Execution Time0.05s

Test: TMVA-DNN-BatchNormalization (Passed)
Build: master-x86_64-ubuntu16-gcc54 (sft-ubuntu-1604-4) on 2019-11-14 01:16:01
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 

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 var00.242348
output DL 

10x2 matrix is as follows

     |      0    |      1    |
-------------------------------
   0 |     0.6165     -0.4796 
   1 |    -0.2532      0.6391 
   2 |     0.4434      -1.063 
   3 |     -1.361       1.001 
   4 |     0.2562     -0.2875 
   5 |      0.834     -0.6201 
   6 |    0.03709       0.611 
   7 |     0.5199     -0.6964 
   8 |       1.63      -1.001 
   9 |    -0.2992      0.2639 

output BN 
output DL feature 0 mean 0.242348	output DL std 0.794948
output DL feature 1 mean -0.163196	output DL std 0.737993
output of BN 

10x2 matrix is as follows

     |      0    |      1    |
-------------------------------
   0 |     0.4961     -0.4519 
   1 |    -0.6571       1.146 
   2 |     0.2665      -1.285 
   3 |     -2.126       1.663 
   4 |    0.01837     -0.1775 
   5 |     0.7844     -0.6525 
   6 |    -0.2721       1.106 
   7 |      0.368     -0.7615 
   8 |       1.84      -1.196 
   9 |     -0.718      0.6099 

output BN feature 0 mean -2.22045e-17	output BN std 1.054
output BN feature 1 mean -2.22045e-17	output BN std 1.05399
Testing weight gradients   for    layer 0
weight gradient for layer 0

2x4 matrix is as follows

     |      0    |      1    |      2    |      3    |
---------------------------------------------------------
   0 |    0.07178    -0.04072     -0.1104      0.2649 
   1 |     0.6035      0.4874       1.247      0.3178 

weights for layer 0

2x4 matrix is as follows

     |      0    |      1    |      2    |      3    |
---------------------------------------------------------
   0 |     0.3383      0.3407      0.5532      0.1919 
   1 |   -0.03582     -0.3153     -0.7365      0.1704 

 training batch 2 mu var00.242351
compute loss for weight  0.338286  0.338276 result 0.350013
 training batch 3 mu var00.242348
compute loss for weight  0.338266  0.338276 result 0.350012
 training batch 4 mu var00.242349
compute loss for weight  0.338281  0.338276 result 0.350013
 training batch 5 mu var00.242348
compute loss for weight  0.338271  0.338276 result 0.350012
   --dy = 0.0717824 dy_ref = 0.0717824
 training batch 6 mu var00.242348
compute loss for weight  0.340735  0.340725 result 0.350012
 training batch 7 mu var00.242348
compute loss for weight  0.340715  0.340725 result 0.350013
 training batch 8 mu var00.242348
compute loss for weight  0.34073  0.340725 result 0.350012
 training batch 9 mu var00.242348
compute loss for weight  0.34072  0.340725 result 0.350013
   --dy = -0.0407181 dy_ref = -0.0407181
 training batch 10 mu var00.242348
compute loss for weight  0.553234  0.553224 result 0.350011
 training batch 11 mu var00.242348
compute loss for weight  0.553214  0.553224 result 0.350014
 training batch 12 mu var00.242348
compute loss for weight  0.553229  0.553224 result 0.350012
 training batch 13 mu var00.242348
compute loss for weight  0.553219  0.553224 result 0.350013
   --dy = -0.110391 dy_ref = -0.110391
 training batch 14 mu var00.242348
compute loss for weight  0.191871  0.191861 result 0.350015
 training batch 15 mu var00.242348
compute loss for weight  0.191851  0.191861 result 0.35001
 training batch 16 mu var00.242348
compute loss for weight  0.191866  0.191861 result 0.350014
 training batch 17 mu var00.242348
compute loss for weight  0.191856  0.191861 result 0.350011
   --dy = 0.264944 dy_ref = 0.264944
Testing weight gradients   for    layer 1
weight gradient for layer 1

1x2 matrix is as follows

     |      0    |      1    |
-------------------------------
   0 |     0.9652     -0.2652 

weights for layer 1

1x2 matrix is as follows

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

 training batch 18 mu var00.242348
compute loss for weight  1.00001  1 result 0.350022
 training batch 19 mu var00.242348
compute loss for weight  0.99999  1 result 0.350003
 training batch 20 mu var00.242348
compute loss for weight  1.00001  1 result 0.350017
 training batch 21 mu var00.242348
compute loss for weight  0.999995  1 result 0.350008
   --dy = 0.965245 dy_ref = 0.965245
 training batch 22 mu var00.242348
compute loss for weight  1.00001  1 result 0.35001
 training batch 23 mu var00.242348
compute loss for weight  0.99999  1 result 0.350015
 training batch 24 mu var00.242348
compute loss for weight  1.00001  1 result 0.350011
 training batch 25 mu var00.242348
compute loss for weight  0.999995  1 result 0.350014
   --dy = -0.26522 dy_ref = -0.26522
Testing weight gradients   for    layer 1
weight gradient for layer 1

1x2 matrix is as follows

     |      0    |      1    |
-------------------------------
   0 |  4.163e-17   1.735e-17 

weights for layer 1

1x2 matrix is as follows

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

 training batch 26 mu var00.242348
compute loss for weight  1e-05  0 result 0.350012
 training batch 27 mu var00.242348
compute loss for weight  -1e-05  0 result 0.350012
 training batch 28 mu var00.242348
compute loss for weight  5e-06  0 result 0.350012
 training batch 29 mu var00.242348
compute loss for weight  -5e-06  0 result 0.350012
   --dy = 1.57282e-11 dy_ref = 4.16334e-17
 training batch 30 mu var00.242348
compute loss for weight  1e-05  0 result 0.350012
 training batch 31 mu var00.242348
compute loss for weight  -1e-05  0 result 0.350012
 training batch 32 mu var00.242348
compute loss for weight  5e-06  0 result 0.350012
 training batch 33 mu var00.242348
compute loss for weight  -5e-06  0 result 0.350012
   --dy = 6.4763e-12 dy_ref = 1.73472e-17
Testing weight gradients   for    layer 2
weight gradient for layer 2

1x2 matrix is as follows

     |      0    |      1    |
-------------------------------
   0 |     -1.134      0.8011 

weights for layer 2

1x2 matrix is as follows

     |      0    |      1    |
-------------------------------
   0 |    -0.8514     -0.3311 

 training batch 34 mu var00.242348
compute loss for weight  -0.851429  -0.851439 result 0.350001
 training batch 35 mu var00.242348
compute loss for weight  -0.851449  -0.851439 result 0.350024
 training batch 36 mu var00.242348
compute loss for weight  -0.851434  -0.851439 result 0.350007
 training batch 37 mu var00.242348
compute loss for weight  -0.851444  -0.851439 result 0.350018
   --dy = -1.13366 dy_ref = -1.13366
 training batch 38 mu var00.242348
compute loss for weight  -0.331065  -0.331075 result 0.35002
 training batch 39 mu var00.242348
compute loss for weight  -0.331085  -0.331075 result 0.350004
 training batch 40 mu var00.242348
compute loss for weight  -0.33107  -0.331075 result 0.350016
 training batch 41 mu var00.242348
compute loss for weight  -0.33108  -0.331075 result 0.350008
   --dy = 0.801089 dy_ref = 0.801089
Testing weight gradients:      maximum relative error: [NON-XML-CHAR-0x1B][32m2.48386e-10[NON-XML-CHAR-0x1B][39m