Author: bugman Date: Tue Aug 26 12:20:49 2014 New Revision: 25279 URL: http://svn.gna.org/viewcvs/relax?rev=25279&view=rev Log: The exponential curve numeric gradient script now uses only floating point numbers. This is to avoid integer truncation problems. Modified: trunk/test_suite/shared_data/curve_fitting/numeric_gradient/integrate.log trunk/test_suite/shared_data/curve_fitting/numeric_gradient/integrate.py Modified: trunk/test_suite/shared_data/curve_fitting/numeric_gradient/integrate.log URL: http://svn.gna.org/viewcvs/relax/trunk/test_suite/shared_data/curve_fitting/numeric_gradient/integrate.log?rev=25279&r1=25278&r2=25279&view=diff ============================================================================== --- trunk/test_suite/shared_data/curve_fitting/numeric_gradient/integrate.log (original) +++ trunk/test_suite/shared_data/curve_fitting/numeric_gradient/integrate.log Tue Aug 26 12:20:49 2014 @@ -1,4 +1,4 @@ -The gradient at [1, 1000] is: +The gradient at [1.0, 1000.0] is: [-1.0995282792650802e-09, 2.1826111665238544e-12] -The gradient at [2, 500] is: +The gradient at [2.0, 500.0] is: [722.67864120737488, -11.564651301654292] Modified: trunk/test_suite/shared_data/curve_fitting/numeric_gradient/integrate.py URL: http://svn.gna.org/viewcvs/relax/trunk/test_suite/shared_data/curve_fitting/numeric_gradient/integrate.py?rev=25279&r1=25278&r2=25279&view=diff ============================================================================== --- trunk/test_suite/shared_data/curve_fitting/numeric_gradient/integrate.py (original) +++ trunk/test_suite/shared_data/curve_fitting/numeric_gradient/integrate.py Tue Aug 26 12:20:49 2014 @@ -45,17 +45,17 @@ # The real parameters. -R = 1 -I0 = 1000 +R = 1.0 +I0 = 1000.0 # The time points. -times = [0, 1, 2, 3, 4] +times = [0.0, 1.0, 2.0, 3.0, 4.0] # The intensities for the above I0 and R. I = [1000.0, 367.879441171, 135.335283237, 49.7870683679, 18.3156388887] # The intensity errors. -errors = [10, 10, 10, 10, 10] +errors = [10.0, 10.0, 10.0, 10.0, 10.0] # The numeric gradient at the minimum. grad_R = derivative(func_R, R, dx=1e-5, order=11) @@ -63,8 +63,8 @@ print("The gradient at %s is:\n %s" % ([R, I0], [grad_R, grad_I])) # The numeric gradient off the minimum. -R_off = 2 -I0_off = 500 +R_off = 2.0 +I0_off = 500.0 grad_R = derivative(func_R, R_off, dx=1e-5, order=11) grad_I = derivative(func_I, I0_off, dx=1e-5, order=11) print("The gradient at %s is:\n %s" % ([R_off, I0_off], [grad_R, grad_I]))