Execution Time0.05s

Test: TMVA-DNN-BatchNormalization (Passed)
Build: master-x86_64-ubuntu19-gcc9 (root-ubuntu-1910-1) on 2019-11-13 00:48:32
Repository revision: 30660dce2d9e89e4852dbf83dbd8b2cfcc137eff

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 var00.386124
output DL 

10x2 matrix is as follows

     |      0    |      1    |
-------------------------------
   0 |      0.751     -0.9445 
   1 |     0.1906       1.594 
   2 |   -0.03881      -1.586 
   3 |    -0.3221       1.041 
   4 |      0.558     -0.8161 
   5 |       1.41      -1.388 
   6 |     0.9565       -1.45 
   7 |     -1.065       1.899 
   8 |      1.518      -2.035 
   9 |   -0.09666      0.2337 

output BN 
output DL feature 0 mean 0.386124	output DL std 0.80817
output DL feature 1 mean -0.345227	output DL std 1.42744
output of BN 

10x2 matrix is as follows

     |      0    |      1    |
-------------------------------
   0 |     0.4759     -0.4425 
   1 |     -0.255       1.432 
   2 |    -0.5542     -0.9162 
   3 |    -0.9236       1.023 
   4 |     0.2242     -0.3477 
   5 |      1.336     -0.7701 
   6 |     0.7439     -0.8158 
   7 |     -1.893       1.657 
   8 |      1.476      -1.248 
   9 |    -0.6296      0.4275 

output BN feature 0 mean -1.33227e-16	output BN std 1.054
output BN feature 1 mean -5.55112e-18	output BN std 1.05406
Testing weight gradients   for    layer 0
weight gradient for layer 0

2x4 matrix is as follows

     |      0    |      1    |      2    |      3    |
---------------------------------------------------------
   0 |    0.04231    0.008956     -0.1025    0.007493 
   1 |    0.03166      0.0046    -0.06951     0.00716 

weights for layer 0

2x4 matrix is as follows

     |      0    |      1    |      2    |      3    |
---------------------------------------------------------
   0 |     0.4532     -0.3801      0.1882      0.4674 
   1 |     0.4669       0.874      -1.344     -0.6482 

 training batch 2 mu var00.386127
compute loss for weight  0.453164  0.453154 result 2.99668
 training batch 3 mu var00.386124
compute loss for weight  0.453144  0.453154 result 2.99668
 training batch 4 mu var00.386125
compute loss for weight  0.453159  0.453154 result 2.99668
 training batch 5 mu var00.386124
compute loss for weight  0.453149  0.453154 result 2.99668
   --dy = 0.0423068 dy_ref = 0.0423068
 training batch 6 mu var00.386124
compute loss for weight  -0.380099  -0.380109 result 2.99668
 training batch 7 mu var00.386124
compute loss for weight  -0.380119  -0.380109 result 2.99668
 training batch 8 mu var00.386124
compute loss for weight  -0.380104  -0.380109 result 2.99668
 training batch 9 mu var00.386124
compute loss for weight  -0.380114  -0.380109 result 2.99668
   --dy = 0.00895638 dy_ref = 0.00895638
 training batch 10 mu var00.386125
compute loss for weight  0.18824  0.18823 result 2.99668
 training batch 11 mu var00.386124
compute loss for weight  0.18822  0.18823 result 2.99668
 training batch 12 mu var00.386125
compute loss for weight  0.188235  0.18823 result 2.99668
 training batch 13 mu var00.386124
compute loss for weight  0.188225  0.18823 result 2.99668
   --dy = -0.102541 dy_ref = -0.102541
 training batch 14 mu var00.386124
compute loss for weight  0.467388  0.467378 result 2.99668
 training batch 15 mu var00.386124
compute loss for weight  0.467368  0.467378 result 2.99668
 training batch 16 mu var00.386124
compute loss for weight  0.467383  0.467378 result 2.99668
 training batch 17 mu var00.386124
compute loss for weight  0.467373  0.467378 result 2.99668
   --dy = 0.00749298 dy_ref = 0.00749298
Testing weight gradients   for    layer 1
weight gradient for layer 1

1x2 matrix is as follows

     |      0    |      1    |
-------------------------------
   0 |    -0.1894       6.183 

weights for layer 1

1x2 matrix is as follows

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

 training batch 18 mu var00.386124
compute loss for weight  1.00001  1 result 2.99668
 training batch 19 mu var00.386124
compute loss for weight  0.99999  1 result 2.99668
 training batch 20 mu var00.386124
compute loss for weight  1.00001  1 result 2.99668
 training batch 21 mu var00.386124
compute loss for weight  0.999995  1 result 2.99668
   --dy = -0.189384 dy_ref = -0.189384
 training batch 22 mu var00.386124
compute loss for weight  1.00001  1 result 2.99674
 training batch 23 mu var00.386124
compute loss for weight  0.99999  1 result 2.99662
 training batch 24 mu var00.386124
compute loss for weight  1.00001  1 result 2.99671
 training batch 25 mu var00.386124
compute loss for weight  0.999995  1 result 2.99665
   --dy = 6.18274 dy_ref = 6.18274
Testing weight gradients   for    layer 1
weight gradient for layer 1

1x2 matrix is as follows

     |      0    |      1    |
-------------------------------
   0 | -1.561e-17  -2.776e-16 

weights for layer 1

1x2 matrix is as follows

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

 training batch 26 mu var00.386124
compute loss for weight  1e-05  0 result 2.99668
 training batch 27 mu var00.386124
compute loss for weight  -1e-05  0 result 2.99668
 training batch 28 mu var00.386124
compute loss for weight  5e-06  0 result 2.99668
 training batch 29 mu var00.386124
compute loss for weight  -5e-06  0 result 2.99668
   --dy = 0 dy_ref = -1.56125e-17
 training batch 30 mu var00.386124
compute loss for weight  1e-05  0 result 2.99668
 training batch 31 mu var00.386124
compute loss for weight  -1e-05  0 result 2.99668
 training batch 32 mu var00.386124
compute loss for weight  5e-06  0 result 2.99668
 training batch 33 mu var00.386124
compute loss for weight  -5e-06  0 result 2.99668
   --dy = -7.40149e-12 dy_ref = -2.77556e-16
Testing weight gradients   for    layer 2
weight gradient for layer 2

1x2 matrix is as follows

     |      0    |      1    |
-------------------------------
   0 |      -2.71       3.461 

weights for layer 2

1x2 matrix is as follows

     |      0    |      1    |
-------------------------------
   0 |    0.06989       1.786 

 training batch 34 mu var00.386124
compute loss for weight  0.0699048  0.0698948 result 2.99665
 training batch 35 mu var00.386124
compute loss for weight  0.0698848  0.0698948 result 2.99671
 training batch 36 mu var00.386124
compute loss for weight  0.0698998  0.0698948 result 2.99667
 training batch 37 mu var00.386124
compute loss for weight  0.0698898  0.0698948 result 2.99669
   --dy = -2.70956 dy_ref = -2.70956
 training batch 38 mu var00.386124
compute loss for weight  1.78638  1.78637 result 2.99671
 training batch 39 mu var00.386124
compute loss for weight  1.78636  1.78637 result 2.99664
 training batch 40 mu var00.386124
compute loss for weight  1.78638  1.78637 result 2.9967
 training batch 41 mu var00.386124
compute loss for weight  1.78637  1.78637 result 2.99666
   --dy = 3.46106 dy_ref = 3.46106
Testing weight gradients:      maximum relative error: [NON-XML-CHAR-0x1B][33m7.58966e-09[NON-XML-CHAR-0x1B][39m