Proof reading, variable inconsistency fix

code/dA.py

@@ -195,7 +195,7 @@ def __init__(
 
     def get_corrupted_input(self, input, corruption_level):
         """This function keeps ``1-corruption_level`` entries of the inputs the
-        same and zero-out randomly selected subset of size ``coruption_level``
+        same and zero-out randomly selected subset of size ``corruption_level``
         Note : first argument of theano.rng.binomial is the shape(size) of
                random numbers that it should produce
                second argument is the number of trials

doc/lenet.txt

@@ -543,7 +543,7 @@ the task.
 
 Filter Shape
 ************
-Common filter shapes found in the litterature vary greatly, usually based on
+Common filter shapes found in the literature vary greatly, usually based on
 the dataset. Best results on MNIST-sized images (28x28) are usually in the 5x5
 range on the first layer, while natural image datasets (often with hundreds of pixels in each
 dimension) tend to use larger first-layer filters of shape 12x12 or 15x15.

doc/mlp.txt

@@ -90,8 +90,8 @@ The set of parameters to learn is the set :math:`\theta =
 \{W^{(2)},b^{(2)},W^{(1)},b^{(1)}\}`.  Obtaining the gradients
 :math:`\partial{\ell}/\partial{\theta}` can be achieved through the
 **backpropagation algorithm** (a special case of the chain-rule of derivation).
-Thankfully, since Theano performs automatic differentation, we will not need to
-cover this in the tutorial !
+Thankfully, since Theano performs automatic differentiation, we will not need to
+cover this in the tutorial!
 
 
 Going from logistic regression to MLP