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 3632965142976038912

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, 1.108, 1.612 ], [ -0.608, -1.616, -0.556 ], [ 1.208, 1.032, 1.552 ], [ -0.852, 1.556, -1.572 ] ],
[ [ 0.7, 0.028, 1.64 ], [ 1.764, 1.356, -1.476 ], [ -1.72, 0.3, 1.876 ], [ 1.912, 0.788, -1.516 ] ],
[ [ -0.128, -0.712, 0.392 ], [ 0.048, 1.512, 1.704 ], [ -1.028, 0.636, -0.408 ], [ -1.688, -0.768, -0.636 ] ],
[ [ 0.496, 1.048, 0.092 ], [ 1.524, 1.556, -1.228 ], [ -0.384, -0.176, -0.384 ], [ -0.804, -0.068, -1.492 ] ]
],
[
[ [ -0.804, 0.148, -0.032 ] ]
]
Inputs Statistics: {meanExponent=-0.15927984553930932, negative=22, min=-1.72, max=1.912, mean=0.1604166666666667, count=48, sum=7.700000000000002, positive=26, stdDev=1.1368718015042663, zeros=0},
{meanExponent=-0.8064440858455617, negative=2, min=-0.804, max=0.148, mean=-0.22933333333333336, count=3, sum=-0.6880000000000001, positive=1, stdDev=0.41294174999494654, zeros=0}
Output: [
[ [ -0.7240000367164612, 1.25600004196167, 1.5800000429153442 ], [ -1.4119999408721924, -1.468000054359436, -0.5879999995231628 ], [ 0.403999924659729, 1.1799999475479126, 1.5200001001358032 ], [ -1.656000018119812, 1.7039999961853027, -1.6039999723434448 ] ],
[ [ -0.1040000319480896, 0.17599999904632568, 1.6080000400543213 ], [ 0.9600000381469727, 1.503999948501587, -1.5079998970031738 ], [ -2.5240001678466797, 0.4480000138282776, 1.8440001010894775 ], [ 1.1079999208450317, 0.9359999895095825, -1.5479999780654907 ] ],
[ [ -0.9320000410079956, -0.5640000104904175, 0.35999998450279236 ], [ -0.7560000419616699, 1.659999966621399, 1.6720000505447388 ], [ -1.8320000171661377, 0.7839999794960022, -0.4399999976158142 ], [ -2.492000102996826, -0.6200000047683716, -0.6679999828338623 ] ],
[ [ -0.30800002813339233, 1.1959999799728394, 0.05999999865889549 ], [ 0.7200000286102295, 1.7039999961853027, -1.2599999904632568 ], [ -1.187999963760376, -0.02799999713897705, -0.41600000858306885 ], [ -1.6080000400543213, 0.07999999821186066, -1.5239999294281006 ] ]
]
Outputs Statistics: {meanExponent=-0.09435540221628486, negative=25, min=-2.5240001678466797, max=1.8440001010894775, mean=-0.06891667012435694, count=48, sum=-3.3080001659691334, positive=23, stdDev=1.2515721550971928, zeros=0}

Feedback Validation

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

SingleDerivativeTester.java:117 executed in 0.57 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, 1.108, 1.612 ], [ -0.608, -1.616, -0.556 ], [ 1.208, 1.032, 1.552 ], [ -0.852, 1.556, -1.572 ] ],
[ [ 0.7, 0.028, 1.64 ], [ 1.764, 1.356, -1.476 ], [ -1.72, 0.3, 1.876 ], [ 1.912, 0.788, -1.516 ] ],
[ [ -0.128, -0.712, 0.392 ], [ 0.048, 1.512, 1.704 ], [ -1.028, 0.636, -0.408 ], [ -1.688, -0.768, -0.636 ] ],
[ [ 0.496, 1.048, 0.092 ], [ 1.524, 1.556, -1.228 ], [ -0.384, -0.176, -0.384 ], [ -0.804, -0.068, -1.492 ] ]
]
Value Statistics: {meanExponent=-0.15927984553930932, negative=22, min=-1.72, max=1.912, mean=0.1604166666666667, count=48, sum=7.700000000000002, positive=26, stdDev=1.1368718015042663, zeros=0}
Implemented Feedback: [ [ 1.0, 0.0, 0.0, 0.0, 0.0, 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, 1.0, 0.0, 0.0, 0.0, 0.0, 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, 1.0, 0.0, 0.0, 0.0, ... ], [ 0.0, 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, 1.0, 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=0.020833333333333332, count=2304, sum=48.0, positive=48, stdDev=0.1428261375083551, zeros=2256}
Measured Feedback: [ [ 0.9999871253967285, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, ... ], [ 0.0, 0.9999871253967285, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, ... ], [ 0.0, 0.0, 1.000046730041504, 0.0, 0.0, 0.0, 0.0, 0.0, ... ], [ 0.0, 0.0, 0.0, 1.0000169277191162, 0.0, 0.0, 0.0, 0.0, ... ], [ 0.0, 0.0, 0.0, 0.0, 0.9999275207519531, 0.0, 0.0, 0.0, ... ], [ 0.0, 0.0, 0.0, 0.0, 0.0, 0.9999275207519531, 0.0, 0.0, ... ], [ 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 1.000046730041504, 0.0, ... ], [ 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.9999275207519531, ... ], ... ]
Measured Statistics: {meanExponent=3.306352073519224E-6, negative=0, min=0.0, max=1.0001659393310547, mean=0.020833491968611877, count=2304, sum=48.00036549568176, positive=48, stdDev=0.14282722524773514, zeros=2256}
Feedback Error: [ [ -1.2874603271484375E-5, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, ... ], [ 0.0, -1.2874603271484375E-5, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, ... ], [ 0.0, 0.0, 4.673004150390625E-5, 0.0, 0.0, 0.0, 0.0, 0.0, ... ], [ 0.0, 0.0, 0.0, 1.6927719116210938E-5, 0.0, 0.0, 0.0, 0.0, ... ], [ 0.0, 0.0, 0.0, 0.0, -7.2479248046875E-5, 0.0, 0.0, 0.0, ... ], [ 0.0, 0.0, 0.0, 0.0, 0.0, -7.2479248046875E-5, 0.0, 0.0, ... ], [ 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 4.673004150390625E-5, 0.0, ... ], [ 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, -7.2479248046875E-5, ... ], ... ]
Error Statistics: {meanExponent=-4.525332463740811, negative=19, min=-7.2479248046875E-5, max=1.659393310546875E-4, mean=1.5863527854283652E-7, count=2304, sum=3.654956817626953E-4, positive=29, stdDev=7.460422102476305E-6, zeros=2256}
Feedback for input 1
Inputs Values: [
[ [ -0.804, 0.148, -0.032 ] ]
]
Value Statistics: {meanExponent=-0.8064440858455617, negative=2, min=-0.804, max=0.148, mean=-0.22933333333333336, count=3, sum=-0.6880000000000001, positive=1, stdDev=0.41294174999494654, zeros=0}
Implemented Feedback: [ [ 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, ... ], [ 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, ... ], [ 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, ... ] ]
Implemented Statistics: {meanExponent=0.0, negative=0, min=0.0, max=1.0, mean=0.3333333333333333, count=144, sum=48.0, positive=48, stdDev=0.4714045207910317, zeros=96}
Measured Feedback: [ [ 1.000046730041504, 1.000046730041504, 1.000046730041504, 1.000046730041504, 0.9999275207519531, 1.000046730041504, 1.000046730041504, 1.000046730041504, ... ], [ 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, ... ], [ 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, ... ] ]
Measured Statistics: {meanExponent=3.3061913634620996E-6, negative=0, min=0.0, max=1.0001659393310547, mean=0.33333587149779004, count=144, sum=48.00036549568176, positive=48, stdDev=0.47140811148377665, zeros=96}
Feedback Error: [ [ 4.673004150390625E-5, 4.673004150390625E-5, 4.673004150390625E-5, 4.673004150390625E-5, -7.2479248046875E-5, 4.673004150390625E-5, 4.673004150390625E-5, 4.673004150390625E-5, ... ], [ 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, ... ], [ 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, ... ] ]
Error Statistics: {meanExponent=-4.482596780077778, negative=19, min=-7.2479248046875E-5, max=1.659393310546875E-4, mean=2.5381644566853843E-6, count=144, sum=3.654956817626953E-4, positive=29, stdDev=3.3633360796458385E-5, zeros=96}

Returns

    {
      "absoluteTol" : {
        "count" : 2448,
        "sum" : 0.004332065582275391,
        "min" : 0.0,
        "max" : 1.659393310546875E-4,
        "sumOfSquare" : 2.921146915468853E-7,
        "standardDeviation" : 1.0779438303474584E-5,
        "average" : 1.7696346332824308E-6
      },
      "relativeTol" : {
        "count" : 96,
        "sum" : 0.0021660188774547764,
        "min" : 1.0132779344484278E-6,
        "max" : 8.296278213305896E-5,
        "sumOfSquare" : 7.302629592255874E-8,
        "standardDeviation" : 1.5862386417277843E-5,
        "average" : 2.256269664015392E-5
      }
    }

Learning Validation

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" : 2448,
        "sum" : 0.004332065582275391,
        "min" : 0.0,
        "max" : 1.659393310546875E-4,
        "sumOfSquare" : 2.921146915468853E-7,
        "standardDeviation" : 1.0779438303474584E-5,
        "average" : 1.7696346332824308E-6
      },
      "relativeTol" : {
        "count" : 96,
        "sum" : 0.0021660188774547764,
        "min" : 1.0132779344484278E-6,
        "max" : 8.296278213305896E-5,
        "sumOfSquare" : 7.302629592255874E-8,
        "standardDeviation" : 1.5862386417277843E-5,
        "average" : 2.256269664015392E-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: 1.7696e-06 +- 1.0779e-05 [0.0000e+00 - 1.6594e-04] (2448#)
relativeTol: 2.2563e-05 +- 1.5862e-05 [1.0133e-06 - 8.2963e-05] (96#)

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=1.7696e-06 +- 1.0779e-05 [0.0000e+00 - 1.6594e-04] (2448#), relativeTol=2.2563e-05 +- 1.5862e-05 [1.0133e-06 - 8.2963e-05] (96#)}OK
 
  {
    "result": "OK",
    "performance": {
      "execution_time": "0.804",
      "gc_time": "0.169"
    },
    "created_on": 1586740001294,
    "file_name": "derivativeTest",
    "report": {
      "simpleName": "Float",
      "canonicalName": "com.simiacryptus.mindseye.layers.cudnn.GateBiasLayerTest.Float",
      "link": "https://github.com/SimiaCryptus/mindseye-cudnn/tree/59d5b3318556370acb2d83ee6ec123ce0fc6974f/src/test/java/com/simiacryptus/mindseye/layers/cudnn/GateBiasLayerTest.java",
      "javaDoc": ""
    },
    "archive": "s3://code.simiacrypt.us/tests/com/simiacryptus/mindseye/layers/cudnn/GateBiasLayer/Float/derivativeTest/202004130641",
    "id": "190e3d03-d559-48c6-ac87-3955591c3669",
    "report_type": "Components",
    "display_name": "Derivative Validation",
    "target": {
      "simpleName": "GateBiasLayer",
      "canonicalName": "com.simiacryptus.mindseye.layers.cudnn.GateBiasLayer",
      "link": "https://github.com/SimiaCryptus/mindseye-cudnn/tree/59d5b3318556370acb2d83ee6ec123ce0fc6974f/src/main/java/com/simiacryptus/mindseye/layers/cudnn/GateBiasLayer.java",
      "javaDoc": ""
    }
  }