Execution Time1.06s

Test: TMVA-DNN-RNN-Backpropagation (Passed)
Build: PR-4279-x86_64-fedora29-gcc8-opt (root-fedora29-3.cern.ch) on 2019-11-14 21:03:34

Test Timing: Passed
Processors1

Show Command Line
Display graphs:

Test output
Testing RNN backward pass
Testing Weight Backprop using RNN with batchsize = 2 input = 10 state = 1 time = 1	using a random input
Testing weight input gradients:      maximum error (relative): [NON-XML-CHAR-0x1B][32m3.67823e-10[NON-XML-CHAR-0x1B][39m
Testing weight state gradients:      maximum error (absolute): [NON-XML-CHAR-0x1B][32m0[NON-XML-CHAR-0x1B][39m
Testing bias gradients:      maximum error (relative): [NON-XML-CHAR-0x1B][32m9.29155e-12[NON-XML-CHAR-0x1B][39m
Testing Weight Backprop using RNN with batchsize = 2 input = 10 state = 1 time = 2	using a random input
Testing weight input gradients:      maximum error (relative): [NON-XML-CHAR-0x1B][32m1.5212e-10[NON-XML-CHAR-0x1B][39m
Testing weight state gradients:      maximum error (relative): [NON-XML-CHAR-0x1B][33m1.6081e-09[NON-XML-CHAR-0x1B][39m
Testing bias gradients:      maximum error (relative): [NON-XML-CHAR-0x1B][32m2.79163e-11[NON-XML-CHAR-0x1B][39m
Testing Weight Backprop using RNN with batchsize = 2 input = 10 state = 2 time = 1	using a random input
Testing weight input gradients:      maximum error (relative): [NON-XML-CHAR-0x1B][32m3.61551e-10[NON-XML-CHAR-0x1B][39m
Testing weight state gradients:      maximum error (absolute): [NON-XML-CHAR-0x1B][32m0[NON-XML-CHAR-0x1B][39m
Testing bias gradients:      maximum error (relative): [NON-XML-CHAR-0x1B][32m9.0407e-12[NON-XML-CHAR-0x1B][39m
Testing Weight Backprop using RNN with batchsize = 1 input = 5 state = 2 time = 2	using a random input
Testing weight input gradients:      maximum error (relative): [NON-XML-CHAR-0x1B][32m4.15989e-10[NON-XML-CHAR-0x1B][39m
Testing weight state gradients:      maximum error (relative): [NON-XML-CHAR-0x1B][32m2.15336e-10[NON-XML-CHAR-0x1B][39m
Testing bias gradients:      maximum error (relative): [NON-XML-CHAR-0x1B][33m1.44311e-08[NON-XML-CHAR-0x1B][39m
Testing Weight Backprop using RNN with batchsize = 2 input = 10 state = 3 time = 4	using a random input
Testing weight input gradients:      maximum error (relative): [NON-XML-CHAR-0x1B][33m1.56439e-07[NON-XML-CHAR-0x1B][39m
Testing weight state gradients:      maximum error (relative): [NON-XML-CHAR-0x1B][33m1.11185e-09[NON-XML-CHAR-0x1B][39m
Testing bias gradients:      maximum error (relative): [NON-XML-CHAR-0x1B][32m2.15573e-10[NON-XML-CHAR-0x1B][39m
Testing Weight Backprop using RNN with batchsize = 1 input = 5 state = 4 time = 3	with a fixed input
Testing weight input gradients:      maximum error (relative): [NON-XML-CHAR-0x1B][33m6.18077e-08[NON-XML-CHAR-0x1B][39m
Testing weight state gradients:      maximum error (relative): [NON-XML-CHAR-0x1B][33m8.12019e-09[NON-XML-CHAR-0x1B][39m
Testing bias gradients:      maximum error (relative): [NON-XML-CHAR-0x1B][33m1.36973e-09[NON-XML-CHAR-0x1B][39m
Testing Weight Backprop using RNN with batchsize = 32 input = 5 state = 10 time = 4	using a random input and a dense layer
Testing weight input gradients:      maximum error (relative): [NON-XML-CHAR-0x1B][33m1.22113e-06[NON-XML-CHAR-0x1B][39m
Testing weight state gradients:      maximum error (relative): [NON-XML-CHAR-0x1B][33m7.96599e-08[NON-XML-CHAR-0x1B][39m
Testing bias gradients:      maximum error (relative): [NON-XML-CHAR-0x1B][33m5.09956e-09[NON-XML-CHAR-0x1B][39m
Testing Weight Backprop using RNN with batchsize = 32 input = 5 state = 10 time = 4	using a random input and a dense layer and an extra RNN
Testing weight input gradients:      maximum error (relative): [NON-XML-CHAR-0x1B][33m8.59074e-08[NON-XML-CHAR-0x1B][39m
Testing weight state gradients:      maximum error (relative): [NON-XML-CHAR-0x1B][33m1.54566e-06[NON-XML-CHAR-0x1B][39m
Testing bias gradients:      maximum error (relative): [NON-XML-CHAR-0x1B][32m6.89407e-10[NON-XML-CHAR-0x1B][39m