--- a/dataset.py	Sat Jul 19 10:11:22 2008 -0400
+++ b/dataset.py	Sat Jul 19 14:12:41 2008 -0400
@@ -220,7 +220,8 @@
         Sub-classes which implement finite-length datasets should redefine this method.
         Some methods only make sense for finite-length datasets.
         """
-        return None
+        from sys import maxint
+        return maxint
 
 
     class MinibatchToSingleExampleIterator(object):
@@ -943,6 +944,9 @@
             del self.fieldname2dataset[fieldname]
             self.fieldname2dataset[rename_field(fieldname,self.datasets[i],i)]=i
             
+    def __len__(self):
+        return len(self.datasets[0])
+    
     def hasFields(self,*fieldnames):
         for fieldname in fieldnames:
             if not fieldname in self.fieldname2dataset:

--- a/kernel_regression.py	Sat Jul 19 10:11:22 2008 -0400
+++ b/kernel_regression.py	Sat Jul 19 14:12:41 2008 -0400
@@ -48,7 +48,7 @@
 
     Usage:
 
-       kernel_regressor=KernelRegression(L2_regularizer=0.1,kernel=GaussianKernel(gamma=0.5))
+       kernel_regressor=KernelRegression(L2_regularizer=0.1,gamma=0.5) (kernel=GaussianKernel(gamma=0.5))
        kernel_predictor=kernel_regressor(training_set)
        all_results_dataset=kernel_predictor(test_set) # creates a dataset with "output" and "squared_error" field
        outputs = kernel_predictor.compute_outputs(inputs) # inputs and outputs are numpy arrays
@@ -96,7 +96,11 @@
         train_inputs = numpy.array(data['input'])
         Y[0]=1
         Y[1:,:] = numpy.array(data['target'])
-        M,train_inputs_square=self.equations.compute_system_matrix(train_inputs,M)
+        train_inputs_square,sumG=self.equations.compute_system_matrix(train_inputs,M)
+        M[0,1:] = sumG
+        M[1:,0] = 1
+        M[0,0] = M.shape[0]
+        print M
         theta=numpy.linalg.solve(M,Y)
         return KernelPredictor(theta,self.gamma, train_inputs, train_inputs_square)
 
@@ -111,7 +115,7 @@
     b = theta[0]
     alpha = theta[1:,:]
     inputs_square = T.sum(inputs*inputs,axis=1)
-    Kx = exp(-(train_inputs_square-2*dot(inputs,train_inputs.T)+inputs_square)*inv_gamma2)
+    Kx = T.exp(-(train_inputs_square-2*T.dot(inputs,train_inputs.T)+inputs_square)*inv_gamma2)
     outputs = T.dot(Kx,alpha) + b # minibatchsize x n_outputs
     squared_errors = T.sum(T.sqr(targets-outputs),axis=1)
 
@@ -132,15 +136,11 @@
         self.compile()
         
 class KernelRegressionEquations(KernelPredictorEquations):
-    # P = KernelPredictorEquations
     M = T.matrix() # (n_examples+1) x (n_examples+1)
     inputs = T.matrix() # n_examples x n_inputs
     G = M[1:,1:]
-    new_G = gemm(G,1.,inputs,inputs.T,1.)
-    M2 = T.add_inplace(M,new_G)
-    M2[0,0] = M.shape[0]
-    M2[1:,0] = 1
-    M2[0,1:] = T.sum(G,axis=0)
+    new_G = T.gemm(G,1.,inputs,inputs.T,1.)
+    sumG = T.sum(new_G,axis=0)
     inputs_square = T.sum(inputs*inputs,axis=1)
     
     __compiled = False
@@ -152,7 +152,7 @@
         def fn(input_vars,output_vars):
             return staticmethod(theano.function(input_vars,output_vars, linker=linker))
 
-        cls.compute_system_matrix = fn([cls.inputs,cls.M],[cls.M2,cls.inputs_square])
+        cls.compute_system_matrix = fn([cls.inputs,cls.M],[cls.inputs_square,cls.sumG])
 
         cls.__compiled = True
 

--- a/lookup_list.py	Sat Jul 19 10:11:22 2008 -0400
+++ b/lookup_list.py	Sat Jul 19 14:12:41 2008 -0400
@@ -29,6 +29,7 @@
     U{http://epydoc.sourceforge.net/manual-epytext.html#doctest-blocks}
     """
     def __init__(self,names=[],values=[]):
+        print names,values
         assert len(values)==len(names)
         self.__dict__['_values']=values
         self.__dict__['_name2index']={}