Execution Time1.54s

Test: TMVA-DNN-Optimization-Cpu (Passed)
Build: PR-4279-x86_64-fedora29-gcc8-opt (root-fedora29-3.cern.ch) on 2019-11-14 21:03:34

Test Timing: Passed
Processors1

Show Command Line
Display graphs:

Test output
Testing optimization: (single precision)
Stochastic Gradient Descent:  Before Training: Mean Absolute Error = 0.75221, No of Epochs = 200, After Training: Mean Absolute error = 9.37405e-07
Stochastic Gradient Descent with momentum:  Before Training: Mean Absolute Error = 0.840328, No of Epochs = 200, After Training: Mean Absolute error = 1.38267e-07
Adam Optimizer:  Before Training: Mean Absolute Error = 0.74293, No of Epochs = 200, After Training: Mean Absolute error = 1.38487e-07
Adagrad Optimizer:  Before Training: Mean Absolute Error = 0.797774, No of Epochs = 200, After Training: Mean Absolute error = 1.00094e-07
RMSProp Optimizer without momentum:  Before Training: Mean Absolute Error = 0.867554, No of Epochs = 119, After Training: Mean Absolute error = 0.011635
RMSProp Optimizer with momentum:  Before Training: Mean Absolute Error = 0.864047, No of Epochs = 200, After Training: Mean Absolute error = 0.02839
Adadelta Optimizer:  Before Training: Mean Absolute Error = 1.04865, No of Epochs = 165, After Training: Mean Absolute error = 0.0062976