Update README.rst

Author: tariqdaouda

README.rst

@@ -16,7 +16,7 @@ Networks are graphs of connected layers and that allows for the craziest deepest
 you can think of, as well as for a super light and clean interface.
 
 There's no need for an MLP or a Perceptron or an Auto-Encoder class,
-because if you know what these things are, you can turn one into the other in a 2 seconds.
+because if you know what these things are, you can turn one into the other in 2 seconds.
 
 So in short:
 
@@ -121,6 +121,9 @@ Training, Testing and Propagating:
 	ae.init()
 	o.W = h.W.T
 
+Another way is to use the Autoencode layer as output::
+
+	o = ML.Autoencode(i, activation = MA.tanh, costObject = cost, learningScenario = ls)
 
 Can it run on GPU?
 ==================