Merge pull request #2352 from lamblin/fix_conv2d_stride_check

theano/tensor/nnet/conv.py

@@ -964,7 +964,7 @@ class ConvOp(OpenMPOp):
         return ['<numpy/noprefix.h>', '<iostream>', '<sstream>']
 
     def c_code_cache_version(self):
-        return (11, self.openmp, blas.blas_header_version())
+        return (12, self.openmp, blas.blas_header_version())
 
     def c_support_code(self):
         return """
@@ -2157,11 +2157,12 @@ if ((!%(z)s)
 }
 z_arr = (PyArrayObject*) %(z)s;
 
-//assertions
-if (PyArray_STRIDES(%(z)s)[0] != PyArray_DIMS(%(z)s)[1] *PyArray_DIMS(%(z)s)[2] *PyArray_DIMS(%(z)s)[3] * sizeof(%(type)s)) %(fail)s;
-if (PyArray_STRIDES(%(z)s)[1] != PyArray_DIMS(%(z)s)[2] * PyArray_DIMS(%(z)s)[3] * sizeof(%(type)s)) %(fail)s;
-if (PyArray_STRIDES(%(z)s)[2] != PyArray_DIMS(%(z)s)[3] * sizeof(%(type)s)) %(fail)s;
-if (PyArray_STRIDES(%(z)s)[3] != sizeof(%(type)s)) %(fail)s;
+// assert the output is C-contiguous
+if (!PyArray_ISCONTIGUOUS(%(z)s))
+{
+    PyErr_SetString(PyExc_AssertionError, "Output (%(z)s) not contiguous");
+    %(fail)s;
+}
 
 //The if on the number of loop make a speed up for small array.
 //with g++ 4.5.1. The compiler should be smart enough to do this himself!