fmcquillan99 commented on issue #463: DL: Add asymmetric cluster check for predict URL: https://github.com/apache/madlib/pull/463#issuecomment-566343228 looks better ``` Time for training in iteration 4: 8.73104310036 sec CONTEXT: PL/Python function "madlib_keras_fit_multiple_model" INFO: Time for training in iteration 5: 8.39556002617 sec CONTEXT: PL/Python function "madlib_keras_fit_multiple_model" INFO: Time for training in iteration 6: 8.93426513672 sec DETAIL: Training set after iteration 6: mst_key=2: metric=0.691666662693, loss=0.472569614649 mst_key=11: metric=0.975000023842, loss=0.136468976736 mst_key=10: metric=0.949999988079, loss=0.109081745148 mst_key=4: metric=0.975000023842, loss=0.178229421377 mst_key=3: metric=0.925000011921, loss=0.184964418411 mst_key=12: metric=0.975000023842, loss=0.141136452556 mst_key=9: metric=0.958333313465, loss=0.0976895838976 mst_key=1: metric=0.375, loss=1.09557545185 mst_key=8: metric=0.975000023842, loss=0.0692544877529 mst_key=7: metric=0.958333313465, loss=0.153927713633 mst_key=5: metric=0.841666638851, loss=0.468217223883 mst_key=6: metric=0.741666674614, loss=0.504250705242 Validation set after iteration 6: mst_key=2: metric=0.600000023842, loss=0.418071746826 mst_key=11: metric=1.0, loss=0.0674076601863 mst_key=10: metric=1.0, loss=0.018841393292 mst_key=4: metric=1.0, loss=0.101651079953 mst_key=3: metric=0.933333337307, loss=0.149523466825 mst_key=12: metric=1.0, loss=0.0722744837403 mst_key=9: metric=1.0, loss=0.0356690809131 mst_key=1: metric=0.166666671634, loss=1.12900209427 mst_key=8: metric=1.0, loss=0.00220014923252 mst_key=7: metric=1.0, loss=0.0267392881215 mst_key=5: metric=0.800000011921, loss=0.491473942995 mst_key=6: metric=0.600000023842, loss=0.556908965111 CONTEXT: PL/Python function "madlib_keras_fit_multiple_model" INFO: Time for training in iteration 7: 8.56644105911 sec CONTEXT: PL/Python function "madlib_keras_fit_multiple_model" INFO: ``` LGTM
---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: [email protected] With regards, Apache Git Services
