Implemented non-sparse grad.

theano/sparse/sandbox/sp.py

@@ -31,7 +31,7 @@ class SpSum(Op):
         :param sparse_grad: if True, this instance ignores the gradient on matrix elements which are implicitly 0.
         """
         super(SpSum, self).__init__()
-        self.axis=axis
+        self.axis = axis
         self.sparse_grad = sparse_grad
         if self.axis not in (None, 0, 1):
             raise ValueError('illegal value for self.axis')
@@ -54,9 +54,8 @@ class SpSum(Op):
         # At least for small matrices (5x5), the .sum() method of a csc matrix returns a dense matrix
         # as the result whether axis is 0 or 1... weird!
         ###
-        #make sure a sparse type has been given in input
         assert isinstance(x.type, theano.sparse.SparseType)
-        b=()
+        b = ()
         if self.axis is not None:
             b=(False,)
         z = tensor.tensor(broadcastable=b, dtype=x.dtype)
@@ -76,7 +75,7 @@ class SpSum(Op):
         if self.axis is None:
             z[0] = numpy.asarray(x.sum())
         else: 
-            s = set(xrange(len(x.shape)))-set([self.axis])
+            s = set(xrange(len(x.shape))) - set([self.axis])
             myreshape = map((lambda i: x.shape[i]), s)
             if x.format not in ('csc', 'csr'):
                 x = x.asformat(x.format)
@@ -95,31 +94,20 @@ class SpSum(Op):
             #        z[0] = numpy.asarray(x.asformat(x.format).sum(axis=self.axis)).reshape((x.shape[0],))
 
     def grad(self,(x,), (gz,)):
-        print 'grad (sparse:%s):'%self.sparse_grad,x,gz
-        if self.sparse_grad:
         if self.axis is None:
-                return [gz * theano.sparse.sp_ones_like(x)]
+            r = gz * theano.sparse.sp_ones_like(x)
         elif self.axis == 0:
-                return col_scale(theano.sparse.sp_ones_like(x), gz)
+            r = col_scale(theano.sparse.sp_ones_like(x), gz)
         elif self.axis == 1:
-                return row_scale(theano.sparse.sp_ones_like(x), gz)
-            else:
-                assert False
-        else:
-            # if sparse_grad is False, this instance does not ignore the
-            # gradient on matrix elements which are implicitly 0,
-            # and return a dense version of the data.
-            if self.axis is None:
-                return [gz * tensor.ones_like(theano.sparse.dense_from_sparse(x))]
-            elif self.axis == 0:
-                raise NotImplementedError('non sparse grad, axis=0')
-                pass
-            elif self.axis == 1:
-                raise NotImplementedError('non sparse grad, axis=1')
-                pass
+            r = row_scale(theano.sparse.sp_ones_like(x), gz)
         else:
             assert False
 
+        if not self.sparse_grad:
+            r = theano.sparse.dense_from_sparse(r)
+
+        return [r]
+
 def sp_sum(x, axis=None, sparse_grad=False):
     return SpSum(axis, sparse_grad)(x)
 

theano/sparse/sandbox/test_sp.py

@@ -372,21 +372,19 @@ class TestSP(unittest.TestCase):
         # TODO: test both grad.
         rng = numpy.random.RandomState(42)
         from theano.sparse.basic import SparseFromDense,DenseFromSparse
-        for format,cast in [("csc",scipy.sparse.csc_matrix), ("csr",scipy.sparse.csr_matrix)]:
+        cases = [("csc", scipy.sparse.csc_matrix), ("csr", scipy.sparse.csr_matrix)]
         
-            print 'format: %(format)s'%locals()
+        for format, cast in cases:
 			
+            print 'format: %(format)s'%locals()
             x = theano.sparse.SparseType(format=format,
                                          dtype=theano.config.floatX)()
             x_data = numpy.arange(20).reshape(5,4).astype(theano.config.floatX)
 
-            #print 'x_data:',x_data
-            #print 'x_data.sum():',x_data.sum()
-
             # Sum on all axis
             print 'sum on all axis...'
             z = theano.sparse.sandbox.sp.sp_sum(x)
-            assert z.type.broadcastable==()
+            assert z.type.broadcastable == ()
             f = theano.function([x], z)
             x_val = cast(x_data)
             out = f(x_val)
@@ -396,17 +394,17 @@ class TestSP(unittest.TestCase):
             # Sum on axis 0
             print 'sum on axis 0...'
             z = theano.sparse.sandbox.sp.sp_sum(x, axis=0)
-            assert z.type.broadcastable==(False,)
+            assert z.type.broadcastable == (False,)
             f = theano.function([x], z)
             x_val = cast(x_data)
             out = f(x_val)
-            expected=x_val.sum(axis=0)
+            expected = x_val.sum(axis=0)
             assert (out == expected).all()
 
             # Sum on axis 1
             print 'sum on axis 1...'
             z = theano.sparse.sandbox.sp.sp_sum(x, axis=1)
-            assert z.type.broadcastable==(False,)
+            assert z.type.broadcastable == (False,)
             f = theano.function([x], z)
             x_val = cast(x_data)
             out = f(x_val)
@@ -414,7 +412,7 @@ class TestSP(unittest.TestCase):
             assert (out == expected).all()
 
             # Sparse gradient on Sum on all axis
-            # suspended until som estuff get fixed =/
+            # unfinished, and suspended until verify_grad get fixed
             if False:
                 print 'grad on sum on all axis...'
                 def fun(x):
@@ -422,7 +420,7 @@ class TestSP(unittest.TestCase):
                     # x is a dense matrix: make it sparse
                     sparse_var = SparseFromDense(format)(x)
                     # apply op
-                    dense_sum = theano.sparse.sandbox.sp.SpSum(axis=None,sparse_grad=False)(sparse_var)
+                    dense_sum = theano.sparse.sandbox.sp.SpSum(axis=None, sparse_grad=False)(sparse_var)
                     return dense_sum
                     # cast back to dense so that verify_grad can work
                     dense_sum = theano.sparse.DenseFromSparse()(sparse_sum)