Execution Time0.19s

Test: TMVA-DNN-BatchNormalization-Cpu (Passed)
Build: master-x86_64-centos7-clang100-opt-no-rt-cxxmodules (olsnba08.cern.ch) on 2019-11-14 01:39:53
Repository revision: 32b17abcda23e44b64218a42d0ca69cb30cda7e0

Test Timing: Passed
Processors1

Show Command Line
Display graphs:

Test output
Testing Backpropagation:
DEEP NEURAL NETWORK:   Depth = 3  Input = ( 1, 10, 4 )  Batch size = 10  Loss function = R
	Layer 0	 DENSE Layer: 	 ( Input =     4 , Width =     2 ) 	Output = (  1 ,    10 ,     2 ) 	 Activation Function = Identity
	Layer 1	 BATCH NORM Layer: 	 ( Input =     2 ) 
	Layer 2	 DENSE Layer: 	 ( Input =     2 , Width =     1 ) 	Output = (  1 ,    10 ,     1 ) 	 Activation Function = Identity
input 
 training batch 1 mu var00.495578
output DL 
output BN 
output DL feature 0 mean 0.495578	output DL std 0.74253
output DL feature 1 mean -0.0211552	output DL std 0.549256
output of BN 
output BN feature 0 mean 2.22045e-17	output BN std 1.05399
output BN feature 1 mean 6.10623e-17	output BN std 1.0539
Testing weight gradients   for    layer 0
weight gradient for layer 0
weights for layer 0
 training batch 2 mu var00.495581
compute loss for weight  0.634091  0.634081 result 1.73373
 training batch 3 mu var00.495578
compute loss for weight  0.634071  0.634081 result 1.73372
 training batch 4 mu var00.495579
compute loss for weight  0.634086  0.634081 result 1.73372
 training batch 5 mu var00.495578
compute loss for weight  0.634076  0.634081 result 1.73372
   --dy = 0.41774 dy_ref = 0.41774
 training batch 6 mu var00.495578
compute loss for weight  -0.0956795  -0.0956895 result 1.73371
 training batch 7 mu var00.495578
compute loss for weight  -0.0956995  -0.0956895 result 1.73373
 training batch 8 mu var00.495578
compute loss for weight  -0.0956845  -0.0956895 result 1.73372
 training batch 9 mu var00.495578
compute loss for weight  -0.0956945  -0.0956895 result 1.73373
   --dy = -1.12248 dy_ref = -1.12248
 training batch 10 mu var00.495579
compute loss for weight  0.330057  0.330047 result 1.7337
 training batch 11 mu var00.495578
compute loss for weight  0.330037  0.330047 result 1.73374
 training batch 12 mu var00.495579
compute loss for weight  0.330052  0.330047 result 1.73371
 training batch 13 mu var00.495578
compute loss for weight  0.330042  0.330047 result 1.73373
   --dy = -1.83783 dy_ref = -1.83783
 training batch 14 mu var00.495578
compute loss for weight  -0.136379  -0.136389 result 1.7337
 training batch 15 mu var00.495578
compute loss for weight  -0.136399  -0.136389 result 1.73374
 training batch 16 mu var00.495578
compute loss for weight  -0.136384  -0.136389 result 1.73371
 training batch 17 mu var00.495578
compute loss for weight  -0.136394  -0.136389 result 1.73373
   --dy = -1.72068 dy_ref = -1.72068
Testing weight gradients   for    layer 1
weight gradient for layer 1
weights for layer 1
 training batch 18 mu var00.495578
compute loss for weight  1.00001  1 result 1.73374
 training batch 19 mu var00.495578
compute loss for weight  0.99999  1 result 1.7337
 training batch 20 mu var00.495578
compute loss for weight  1.00001  1 result 1.73373
 training batch 21 mu var00.495578
compute loss for weight  0.999995  1 result 1.73371
   --dy = 1.98907 dy_ref = 1.98907
 training batch 22 mu var00.495578
compute loss for weight  1.00001  1 result 1.73374
 training batch 23 mu var00.495578
compute loss for weight  0.99999  1 result 1.73371
 training batch 24 mu var00.495578
compute loss for weight  1.00001  1 result 1.73373
 training batch 25 mu var00.495578
compute loss for weight  0.999995  1 result 1.73371
   --dy = 1.47837 dy_ref = 1.47837
Testing weight gradients   for    layer 1
weight gradient for layer 1
weights for layer 1
 training batch 26 mu var00.495578
compute loss for weight  1e-05  0 result 1.73372
 training batch 27 mu var00.495578
compute loss for weight  -1e-05  0 result 1.73372
 training batch 28 mu var00.495578
compute loss for weight  5e-06  0 result 1.73372
 training batch 29 mu var00.495578
compute loss for weight  -5e-06  0 result 1.73372
   --dy = 0 dy_ref = 1.11022e-16
 training batch 30 mu var00.495578
compute loss for weight  1e-05  0 result 1.73372
 training batch 31 mu var00.495578
compute loss for weight  -1e-05  0 result 1.73372
 training batch 32 mu var00.495578
compute loss for weight  5e-06  0 result 1.73372
 training batch 33 mu var00.495578
compute loss for weight  -5e-06  0 result 1.73372
   --dy = 0 dy_ref = 1.11022e-16
Testing weight gradients   for    layer 2
weight gradient for layer 2
weights for layer 2
 training batch 34 mu var00.495578
compute loss for weight  1.18074  1.18073 result 1.73374
 training batch 35 mu var00.495578
compute loss for weight  1.18072  1.18073 result 1.73371
 training batch 36 mu var00.495578
compute loss for weight  1.18074  1.18073 result 1.73373
 training batch 37 mu var00.495578
compute loss for weight  1.18073  1.18073 result 1.73371
   --dy = 1.68461 dy_ref = 1.68461
 training batch 38 mu var00.495578
compute loss for weight  -1.06719  -1.0672 result 1.73371
 training batch 39 mu var00.495578
compute loss for weight  -1.06721  -1.0672 result 1.73374
 training batch 40 mu var00.495578
compute loss for weight  -1.06719  -1.0672 result 1.73371
 training batch 41 mu var00.495578
compute loss for weight  -1.0672  -1.0672 result 1.73373
   --dy = -1.38528 dy_ref = -1.38528
Testing weight gradients:      maximum relative error: [NON-XML-CHAR-0x1B][32m1.62408e-10[NON-XML-CHAR-0x1B][39m