1. Reported is the test error (TeVErr), for the model selected such
that the cross validation error (see average over 10 folds as reported
in ip_don_all_results.txt) is smallest (independent on each of the
five realizations of the data set.).

run:1 d1:3 d2:1 C:4.00e+00 l:5 => TeVErr:5.1771e-02 #svs=352
run:2 d1:3 d2:1 C:2.00e+00 l:6 => TeVErr:5.6548e-02 #svs=357
run:3 d1:3 d2:1 C:7.00e+00 l:5 => TeVErr:6.5963e-02 #svs=341
run:4 d1:3 d2:1 C:2.00e+00 l:4 => TeVErr:4.3956e-02 #svs=347
run:5 d1:3 d2:1 C:8.00e-01 l:3 => TeVErr:5.9880e-02 #svs=348
run:6 d1:3 d2:1 C:7.00e+00 l:6 => TeVErr:6.6116e-02 #svs=338
run:7 d1:3 d2:1 C:1.50e+00 l:5 => TeVErr:4.9451e-02 #svs=361
run:8 d1:3 d2:1 C:4.00e+00 l:4 => TeVErr:6.9767e-02 #svs=340
run:9 d1:3 d2:1 C:7.00e+00 l:6 => TeVErr:4.3103e-02 #svs=367
run:10 d1:2 d2:1 C:1.50e+00 l:5 => TeVErr:8.5960e-02 #svs=288

TeVErr average+/-std. deviation: 5.93 +/- 1.32 %