Execution Time0.20s

Test: TMVA-DNN-BatchNormalization (Passed)
Build: master-x86_64-centos7-gcc62-opt-no-rt-cxxmodules (olsnba08.cern.ch) on 2019-11-12 16:51:37
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 var00.527281
output DL 

10x2 matrix is as follows

     |      0    |      1    |
-------------------------------
   0 |     0.8058       0.421 
   1 |     0.9955     -0.4272 
   2 |    -0.2976     -0.2408 
   3 |      1.126      -2.593 
   4 |      0.852     -0.2126 
   5 |      1.997     0.09185 
   6 |     0.4492        1.13 
   7 |     -1.522      0.7014 
   8 |     0.7575       2.318 
   9 |     0.1095     -0.4125 

output BN 
output DL feature 0 mean 0.527281	output DL std 0.944922
output DL feature 1 mean 0.0775231	output DL std 1.26689
output of BN 

10x2 matrix is as follows

     |      0    |      1    |
-------------------------------
   0 |     0.3107      0.2858 
   1 |     0.5223       -0.42 
   2 |    -0.9201     -0.2648 
   3 |     0.6679      -2.222 
   4 |     0.3622     -0.2414 
   5 |       1.64     0.01192 
   6 |   -0.08713      0.8753 
   7 |     -2.286      0.5191 
   8 |     0.2568       1.864 
   9 |     -0.466     -0.4077 

output BN feature 0 mean -2.22045e-17	output BN std 1.05403
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 |    -0.3674      -0.273     -0.5374     -0.7233 
   1 |    -0.5652      0.2894     -0.3658     -0.4694 

weights for layer 0

2x4 matrix is as follows

     |      0    |      1    |      2    |      3    |
---------------------------------------------------------
   0 |     0.7971     -0.6449     -0.3428     0.09318 
   1 |     0.2324      0.5552      0.5825      0.8374 

 training batch 2 mu var00.527284
compute loss for weight  0.797113  0.797103 result 1.16245
 training batch 3 mu var00.527281
compute loss for weight  0.797093  0.797103 result 1.16245
 training batch 4 mu var00.527282
compute loss for weight  0.797108  0.797103 result 1.16245
 training batch 5 mu var00.527281
compute loss for weight  0.797098  0.797103 result 1.16245
   --dy = -0.367353 dy_ref = -0.367353
 training batch 6 mu var00.527281
compute loss for weight  -0.644847  -0.644857 result 1.16245
 training batch 7 mu var00.527281
compute loss for weight  -0.644867  -0.644857 result 1.16245
 training batch 8 mu var00.527281
compute loss for weight  -0.644852  -0.644857 result 1.16245
 training batch 9 mu var00.527281
compute loss for weight  -0.644862  -0.644857 result 1.16245
   --dy = -0.273045 dy_ref = -0.273045
 training batch 10 mu var00.527282
compute loss for weight  -0.342833  -0.342843 result 1.16245
 training batch 11 mu var00.527281
compute loss for weight  -0.342853  -0.342843 result 1.16246
 training batch 12 mu var00.527281
compute loss for weight  -0.342838  -0.342843 result 1.16245
 training batch 13 mu var00.527281
compute loss for weight  -0.342848  -0.342843 result 1.16245
   --dy = -0.537385 dy_ref = -0.537385
 training batch 14 mu var00.527281
compute loss for weight  0.0931884  0.0931784 result 1.16244
 training batch 15 mu var00.527281
compute loss for weight  0.0931684  0.0931784 result 1.16246
 training batch 16 mu var00.527281
compute loss for weight  0.0931834  0.0931784 result 1.16245
 training batch 17 mu var00.527281
compute loss for weight  0.0931734  0.0931784 result 1.16245
   --dy = -0.723336 dy_ref = -0.723336
Testing weight gradients   for    layer 1
weight gradient for layer 1

1x2 matrix is as follows

     |      0    |      1    |
-------------------------------
   0 |     0.7736       1.551 

weights for layer 1

1x2 matrix is as follows

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

 training batch 18 mu var00.527281
compute loss for weight  1.00001  1 result 1.16246
 training batch 19 mu var00.527281
compute loss for weight  0.99999  1 result 1.16244
 training batch 20 mu var00.527281
compute loss for weight  1.00001  1 result 1.16245
 training batch 21 mu var00.527281
compute loss for weight  0.999995  1 result 1.16245
   --dy = 0.773618 dy_ref = 0.773618
 training batch 22 mu var00.527281
compute loss for weight  1.00001  1 result 1.16247
 training batch 23 mu var00.527281
compute loss for weight  0.99999  1 result 1.16244
 training batch 24 mu var00.527281
compute loss for weight  1.00001  1 result 1.16246
 training batch 25 mu var00.527281
compute loss for weight  0.999995  1 result 1.16244
   --dy = 1.55128 dy_ref = 1.55128
Testing weight gradients   for    layer 1
weight gradient for layer 1

1x2 matrix is as follows

     |      0    |      1    |
-------------------------------
   0 |  4.857e-17  -9.194e-17 

weights for layer 1

1x2 matrix is as follows

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

 training batch 26 mu var00.527281
compute loss for weight  1e-05  0 result 1.16245
 training batch 27 mu var00.527281
compute loss for weight  -1e-05  0 result 1.16245
 training batch 28 mu var00.527281
compute loss for weight  5e-06  0 result 1.16245
 training batch 29 mu var00.527281
compute loss for weight  -5e-06  0 result 1.16245
   --dy = 5.55112e-11 dy_ref = 4.85723e-17
 training batch 30 mu var00.527281
compute loss for weight  1e-05  0 result 1.16245
 training batch 31 mu var00.527281
compute loss for weight  -1e-05  0 result 1.16245
 training batch 32 mu var00.527281
compute loss for weight  5e-06  0 result 1.16245
 training batch 33 mu var00.527281
compute loss for weight  -5e-06  0 result 1.16245
   --dy = -2.96059e-11 dy_ref = -9.19403e-17
Testing weight gradients   for    layer 2
weight gradient for layer 2

1x2 matrix is as follows

     |      0    |      1    |
-------------------------------
   0 |      1.423      -1.875 

weights for layer 2

1x2 matrix is as follows

     |      0    |      1    |
-------------------------------
   0 |     0.5435     -0.8272 

 training batch 34 mu var00.527281
compute loss for weight  0.543551  0.543541 result 1.16247
 training batch 35 mu var00.527281
compute loss for weight  0.543531  0.543541 result 1.16244
 training batch 36 mu var00.527281
compute loss for weight  0.543546  0.543541 result 1.16246
 training batch 37 mu var00.527281
compute loss for weight  0.543536  0.543541 result 1.16244
   --dy = 1.42329 dy_ref = 1.42329
 training batch 38 mu var00.527281
compute loss for weight  -0.827203  -0.827213 result 1.16243
 training batch 39 mu var00.527281
compute loss for weight  -0.827223  -0.827213 result 1.16247
 training batch 40 mu var00.527281
compute loss for weight  -0.827208  -0.827213 result 1.16244
 training batch 41 mu var00.527281
compute loss for weight  -0.827218  -0.827213 result 1.16246
   --dy = -1.87532 dy_ref = -1.87532
Testing weight gradients:      maximum relative error: [NON-XML-CHAR-0x1B][32m5.62131e-11[NON-XML-CHAR-0x1B][39m