Merge pull request pytorch#148 from adham/patch-1

update softmax code example comments

beginner_source/nlp/deep_learning_tutorial.py

@@ -120,7 +120,7 @@
 # then dividing by the normalization constant.
 #
 
-# Softmax is also in torch.functional
+# Softmax is also in torch.nn.functional
 data = autograd.Variable(torch.randn(5))
 print(data)
 print(F.softmax(data))