1. Test Modules
  2. Differential Validation
    1. Feedback Validation
    2. Learning Validation
    3. Total Accuracy
    4. Frozen and Alive Status
  3. Results

Subreport: Logs for com.simiacryptus.ref.lang.ReferenceCountingBase

Test Modules

Using Seed 2339369466957845504

Differential Validation

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)));
Logging 
Inputs: [
[ [ 0.08 ], [ -0.128 ], [ -0.608 ] ],
[ [ 0.7 ], [ 0.496 ], [ 1.764 ] ]
]
Inputs Statistics: {meanExponent=-0.403119694464533, negative=2, min=-0.608, max=1.764, mean=0.38399999999999995, count=6, sum=2.304, positive=4, stdDev=0.747821725636086, zeros=0}
Output: [
[ [ 0.0031948963187562462 ], [ 0.008158717663047321 ], [ 0.17032645018387926 ] ],
[ [ 0.2206555615733703 ], [ 0.11625086786080474 ], [ 1.027731737681294 ] ]
]
Outputs Statistics: {meanExponent=-1.1552747301864812, negative=0, min=0.0031948963187562462, max=1.027731737681294, mean=0.25771970521352533, count=6, sum=1.5463182312811519, positive=6, stdDev=0.35331922740057087, zeros=0}

Feedback Validation

We validate the agreement between the implemented derivative of the inputs apply finite difference estimations:

SingleDerivativeTester.java:117 executed in 0.02 seconds (0.000 gc):

        return testFeedback(
            statistics,
            component.addRef(),
            RefUtil.addRef(inputPrototype),
            outputPrototype.addRef());
      },
      outputPrototype.addRef(),
      RefUtil.addRef(inputPrototype),
      component.addRef()));
Logging 
Feedback for input 0
Inputs Values: [
[ [ 0.08 ], [ -0.128 ], [ -0.608 ] ],
[ [ 0.7 ], [ 0.496 ], [ 1.764 ] ]
]
Value Statistics: {meanExponent=-0.403119694464533, negative=2, min=-0.608, max=1.764, mean=0.38399999999999995, count=6, sum=2.304, positive=4, stdDev=0.747821725636086, zeros=0}
Implemented Feedback: [ [ 0.07974522228289, 0.0, 0.0, 0.0, 0.0, 0.0 ], [ 0.0, 0.5734623443633283, 0.0, 0.0, 0.0, 0.0 ], [ 0.0, 0.0, -0.12696413546540486, 0.0, 0.0, 0.0 ], [ 0.0, 0.0, 0.0, 0.44434455934671724, 0.0, 0.0 ], [ 0.0, 0.0, 0.0, 0.0, -0.5195131665224453, 0.0 ], [ 0.0, 0.0, 0.0, 0.0, 0.0, 0.8699375598949443 ] ]
Implemented Statistics: {meanExponent=-0.48888414227858834, negative=2, min=-0.5195131665224453, max=0.8699375598949443, mean=0.036694788441667486, count=36, sum=1.3210123839000296, positive=4, stdDev=0.20595248631832885, zeros=30}
Measured Feedback: [ [ 0.07979474569985712, 0.0, 0.0, 0.0, 0.0, 0.0 ], [ 0.0, 0.5734898340747918, 0.0, 0.0, 0.0, 0.0 ], [ 0.0, 0.0, -0.12691533895603513, 0.0, 0.0, 0.0 ], [ 0.0, 0.0, 0.0, 0.4443805067300133, 0.0, 0.0 ], [ 0.0, 0.0, 0.0, 0.0, -0.5194819727272204, 0.0 ], [ 0.0, 0.0, 0.0, 0.0, 0.0, 0.8699435566983382 ] ]
Measured Statistics: {meanExponent=-0.48886155173714546, negative=2, min=-0.5194819727272204, max=0.8699435566983382, mean=0.03670031476443736, count=36, sum=1.321211331519745, positive=4, stdDev=0.20595399765109285, zeros=30}
Feedback Error: [ [ 4.952341696712326E-5, 0.0, 0.0, 0.0, 0.0, 0.0 ], [ 0.0, 2.7489711463490885E-5, 0.0, 0.0, 0.0, 0.0 ], [ 0.0, 0.0, 4.879650936973068E-5, 0.0, 0.0, 0.0 ], [ 0.0, 0.0, 0.0, 3.594738329604219E-5, 0.0, 0.0 ], [ 0.0, 0.0, 0.0, 0.0, 3.119379522487087E-5, 0.0 ], [ 0.0, 0.0, 0.0, 0.0, 0.0, 5.996803393970573E-6 ] ]
Error Statistics: {meanExponent=-4.558329191738562, negative=0, min=0.0, max=4.952341696712326E-5, mean=5.526322769867457E-6, count=36, sum=1.9894761971522845E-4, positive=6, stdDev=1.3734721316978445E-5, zeros=30}

Returns

    {
      "absoluteTol" : {
        "count" : 36,
        "sum" : 1.9894761971522845E-4,
        "min" : 0.0,
        "max" : 4.952341696712326E-5,
        "sumOfSquare" : 7.890581268425434E-9,
        "standardDeviation" : 1.3734721316978445E-5,
        "average" : 5.526322769867457E-6
      },
      "relativeTol" : {
        "count" : 6,
        "sum" : 6.005028974880725E-4,
        "min" : 3.4466742220367437E-6,
        "max" : 3.104138580025213E-4,
        "sumOfSquare" : 1.3642269831701872E-7,
        "standardDeviation" : 1.1278451184184794E-4,
        "average" : 1.0008381624801209E-4
      }
    }

Learning Validation

We validate the agreement between the implemented derivative of the internal weights apply finite difference estimations:

SingleDerivativeTester.java:133 executed in 0.01 seconds (0.000 gc):

        return testLearning(
            statistics,
            component.addRef(),
            RefUtil.addRef(inputPrototype),
            outputPrototype.addRef());
      },
      outputPrototype.addRef(),
      RefUtil.addRef(inputPrototype),
      component.addRef()));
Logging 
Learning Gradient for weight setByCoord 0
Weights: [ 1.0, 1.0 ]
Implemented Gradient: [ [ -0.996815278536125, -0.8192319205190405, 0.0, -0.8958559664248332, 0.0, -0.4931618820266124 ], [ 0.0, 0.0, -0.9919073083234755, 0.0, -0.8544624449382324, 0.0 ] ]
Implemented Statistics: {meanExponent=-0.08576444781405546, negative=6, min=-0.996815278536125, max=0.0, mean=-0.4209529000640266, count=12, sum=-5.051434800768319, positive=0, stdDev=0.43761838806115677, zeros=6}
Measured Gradient: [ [ -0.9967152900557932, -0.819136536715781, 0.0, -0.8957575468601653, 0.0, -0.4930939130143308 ], [ 0.0, 0.0, -0.9918073281202713, 0.0, -0.8543654779902266, 0.0 ] ]
Measured Statistics: {meanExponent=-0.08581357667606182, negative=6, min=-0.9967152900557932, max=0.0, mean=-0.4209063410630473, count=12, sum=-5.0508760927565675, positive=0, stdDev=0.4375714794704981, zeros=6}
Gradient Error: [ [ 9.998848033176966E-5, 9.538380325946338E-5, 0.0, 9.841956466793977E-5, 0.0, 6.796901228156571E-5 ], [ 0.0, 0.0, 9.998020320423073E-5, 0.0, 9.696694800587213E-5, 0.0 ] ]
Error Statistics: {meanExponent=-4.034774209458487, negative=0, min=0.0, max=9.998848033176966E-5, mean=4.6559000979236785E-5, count=12, sum=5.587080117508414E-4, positive=6, stdDev=4.72473161807217E-5, zeros=6}

Returns

    {
      "absoluteTol" : {
        "count" : 48,
        "sum" : 7.576556314660698E-4,
        "min" : 0.0,
        "max" : 9.998848033176966E-5,
        "sumOfSquare" : 6.069117477001333E-8,
        "standardDeviation" : 3.186297657949392E-5,
        "average" : 1.5784492322209787E-5
      },
      "relativeTol" : {
        "count" : 12,
        "sum" : 9.398730529870684E-4,
        "min" : 3.4466742220367437E-6,
        "max" : 3.104138580025213E-4,
        "sumOfSquare" : 1.5585510039168586E-7,
        "standardDeviation" : 8.278569425568536E-5,
        "average" : 7.832275441558904E-5
      }
    }

Total Accuracy

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));
Logging 
Finite-Difference Derivative Accuracy:
absoluteTol: 1.5784e-05 +- 3.1863e-05 [0.0000e+00 - 9.9988e-05] (48#)
relativeTol: 7.8323e-05 +- 8.2786e-05 [3.4467e-06 - 3.1041e-04] (12#)

Frozen and Alive Status

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());

Results

classdetailsresult
com.simiacryptus.mindseye.test.unit.SingleDerivativeTesterToleranceStatistics{absoluteTol=1.5784e-05 +- 3.1863e-05 [0.0000e+00 - 9.9988e-05] (48#), relativeTol=7.8323e-05 +- 8.2786e-05 [3.4467e-06 - 3.1041e-04] (12#)}OK
 
  {
    "result": "OK",
    "performance": {
      "execution_time": "0.163",
      "gc_time": "0.102"
    },
    "created_on": 1586739473886,
    "file_name": "derivativeTest",
    "report": {
      "simpleName": "Basic",
      "canonicalName": "com.simiacryptus.mindseye.layers.java.HyperbolicActivationLayerTest.Basic",
      "link": "https://github.com/SimiaCryptus/mindseye-java/tree/93db34cedee48c0202777a2b25deddf1dfaf5731/src/test/java/com/simiacryptus/mindseye/layers/java/HyperbolicActivationLayerTest.java",
      "javaDoc": ""
    },
    "archive": "s3://code.simiacrypt.us/tests/com/simiacryptus/mindseye/layers/java/HyperbolicActivationLayer/Basic/derivativeTest/202004135753",
    "id": "44c2d39b-392d-4e23-b373-39284a5fe8df",
    "report_type": "Components",
    "display_name": "Derivative Validation",
    "target": {
      "simpleName": "HyperbolicActivationLayer",
      "canonicalName": "com.simiacryptus.mindseye.layers.java.HyperbolicActivationLayer",
      "link": "https://github.com/SimiaCryptus/mindseye-java/tree/93db34cedee48c0202777a2b25deddf1dfaf5731/src/main/java/com/simiacryptus/mindseye/layers/java/HyperbolicActivationLayer.java",
      "javaDoc": ""
    }
  }