fix type and doc formating.

Author: nouiz

code/DBN.py

@@ -41,7 +41,7 @@ def __init__(self, numpy_rng, theano_rng = None, n_ins = 784,
         :param n_ins: dimension of the input to the DBN
 
         :type n_layers_sizes: list of ints
-        :param n_layers_sizes: intermidiate layers size, must contain 
+        :param n_layers_sizes: intermediate layers size, must contain 
                                at least one value
 
         :type n_outs: int
@@ -63,7 +63,7 @@ def __init__(self, numpy_rng, theano_rng = None, n_ins = 784,
         self.y  = T.ivector('y') # the labels are presented as 1D vector of 
                                  # [int] labels
 
-        # The DBN is an MLP, for which all weights of intermidiate layers are shared with a
+        # The DBN is an MLP, for which all weights of intermediate layers are shared with a
         # different RBM.  We will first construct the DBN as a deep multilayer perceptron, and
         # when constructing each sigmoidal layer we also construct an RBM that shares weights
         # with that layer. During pretraining we will train these RBMs (which will lead

code/SdA.py

@@ -70,7 +70,7 @@ def __init__(self, numpy_rng, theano_rng = None, n_ins = 784,
         :param n_ins: dimension of the input to the sdA
 
         :type n_layers_sizes: list of ints
-        :param n_layers_sizes: intermidiate layers size, must contain 
+        :param n_layers_sizes: intermediate layers size, must contain 
                                at least one value
 
         :type n_outs: int
@@ -95,7 +95,7 @@ def __init__(self, numpy_rng, theano_rng = None, n_ins = 784,
         self.y  = T.ivector('y') # the labels are presented as 1D vector of 
                                  # [int] labels
 
-        # The SdA is an MLP, for which all weights of intermidiate layers
+        # The SdA is an MLP, for which all weights of intermediate layers
         # are shared with a different denoising autoencoders 
         # We will first construct the SdA as a deep multilayer perceptron,
         # and when constructing each sigmoidal layer we also construct a 

code/mlp.py

@@ -3,7 +3,7 @@
 
  A multilayer perceptron is a logistic regressor where
 instead of feeding the input to the logistic regression you insert a
-intermidiate layer, called the hidden layer, that has a nonlinear 
+intermediate layer, called the hidden layer, that has a nonlinear 
 activation function (usually tanh or sigmoid) . One can use many such 
 hidden layers making the architecture deep. The tutorial will also tackle 
 the problem of MNIST digit classification.
@@ -101,7 +101,7 @@ class MLP(object):
 
     A multilayer perceptron is a feedforward artificial neural network model 
     that has one layer or more of hidden units and nonlinear activations. 
-    Intermidiate layers usually have as activation function thanh or the 
+    Intermediate layers usually have as activation function thanh or the 
     sigmoid function (defined here by a ``SigmoidalLayer`` class)  while the 
     top layer is a softamx layer (defined here by a ``LogisticRegression`` 
     class). 

doc/DBN.txt

@@ -162,7 +162,7 @@ classification.
             :param n_ins: dimension of the input to the DBN
 
             :type n_layers_sizes: list of ints
-            :param n_layers_sizes: intermidiate layers size, must contain 
+            :param n_layers_sizes: intermediate layers size, must contain 
                                at least one value
 
             :type n_outs: int

doc/SdA.txt

@@ -94,7 +94,7 @@ representations of intermediate layers of the MLP.
         :param n_ins: dimension of the input to the sdA
 
         :type n_layers_sizes: list of ints
-        :param n_layers_sizes: intermidiate layers size, must contain 
+        :param n_layers_sizes: intermediate layers size, must contain 
                                at least one value
 
         :type n_outs: int

doc/mlp.txt

@@ -138,7 +138,7 @@ layer on top.
 
 The initial values for the weights of a hidden layer :math:`i` should be uniformly
 sampled from a symmetric interval that depends on the activation function. For
-:math:`tanh` activation function results obtained in [Xavier10] show that the
+:math:`tanh` activation function results obtained in [Xavier10]_ show that the
 interval should be 
 :math:`[-\sqrt{\frac{6}{fan_{in}+fan_{out}}},\sqrt{\frac{6}{fan_{in}+fan_{out}}}]`, where 
 :math:`fan_{in}` is the number of units in the :math:`(i-1)`-th layer,
@@ -154,11 +154,11 @@ both upward (activations flowing from inputs to outputs) and backward
         # `W` is initialized with `W_values` which is uniformely sampled
         # from sqrt(-6./(n_in+n_hidden)) and sqrt(6./(n_in+n_hidden))
         # for tanh activation function
-        # the output of uniform if converted using asarray to dtype 
+        # the output of uniform is converted using asarray to dtype 
         # theano.config.floatX so that the code is runable on GPU
         # Note : optimal initialization of weights is dependent on the
         #        activation function used (among other things).
-        #        For example, results presented in [Xavier10] suggest that you 
+        #        For example, results presented in [Xavier10]_ suggest that you 
         #        should use 4 times larger initial weights for sigmoid 
         #        compared to tanh
         if activation == theano.tensor.tanh:
@@ -207,7 +207,7 @@ the ``MLP`` class :
 
     A multilayer perceptron is a feedforward artificial neural network model 
     that has one layer or more of hidden units and nonlinear activations. 
-    Intermidiate layers usually have as activation function thanh or the 
+    Intermediate layers usually have as activation function tanh or the 
     sigmoid function (defined here by a ``HiddenLayer`` class)  while the 
     top layer is a softamx layer (defined here by a ``LogisticRegression`` 
     class). 
@@ -412,7 +412,7 @@ Under some assumptions, a compromise between these two constraints leads to the
 initialization: :math:`uniform[-\frac{6}{\sqrt{fan_{in}+fan_{out}}},\frac{6}{\sqrt{fan_{in}+fan_{out}}}]`
 for tanh and :math:`uniform[-4*\frac{6}{\sqrt{fan_{in}+fan_{out}}},4*\frac{6}{\sqrt{fan_{in}+fan_{out}}}]`
 for sigmoid. Where :math:`fan_{in}` is the number of inputs and :math:`fan_{out}` the number of hidden units.
-For mathematical considerations please refer to [Xavier10].
+For mathematical considerations please refer to [Xavier10]_.
 
 Learning rate
 --------------

doc/rbm.txt

@@ -574,7 +574,7 @@ op provided by Theano, therefore we urge the reader to look it up by following t
 
 Once we have the generated the chain we take the sample at the end of the 
 chain to get the free energy of the negative phase. Note that the
-``chain_end`` is a symbolical Theano variable express in terms of the model
+``chain_end`` is a symbolical Theano variable expressed in terms of the model
 parameters, and if we would apply ``T.grad`` naively, the function will 
 try to go through the Gibbs chain to get the gradients. This is not what we 
 want (it will mess up our gradients) and therefire we need to indicate to
@@ -585,7 +585,7 @@ want (it will mess up our gradients) and therefire we need to indicate to
 
 
         # determine gradients on RBM parameters
-        # not that we only need the sample at the end of the chain
+        # note that we only need the sample at the end of the chain
         chain_end = nv_samples[-1]
 
         cost = T.mean(self.free_energy(self.input)) - T.mean(self.free_energy(chain_end))