convolution.
@@ -338,12 +344,14 @@ class GpuDnnConv(DnnBase, COp):
     """
     __props__ = ('workmem',)
 
-    def __init__(self, workmem='small'):
+    def __init__(self, workmem=None):
         """
         :param workmem: either 'none', 'small' or 'large'.  Default is 'small'.
         """
         COp.__init__(self, ["dnn_base.c", "dnn_conv_base.c", "dnn_fwd.c"],
                      "APPLY_SPECIFIC(conv_fwd)")
+        if workmem is None:
+            workmem = config.dnn.conv.workmem
         self.workmem = workmem
         assert self.workmem in ['none', 'small', 'large']