Execution Time1.01s

Test: TMVA-DNN-BatchNormalization (Passed)
Build: master-x86_64-ubuntu18-clang91-dbg (sft-ubuntu-1804-3) on 2019-11-14 11:42:52
Repository revision: 14de58de35eff907054671888ccc2de0f7f27e77

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 

10x4 matrix is as follows

     |      0    |      1    |      2    |      3    |
---------------------------------------------------------
   0 |     0.9989     -0.4348      0.7818    -0.03005 
   1 |     0.8243    -0.05672     -0.9009     -0.0747 
   2 |   0.007912     -0.4108       1.391     -0.9851 
   3 |   -0.04894      -1.443      -1.061      -1.388 
   4 |     0.7674      -0.736      0.5797     -0.3821 
   5 |      2.061      -1.235       1.165     -0.4542 
   6 |    -0.1348     -0.4996     -0.1824       1.844 
   7 |    -0.2428       1.997    0.004806     -0.4222 
   8 |      1.541     0.09474       1.525       1.217 
   9 |    -0.1363     -0.1992     -0.2938     -0.1184 

 training batch 1 mu var0-0.180541
output DL 

10x2 matrix is as follows

     |      0    |      1    |
-------------------------------
   0 |   -0.07775      -1.323 
   1 |     -1.465     -0.9861 
   2 |     0.5571     0.05964 
   3 |     -1.183       0.815 
   4 |    -0.1622      -0.846 
   5 |    -0.6554      -2.522 
   6 |      1.553      0.1435 
   7 |    -0.9829     -0.2413 
   8 |     0.7244       -2.47 
   9 |    -0.1136      0.2992 

output BN 
output DL feature 0 mean -0.180541	output DL std 0.937406
output DL feature 1 mean -0.707111	output DL std 1.14192
output of BN 

10x2 matrix is as follows

     |      0    |      1    |
-------------------------------
   0 |     0.1156     -0.5688 
   1 |     -1.445     -0.2575 
   2 |     0.8294      0.7078 
   3 |     -1.127       1.405 
   4 |    0.02067     -0.1282 
   5 |    -0.5339      -1.675 
   6 |      1.949      0.7852 
   7 |    -0.9022        0.43 
   8 |      1.018      -1.627 
   9 |    0.07529      0.9289 

output BN feature 0 mean 3.46945e-17	output BN std 1.05403
output BN feature 1 mean -2.22045e-17	output BN std 1.05405
Testing weight gradients   for    layer 0
weight gradient for layer 0

2x4 matrix is as follows

     |      0    |      1    |      2    |      3    |
---------------------------------------------------------
   0 |     -2.786     0.08282      -2.126      -0.949 
   1 |       5.05     -0.3999       5.435       3.882 

weights for layer 0

2x4 matrix is as follows

     |      0    |      1    |      2    |      3    |
---------------------------------------------------------
   0 |    -0.8719     -0.4444      0.7958      0.7367 
   1 |     -1.366     -0.3118     -0.1256     -0.1189 

 training batch 2 mu var0-0.180539
compute loss for weight  -0.871936  -0.871946 result 5.38098
 training batch 3 mu var0-0.180541
compute loss for weight  -0.871956  -0.871946 result 5.38104
 training batch 4 mu var0-0.180541
compute loss for weight  -0.871941  -0.871946 result 5.381
 training batch 5 mu var0-0.180541
compute loss for weight  -0.871951  -0.871946 result 5.38103
   --dy = -2.78569 dy_ref = -2.78569
 training batch 6 mu var0-0.180542
compute loss for weight  -0.444436  -0.444446 result 5.38101
 training batch 7 mu var0-0.180541
compute loss for weight  -0.444456  -0.444446 result 5.38101
 training batch 8 mu var0-0.180542
compute loss for weight  -0.444441  -0.444446 result 5.38101
 training batch 9 mu var0-0.180541
compute loss for weight  -0.444451  -0.444446 result 5.38101
   --dy = 0.0828241 dy_ref = 0.0828241
 training batch 10 mu var0-0.180541
compute loss for weight  0.795835  0.795825 result 5.38099
 training batch 11 mu var0-0.180541
compute loss for weight  0.795815  0.795825 result 5.38103
 training batch 12 mu var0-0.180541
compute loss for weight  0.79583  0.795825 result 5.381
 training batch 13 mu var0-0.180541
compute loss for weight  0.79582  0.795825 result 5.38102
   --dy = -2.1259 dy_ref = -2.1259
 training batch 14 mu var0-0.180541
compute loss for weight  0.736702  0.736692 result 5.381
 training batch 15 mu var0-0.180541
compute loss for weight  0.736682  0.736692 result 5.38102
 training batch 16 mu var0-0.180541
compute loss for weight  0.736697  0.736692 result 5.38101
 training batch 17 mu var0-0.180541
compute loss for weight  0.736687  0.736692 result 5.38102
   --dy = -0.948976 dy_ref = -0.948976
Testing weight gradients   for    layer 1
weight gradient for layer 1

1x2 matrix is as follows

     |      0    |      1    |
-------------------------------
   0 |      9.645       1.117 

weights for layer 1

1x2 matrix is as follows

     |      0    |      1    |
-------------------------------
   0 |          1           1 

 training batch 18 mu var0-0.180541
compute loss for weight  1.00001  1 result 5.38111
 training batch 19 mu var0-0.180541
compute loss for weight  0.99999  1 result 5.38092
 training batch 20 mu var0-0.180541
compute loss for weight  1.00001  1 result 5.38106
 training batch 21 mu var0-0.180541
compute loss for weight  0.999995  1 result 5.38096
   --dy = 9.64477 dy_ref = 9.64477
 training batch 22 mu var0-0.180541
compute loss for weight  1.00001  1 result 5.38102
 training batch 23 mu var0-0.180541
compute loss for weight  0.99999  1 result 5.381
 training batch 24 mu var0-0.180541
compute loss for weight  1.00001  1 result 5.38102
 training batch 25 mu var0-0.180541
compute loss for weight  0.999995  1 result 5.38101
   --dy = 1.11725 dy_ref = 1.11725
Testing weight gradients   for    layer 1
weight gradient for layer 1

1x2 matrix is as follows

     |      0    |      1    |
-------------------------------
   0 |  4.996e-16   8.327e-17 

weights for layer 1

1x2 matrix is as follows

     |      0    |      1    |
-------------------------------
   0 |          0           0 

 training batch 26 mu var0-0.180541
compute loss for weight  1e-05  0 result 5.38101
 training batch 27 mu var0-0.180541
compute loss for weight  -1e-05  0 result 5.38101
 training batch 28 mu var0-0.180541
compute loss for weight  5e-06  0 result 5.38101
 training batch 29 mu var0-0.180541
compute loss for weight  -5e-06  0 result 5.38101
   --dy = 0 dy_ref = 4.996e-16
 training batch 30 mu var0-0.180541
compute loss for weight  1e-05  0 result 5.38101
 training batch 31 mu var0-0.180541
compute loss for weight  -1e-05  0 result 5.38101
 training batch 32 mu var0-0.180541
compute loss for weight  5e-06  0 result 5.38101
 training batch 33 mu var0-0.180541
compute loss for weight  -5e-06  0 result 5.38101
   --dy = 0 dy_ref = 8.32667e-17
Testing weight gradients   for    layer 2
weight gradient for layer 2

1x2 matrix is as follows

     |      0    |      1    |
-------------------------------
   0 |     -4.373      -1.443 

weights for layer 2

1x2 matrix is as follows

     |      0    |      1    |
-------------------------------
   0 |     -2.205     -0.7745 

 training batch 34 mu var0-0.180541
compute loss for weight  -2.20547  -2.20548 result 5.38097
 training batch 35 mu var0-0.180541
compute loss for weight  -2.20549  -2.20548 result 5.38106
 training batch 36 mu var0-0.180541
compute loss for weight  -2.20547  -2.20548 result 5.38099
 training batch 37 mu var0-0.180541
compute loss for weight  -2.20548  -2.20548 result 5.38103
   --dy = -4.3731 dy_ref = -4.3731
 training batch 38 mu var0-0.180541
compute loss for weight  -0.774462  -0.774472 result 5.381
 training batch 39 mu var0-0.180541
compute loss for weight  -0.774482  -0.774472 result 5.38103
 training batch 40 mu var0-0.180541
compute loss for weight  -0.774467  -0.774472 result 5.38101
 training batch 41 mu var0-0.180541
compute loss for weight  -0.774477  -0.774472 result 5.38102
   --dy = -1.4426 dy_ref = -1.4426
Testing weight gradients:      maximum relative error: [NON-XML-CHAR-0x1B][33m1.1472e-09[NON-XML-CHAR-0x1B][39m