Execution Time0.06s

Test: TMVA-DNN-BatchNormalization (Passed)
Build: master-x86_64-ubuntu16-gcc54 (sft-ubuntu-1604-4) on 2019-11-15 01:17:05
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 var00.105192
output DL 

10x2 matrix is as follows

     |      0    |      1    |
-------------------------------
   0 |     0.2625     -0.4613 
   1 |    -0.2904     -0.6922 
   2 |      1.034      0.9247 
   3 |    -0.2416        2.06 
   4 |     0.2377      0.2025 
   5 |     0.4412     -0.4531 
   6 |     -1.457     -0.3507 
   7 |      1.122      -1.431 
   8 |    0.09778      -2.055 
   9 |    -0.1553      0.3601 

output BN 
output DL feature 0 mean 0.105192	output DL std 0.733151
output DL feature 1 mean -0.189698	output DL std 1.16434
output of BN 

10x2 matrix is as follows

     |      0    |      1    |
-------------------------------
   0 |     0.2261     -0.2459 
   1 |    -0.5687     -0.4549 
   2 |      1.336       1.009 
   3 |    -0.4985       2.036 
   4 |     0.1905      0.3551 
   5 |      0.483     -0.2385 
   6 |     -2.246     -0.1458 
   7 |      1.462      -1.124 
   8 |   -0.01066      -1.689 
   9 |    -0.3744      0.4977 

output BN feature 0 mean 4.44089e-17	output BN std 1.05398
output BN feature 1 mean -3.33067e-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 |      0.429      0.7029       0.385      0.8364 
   1 |    0.08303    -0.02088     -0.1976      0.7471 

weights for layer 0

2x4 matrix is as follows

     |      0    |      1    |      2    |      3    |
---------------------------------------------------------
   0 |    0.09646      0.4416      0.4342     -0.6202 
   1 |    -0.9115     -0.9348     0.03522     -0.5065 

 training batch 2 mu var00.105195
compute loss for weight  0.0964669  0.0964569 result 1.10294
 training batch 3 mu var00.105192
compute loss for weight  0.0964469  0.0964569 result 1.10294
 training batch 4 mu var00.105193
compute loss for weight  0.0964619  0.0964569 result 1.10294
 training batch 5 mu var00.105192
compute loss for weight  0.0964519  0.0964569 result 1.10294
   --dy = 0.428975 dy_ref = 0.428975
 training batch 6 mu var00.105192
compute loss for weight  0.441614  0.441604 result 1.10295
 training batch 7 mu var00.105192
compute loss for weight  0.441594  0.441604 result 1.10293
 training batch 8 mu var00.105192
compute loss for weight  0.441609  0.441604 result 1.10294
 training batch 9 mu var00.105192
compute loss for weight  0.441599  0.441604 result 1.10294
   --dy = 0.702915 dy_ref = 0.702915
 training batch 10 mu var00.105193
compute loss for weight  0.434214  0.434204 result 1.10294
 training batch 11 mu var00.105192
compute loss for weight  0.434194  0.434204 result 1.10294
 training batch 12 mu var00.105192
compute loss for weight  0.434209  0.434204 result 1.10294
 training batch 13 mu var00.105192
compute loss for weight  0.434199  0.434204 result 1.10294
   --dy = 0.384955 dy_ref = 0.384955
 training batch 14 mu var00.105192
compute loss for weight  -0.620233  -0.620243 result 1.10295
 training batch 15 mu var00.105192
compute loss for weight  -0.620253  -0.620243 result 1.10293
 training batch 16 mu var00.105192
compute loss for weight  -0.620238  -0.620243 result 1.10294
 training batch 17 mu var00.105192
compute loss for weight  -0.620248  -0.620243 result 1.10294
   --dy = 0.836433 dy_ref = 0.836433
Testing weight gradients   for    layer 1
weight gradient for layer 1

1x2 matrix is as follows

     |      0    |      1    |
-------------------------------
   0 |     0.7015       1.504 

weights for layer 1

1x2 matrix is as follows

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

 training batch 18 mu var00.105192
compute loss for weight  1.00001  1 result 1.10295
 training batch 19 mu var00.105192
compute loss for weight  0.99999  1 result 1.10293
 training batch 20 mu var00.105192
compute loss for weight  1.00001  1 result 1.10294
 training batch 21 mu var00.105192
compute loss for weight  0.999995  1 result 1.10294
   --dy = 0.701486 dy_ref = 0.701486
 training batch 22 mu var00.105192
compute loss for weight  1.00001  1 result 1.10296
 training batch 23 mu var00.105192
compute loss for weight  0.99999  1 result 1.10293
 training batch 24 mu var00.105192
compute loss for weight  1.00001  1 result 1.10295
 training batch 25 mu var00.105192
compute loss for weight  0.999995  1 result 1.10293
   --dy = 1.50439 dy_ref = 1.50439
Testing weight gradients   for    layer 1
weight gradient for layer 1

1x2 matrix is as follows

     |      0    |      1    |
-------------------------------
   0 |          0  -2.776e-17 

weights for layer 1

1x2 matrix is as follows

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

 training batch 26 mu var00.105192
compute loss for weight  1e-05  0 result 1.10294
 training batch 27 mu var00.105192
compute loss for weight  -1e-05  0 result 1.10294
 training batch 28 mu var00.105192
compute loss for weight  5e-06  0 result 1.10294
 training batch 29 mu var00.105192
compute loss for weight  -5e-06  0 result 1.10294
   --dy = 0 dy_ref = 0
 training batch 30 mu var00.105192
compute loss for weight  1e-05  0 result 1.10294
 training batch 31 mu var00.105192
compute loss for weight  -1e-05  0 result 1.10294
 training batch 32 mu var00.105192
compute loss for weight  5e-06  0 result 1.10294
 training batch 33 mu var00.105192
compute loss for weight  -5e-06  0 result 1.10294
   --dy = -3.70074e-12 dy_ref = -2.77556e-17
Testing weight gradients   for    layer 2
weight gradient for layer 2

1x2 matrix is as follows

     |      0    |      1    |
-------------------------------
   0 |      1.271       -1.79 

weights for layer 2

1x2 matrix is as follows

     |      0    |      1    |
-------------------------------
   0 |     0.5519     -0.8403 

 training batch 34 mu var00.105192
compute loss for weight  0.551918  0.551908 result 1.10295
 training batch 35 mu var00.105192
compute loss for weight  0.551898  0.551908 result 1.10293
 training batch 36 mu var00.105192
compute loss for weight  0.551913  0.551908 result 1.10295
 training batch 37 mu var00.105192
compute loss for weight  0.551903  0.551908 result 1.10293
   --dy = 1.27102 dy_ref = 1.27102
 training batch 38 mu var00.105192
compute loss for weight  -0.840259  -0.840269 result 1.10292
 training batch 39 mu var00.105192
compute loss for weight  -0.840279  -0.840269 result 1.10296
 training batch 40 mu var00.105192
compute loss for weight  -0.840264  -0.840269 result 1.10293
 training batch 41 mu var00.105192
compute loss for weight  -0.840274  -0.840269 result 1.10295
   --dy = -1.79037 dy_ref = -1.79037
Testing weight gradients:      maximum relative error: [NON-XML-CHAR-0x1B][32m4.96095e-11[NON-XML-CHAR-0x1B][39m