Execution Time0.06s

Test: TMVA-DNN-BatchNormalization (Passed)
Build: master-x86_64-ubuntu19-gcc8 (root-ubuntu1904-1) on 2019-11-15 01:38:03
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 var00.332873
output DL 

10x2 matrix is as follows

     |      0    |      1    |
-------------------------------
   0 |     0.6677       1.729 
   1 |    -0.6586     -0.3061 
   2 |      1.775       1.538 
   3 |     0.6759      -2.095 
   4 |     0.8814       1.095 
   5 |      1.432       2.846 
   6 |     -1.046     -0.7246 
   7 |    -0.5282      0.8936 
   8 |     0.1751       3.427 
   9 |   -0.04607     -0.6047 

output BN 
output DL feature 0 mean 0.332873	output DL std 0.919677
output DL feature 1 mean 0.779857	output DL std 1.71578
output of BN 

10x2 matrix is as follows

     |      0    |      1    |
-------------------------------
   0 |     0.3838      0.5832 
   1 |     -1.136     -0.6671 
   2 |      1.653      0.4656 
   3 |     0.3931      -1.766 
   4 |     0.6287      0.1938 
   5 |       1.26       1.269 
   6 |      -1.58     -0.9243 
   7 |    -0.9868      0.0699 
   8 |    -0.1808       1.627 
   9 |    -0.4343     -0.8506 

output BN feature 0 mean 7.77156e-17	output BN std 1.05402
output BN feature 1 mean 6.66134e-17	output BN std 1.05407
Testing weight gradients   for    layer 0
weight gradient for layer 0

2x4 matrix is as follows

     |      0    |      1    |      2    |      3    |
---------------------------------------------------------
   0 |     0.7289      0.5338      0.9622      0.6698 
   1 |    -0.2674     -0.5163     -0.2769     -0.6459 

weights for layer 0

2x4 matrix is as follows

     |      0    |      1    |      2    |      3    |
---------------------------------------------------------
   0 |   -0.09272      -0.407      0.7227     -0.6125 
   1 |      1.009      0.5574       1.231     -0.0464 

 training batch 2 mu var00.332876
compute loss for weight  -0.0927065  -0.0927165 result 2.12149
 training batch 3 mu var00.332873
compute loss for weight  -0.0927265  -0.0927165 result 2.12147
 training batch 4 mu var00.332874
compute loss for weight  -0.0927115  -0.0927165 result 2.12149
 training batch 5 mu var00.332873
compute loss for weight  -0.0927215  -0.0927165 result 2.12148
   --dy = 0.728906 dy_ref = 0.728906
 training batch 6 mu var00.332873
compute loss for weight  -0.407002  -0.407012 result 2.12149
 training batch 7 mu var00.332873
compute loss for weight  -0.407022  -0.407012 result 2.12148
 training batch 8 mu var00.332873
compute loss for weight  -0.407007  -0.407012 result 2.12148
 training batch 9 mu var00.332873
compute loss for weight  -0.407017  -0.407012 result 2.12148
   --dy = 0.533755 dy_ref = 0.533755
 training batch 10 mu var00.332874
compute loss for weight  0.722682  0.722672 result 2.12149
 training batch 11 mu var00.332873
compute loss for weight  0.722662  0.722672 result 2.12147
 training batch 12 mu var00.332873
compute loss for weight  0.722677  0.722672 result 2.12149
 training batch 13 mu var00.332873
compute loss for weight  0.722667  0.722672 result 2.12148
   --dy = 0.962177 dy_ref = 0.962177
 training batch 14 mu var00.332873
compute loss for weight  -0.61253  -0.61254 result 2.12149
 training batch 15 mu var00.332873
compute loss for weight  -0.61255  -0.61254 result 2.12148
 training batch 16 mu var00.332873
compute loss for weight  -0.612535  -0.61254 result 2.12149
 training batch 17 mu var00.332873
compute loss for weight  -0.612545  -0.61254 result 2.12148
   --dy = 0.669832 dy_ref = 0.669832
Testing weight gradients   for    layer 1
weight gradient for layer 1

1x2 matrix is as follows

     |      0    |      1    |
-------------------------------
   0 |       1.62       2.623 

weights for layer 1

1x2 matrix is as follows

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

 training batch 18 mu var00.332873
compute loss for weight  1.00001  1 result 2.1215
 training batch 19 mu var00.332873
compute loss for weight  0.99999  1 result 2.12147
 training batch 20 mu var00.332873
compute loss for weight  1.00001  1 result 2.12149
 training batch 21 mu var00.332873
compute loss for weight  0.999995  1 result 2.12147
   --dy = 1.62028 dy_ref = 1.62028
 training batch 22 mu var00.332873
compute loss for weight  1.00001  1 result 2.12151
 training batch 23 mu var00.332873
compute loss for weight  0.99999  1 result 2.12146
 training batch 24 mu var00.332873
compute loss for weight  1.00001  1 result 2.1215
 training batch 25 mu var00.332873
compute loss for weight  0.999995  1 result 2.12147
   --dy = 2.62269 dy_ref = 2.62269
Testing weight gradients   for    layer 1
weight gradient for layer 1

1x2 matrix is as follows

     |      0    |      1    |
-------------------------------
   0 |   1.11e-16   1.665e-16 

weights for layer 1

1x2 matrix is as follows

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

 training batch 26 mu var00.332873
compute loss for weight  1e-05  0 result 2.12148
 training batch 27 mu var00.332873
compute loss for weight  -1e-05  0 result 2.12148
 training batch 28 mu var00.332873
compute loss for weight  5e-06  0 result 2.12148
 training batch 29 mu var00.332873
compute loss for weight  -5e-06  0 result 2.12148
   --dy = 0 dy_ref = 1.11022e-16
 training batch 30 mu var00.332873
compute loss for weight  1e-05  0 result 2.12148
 training batch 31 mu var00.332873
compute loss for weight  -1e-05  0 result 2.12148
 training batch 32 mu var00.332873
compute loss for weight  5e-06  0 result 2.12148
 training batch 33 mu var00.332873
compute loss for weight  -5e-06  0 result 2.12148
   --dy = 0 dy_ref = 1.66533e-16
Testing weight gradients   for    layer 2
weight gradient for layer 2

1x2 matrix is as follows

     |      0    |      1    |
-------------------------------
   0 |     -2.276      -2.612 

weights for layer 2

1x2 matrix is as follows

     |      0    |      1    |
-------------------------------
   0 |    -0.7119      -1.004 

 training batch 34 mu var00.332873
compute loss for weight  -0.711901  -0.711911 result 2.12146
 training batch 35 mu var00.332873
compute loss for weight  -0.711921  -0.711911 result 2.1215
 training batch 36 mu var00.332873
compute loss for weight  -0.711906  -0.711911 result 2.12147
 training batch 37 mu var00.332873
compute loss for weight  -0.711916  -0.711911 result 2.12149
   --dy = -2.27595 dy_ref = -2.27595
 training batch 38 mu var00.332873
compute loss for weight  -1.00398  -1.00399 result 2.12146
 training batch 39 mu var00.332873
compute loss for weight  -1.004  -1.00399 result 2.12151
 training batch 40 mu var00.332873
compute loss for weight  -1.00398  -1.00399 result 2.12147
 training batch 41 mu var00.332873
compute loss for weight  -1.00399  -1.00399 result 2.1215
   --dy = -2.61227 dy_ref = -2.61227
Testing weight gradients:      maximum relative error: [NON-XML-CHAR-0x1B][32m1.68883e-10[NON-XML-CHAR-0x1B][39m