I'm getting some test failures in my code with infs in numpy 1.4, nan treatment has been corrected in numpy.testing
>>> assert_almost_equal(np.array([-5.23722324, -np.nan, -5.23722324]), >>> np.array([-5.23722324, -np.nan, -5.23722324])) but it seems that inf doesn't compare almost equal >>> assert_almost_equal(np.array([-5.23722324, -np.inf, -5.23722324]), >>> np.array([-5.23722324, -np.inf, -5.23722324])) Traceback (most recent call last): ... ValueError: Arrays are not almost equal x: array([-5.23722324, -Inf, -5.23722324]) y: array([-5.23722324, -Inf, -5.23722324]) numpy 1.4.0 has this been changed already? or is it not desired? Josef _______________________________________________ NumPy-Discussion mailing list NumPy-Discussion@scipy.org http://mail.scipy.org/mailman/listinfo/numpy-discussion
