[Doc] Fix GraphConv math notation (#1430)

python/dgl/nn/mxnet/conv/graphconv.py

@@ -103,7 +103,7 @@ class GraphConv(gluon.Block):
           dimensions, :math:`N` is the number of nodes.
         * Output shape: :math:`(N, *, \text{out_feats})` where all but the last dimension are
           the same shape as the input.
-        * Weight shape: "math:`(\text{in_feats}, \text{out_feats})`.
+        * Weight shape: :math:`(\text{in_feats}, \text{out_feats})`.
 
         Parameters
         ----------

python/dgl/nn/pytorch/conv/graphconv.py

@@ -109,7 +109,7 @@ class GraphConv(nn.Module):
           dimensions, :math:`N` is the number of nodes.
         * Output shape: :math:`(N, *, \text{out_feats})` where all but the last dimension are
           the same shape as the input.
-        * Weight shape: "math:`(\text{in_feats}, \text{out_feats})`.
+        * Weight shape: :math:`(\text{in_feats}, \text{out_feats})`.
 
         Parameters
         ----------

python/dgl/nn/tensorflow/conv/graphconv.py

@@ -106,7 +106,7 @@ class GraphConv(layers.Layer):
           dimensions, :math:`N` is the number of nodes.
         * Output shape: :math:`(N, *, \text{out_feats})` where all but the last dimension are
           the same shape as the input.
-        * Weight shape: "math:`(\text{in_feats}, \text{out_feats})`.
+        * Weight shape: :math:`(\text{in_feats}, \text{out_feats})`.
 
         Parameters
         ----------