Subreport: Logs for com.simiacryptus.ref.lang.ReferenceCountingBase
Using Seed 2513584441001221120
SingleDerivativeTester.java:101 executed in 0.00 seconds (0.000 gc):
log.info(RefString.format("Inputs: %s", prettyPrint(inputPrototype)));
log.info(RefString.format("Inputs Statistics: %s", printStats(inputPrototype)));
log.info(RefString.format("Output: %s", outputPrototype.prettyPrint()));
assert outputPrototype != null;
log.info(RefString.format("Outputs Statistics: %s", outputPrototype.getScalarStatistics()));
},
outputPrototype.addRef(),
RefUtil.addRef(inputPrototype)));
Inputs: [
[ [ 0.08 ], [ -0.128 ] ],
[ [ 0.7 ], [ 0.496 ] ]
]
Inputs Statistics: {meanExponent=-0.6122800817139336, negative=1, min=-0.128, max=0.7, mean=0.287, count=4, sum=1.148, positive=3, stdDev=0.3276141022605712, zeros=0}
Output: [
[ [ 0.07999999821186066 ], [ 0.0 ] ],
[ [ 0.699999988079071 ], [ 0.4959999918937683 ] ]
]
Outputs Statistics: {meanExponent=-0.518776773568199, negative=0, min=0.0, max=0.699999988079071, mean=0.318999994546175, count=4, sum=1.2759999781847, positive=3, stdDev=0.28955655272389225, zeros=1}
We validate the agreement between the implemented derivative of the inputs apply finite difference estimations:
SingleDerivativeTester.java:117 executed in 0.06 seconds (0.000 gc):
return testFeedback(
statistics,
component.addRef(),
RefUtil.addRef(inputPrototype),
outputPrototype.addRef());
},
outputPrototype.addRef(),
RefUtil.addRef(inputPrototype),
component.addRef()));
Feedback for input 0
Inputs Values: [
[ [ 0.08 ], [ -0.128 ] ],
[ [ 0.7 ], [ 0.496 ] ]
]
Value Statistics: {meanExponent=-0.6122800817139336, negative=1, min=-0.128, max=0.7, mean=0.287, count=4, sum=1.148, positive=3, stdDev=0.3276141022605712, zeros=0}
Implemented Feedback: [ [ 1.0, 0.0, 0.0, 0.0 ], [ 0.0, 1.0, 0.0, 0.0 ], [ 0.0, 0.0, 0.0, 0.0 ], [ 0.0, 0.0, 0.0, 1.0 ] ]
Implemented Statistics: {meanExponent=0.0, negative=0, min=0.0, max=1.0, mean=0.1875, count=16, sum=3.0, positive=3, stdDev=0.3903123748998999, zeros=13}
Measured Feedback: [ [ 1.0000169277191162, 0.0, 0.0, 0.0 ], [ 0.0, 1.0001659393310547, 0.0, 0.0 ], [ 0.0, 0.0, 0.0, 0.0 ], [ 0.0, 0.0, 0.0, 1.0001659393310547 ] ]
Measured Statistics: {meanExponent=5.049088901553801E-5, negative=0, min=0.0, max=1.0001659393310547, mean=0.1875218003988266, count=16, sum=3.0003488063812256, positive=3, stdDev=0.39035775723396066, zeros=13}
Feedback Error: [ [ 1.6927719116210938E-5, 0.0, 0.0, 0.0 ], [ 0.0, 1.659393310546875E-4, 0.0, 0.0 ], [ 0.0, 0.0, 0.0, 0.0 ], [ 0.0, 0.0, 0.0, 1.659393310546875E-4 ] ]
Error Statistics: {meanExponent=-4.110500961960853, negative=0, min=0.0, max=1.659393310546875E-4, mean=2.180039882659912E-5, count=16, sum=3.4880638122558594E-4, positive=3, stdDev=5.4631808861831544E-5, zeros=13}
Returns
{
"absoluteTol" : {
"count" : 16,
"sum" : 3.4880638122558594E-4,
"min" : 0.0,
"max" : 1.659393310546875E-4,
"sumOfSquare" : 5.5358270856231684E-8,
"standardDeviation" : 5.4631808861831544E-5,
"average" : 2.180039882659912E-5
},
"relativeTol" : {
"count" : 3,
"sum" : 1.7438935218791108E-4,
"min" : 8.463787921793169E-6,
"max" : 8.296278213305896E-5,
"sumOfSquare" : 1.3837282144499906E-8,
"standardDeviation" : 3.5119162665575597E-5,
"average" : 5.8129784062637026E-5
}
}
We validate the agreement between the implemented derivative of the internal weights apply finite difference estimations:
SingleDerivativeTester.java:133 executed in 0.00 seconds (0.000 gc):
return testLearning(
statistics,
component.addRef(),
RefUtil.addRef(inputPrototype),
outputPrototype.addRef());
},
outputPrototype.addRef(),
RefUtil.addRef(inputPrototype),
component.addRef()));
Returns
{
"absoluteTol" : {
"count" : 16,
"sum" : 3.4880638122558594E-4,
"min" : 0.0,
"max" : 1.659393310546875E-4,
"sumOfSquare" : 5.5358270856231684E-8,
"standardDeviation" : 5.4631808861831544E-5,
"average" : 2.180039882659912E-5
},
"relativeTol" : {
"count" : 3,
"sum" : 1.7438935218791108E-4,
"min" : 8.463787921793169E-6,
"max" : 8.296278213305896E-5,
"sumOfSquare" : 1.3837282144499906E-8,
"standardDeviation" : 3.5119162665575597E-5,
"average" : 5.8129784062637026E-5
}
}
The overall agreement accuracy between the implemented derivative and the finite difference estimations:
SingleDerivativeTester.java:148 executed in 0.00 seconds (0.000 gc):
//log.info(String.format("Component: %s\nInputs: %s\noutput=%s", component, Arrays.toStream(inputPrototype), outputPrototype));
log.info(RefString.format("Finite-Difference Derivative Accuracy:"));
log.info(RefString.format("absoluteTol: %s", statistics.absoluteTol));
log.info(RefString.format("relativeTol: %s", statistics.relativeTol));
Finite-Difference Derivative Accuracy:
absoluteTol: 2.1800e-05 +- 5.4632e-05 [0.0000e+00 - 1.6594e-04] (16#)
relativeTol: 5.8130e-05 +- 3.5119e-05 [8.4638e-06 - 8.2963e-05] (3#)
SingleDerivativeTester.java:156 executed in 0.01 seconds (0.000 gc):
testFrozen(component.addRef(), RefUtil.addRef(inputPrototype));
testUnFrozen(component.addRef(), RefUtil.addRef(inputPrototype));
LayerTests.java:425 executed in 0.00 seconds (0.000 gc):
throwException(exceptions.addRef());
| class | details | result |
|---|---|---|
| com.simiacryptus.mindseye.test.unit.SingleDerivativeTester | ToleranceStatistics{absoluteTol=2.1800e-05 +- 5.4632e-05 [0.0000e+00 - 1.6594e-04] (16#), relativeTol=5.8130e-05 +- 3.5119e-05 [8.4638e-06 - 8.2963e-05] (3#)} | OK |
{
"result": "OK",
"performance": {
"execution_time": "0.255",
"gc_time": "0.143"
},
"created_on": 1586740290439,
"file_name": "derivativeTest",
"report": {
"simpleName": "ReLu_Float",
"canonicalName": "com.simiacryptus.mindseye.layers.cudnn.ActivationLayerTest.ReLu_Float",
"link": "https://github.com/SimiaCryptus/mindseye-cudnn/tree/59d5b3318556370acb2d83ee6ec123ce0fc6974f/src/test/java/com/simiacryptus/mindseye/layers/cudnn/ActivationLayerTest.java",
"javaDoc": ""
},
"archive": "s3://code.simiacrypt.us/tests/com/simiacryptus/mindseye/layers/cudnn/ActivationLayer/ReLu_Float/derivativeTest/202004131130",
"id": "b52eb755-20cf-4645-91fb-0196ee364998",
"report_type": "Components",
"display_name": "Derivative Validation",
"target": {
"simpleName": "ActivationLayer",
"canonicalName": "com.simiacryptus.mindseye.layers.cudnn.ActivationLayer",
"link": "https://github.com/SimiaCryptus/mindseye-cudnn/tree/59d5b3318556370acb2d83ee6ec123ce0fc6974f/src/main/java/com/simiacryptus/mindseye/layers/cudnn/ActivationLayer.java",
"javaDoc": ""
}
}