Problem: numpy raised TypeError: Cannot cast ufunc add output from dt…

…ype('complex128') to dtype('float64') with casting rule 'same_kind'

Solution: use numpy.add(x,y,out=x, casting='unsafe') to cast from complex to float if necessary

algopy/utpm/algorithms.py

@@ -1190,9 +1190,9 @@ def _dot(cls, x_data, y_data, out = None):
         for d in range(D):
             for p in range(P):
                 for c in range(d+1):
-                    z_data[d,p,...] += numpy.dot(
-                            x_data[c,p,...],
-                            y_data[d-c,p,...])
+                    tmp = numpy.dot(x_data[c,p,...],
+                                    y_data[d-c,p,...])
+                    numpy.add(z_data[d,p,...], tmp, out=z_data[d,p, ...], casting='unsafe') 
 
         return out