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

Target Description: The type Re lu activation layer.

Report Description: The type Basic.

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

Test Modules

Using Seed 4128077563362634752

Differential Validation

SingleDerivativeTester.java:153 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: [
[ [ 0.08 ], [ 0.0 ], [ 0.0 ] ],
[ [ 0.7 ], [ 0.496 ], [ 1.764 ] ]
]
Outputs Statistics: {meanExponent=-0.32745792892695036, negative=0, min=0.0, max=1.764, mean=0.5066666666666667, count=6, sum=3.04, positive=4, stdDev=0.6209999105385514, zeros=2}

Feedback Validation

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

SingleDerivativeTester.java:169 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: [ [ 1.0, 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, 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 ], [ 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=0.1111111111111111, count=36, sum=4.0, positive=4, stdDev=0.31426968052735443, zeros=32}
Measured Feedback: [ [ 1.0000000000000286, 0.0, 0.0, 0.0, 0.0, 0.0 ], [ 0.0, 0.9999999999998899, 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, 0.9999999999998899, 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, 0.9999999999998899 ] ]
Measured Statistics: {meanExponent=-3.276302567614995E-14, negative=0, min=0.0, max=1.0000000000000286, mean=0.11111111111110272, count=36, sum=3.999999999999698, positive=4, stdDev=0.31426968052733073, zeros=32}
Feedback Error: [ [ 2.864375403532904E-14, 0.0, 0.0, 0.0, 0.0, 0.0 ], [ 0.0, -1.1013412404281553E-13, 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, -1.1013412404281553E-13, 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, -1.1013412404281553E-13 ] ]
Error Statistics: {meanExponent=-13.104301089584563, negative=3, min=-1.1013412404281553E-13, max=2.864375403532904E-14, mean=-8.382183835919932E-15, count=36, sum=-3.0175861809311755E-13, positive=1, stdDev=3.1037452407598356E-14, zeros=32}

Returns

    {
      "absoluteTol" : {
        "count" : 36,
        "sum" : 3.590461261637756E-13,
        "min" : 0.0,
        "max" : 1.1013412404281553E-13,
        "sumOfSquare" : 3.720904048127126E-26,
        "standardDeviation" : 3.0563273477457106E-14,
        "average" : 9.973503504549322E-15
      },
      "relativeTol" : {
        "count" : 4,
        "sum" : 1.7952306308189672E-13,
        "min" : 1.4321877017664317E-14,
        "max" : 5.50670620214108E-14,
        "sumOfSquare" : 9.302260120318813E-27,
        "standardDeviation" : 1.7643182647570603E-14,
        "average" : 4.488076577047418E-14
      }
    }

Learning Validation

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

SingleDerivativeTester.java:185 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 ]
Implemented Gradient: [ [ 0.08, 0.7, 0.0, 0.496, 0.0, 1.764 ] ]
Implemented Statistics: {meanExponent=-0.32745792892695036, negative=0, min=0.0, max=1.764, mean=0.5066666666666667, count=6, sum=3.04, positive=4, stdDev=0.6209999105385514, zeros=2}
Measured Gradient: [ [ 0.07999999999994123, 0.700000000000145, 0.0, 0.4959999999998299, 0.0, 1.7639999999996547 ] ]
Measured Statistics: {meanExponent=-0.32745792892706604, negative=0, min=0.0, max=1.7639999999996547, mean=0.5066666666665952, count=6, sum=3.039999999999571, positive=4, stdDev=0.6209999105384496, zeros=2}
Gradient Error: [ [ -5.877243136609422E-14, 1.4499512701604544E-13, 0.0, -1.7008616737257398E-13, 0.0, -3.452793606584237E-13 ] ]
Error Statistics: {meanExponent=-12.825158330456688, negative=3, min=-3.452793606584237E-13, max=1.4499512701604544E-13, mean=-7.152380539684107E-14, count=6, sum=-4.2914283238104645E-13, positive=1, stdDev=1.5380236146030143E-13, zeros=2}

Returns

    {
      "absoluteTol" : {
        "count" : 42,
        "sum" : 1.078179212576913E-12,
        "min" : 0.0,
        "max" : 3.452793606584237E-13,
        "sumOfSquare" : 2.0983396725653374E-25,
        "standardDeviation" : 6.585628296421304E-14,
        "average" : 2.567093363278364E-14
      },
      "relativeTol" : {
        "count" : 8,
        "sum" : 9.19744836507784E-13,
        "min" : 1.4321877017664317E-14,
        "max" : 3.673276960382238E-13,
        "sumOfSquare" : 1.9393420774371342E-25,
        "standardDeviation" : 1.0499576611009854E-13,
        "average" : 1.14968104563473E-13
      }
    }

Total Accuracy

The overall agreement accuracy between the implemented derivative and the finite difference estimations:

SingleDerivativeTester.java:200 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: 2.5671e-14 +- 6.5856e-14 [0.0000e+00 - 3.4528e-13] (42#)
relativeTol: 1.1497e-13 +- 1.0500e-13 [1.4322e-14 - 3.6733e-13] (8#)

Frozen and Alive Status

SingleDerivativeTester.java:208 executed in 0.01 seconds (0.000 gc):

    testFrozen(component.addRef(), RefUtil.addRef(inputPrototype));
    testUnFrozen(component.addRef(), RefUtil.addRef(inputPrototype));

LayerTests.java:605 executed in 0.00 seconds (0.000 gc):

    throwException(exceptions.addRef());

Results

classdetailsresult
com.simiacryptus.mindseye.test.unit.SingleDerivativeTesterToleranceStatistics{absoluteTol=2.5671e-14 +- 6.5856e-14 [0.0000e+00 - 3.4528e-13] (42#), relativeTol=1.1497e-13 +- 1.0500e-13 [1.4322e-14 - 3.6733e-13] (8#)}OK
  {
    "result": "OK",
    "performance": {
      "execution_time": "0.195",
      "gc_time": "0.131"
    },
    "created_on": 1587004751331,
    "file_name": "derivativeTest",
    "report": {
      "simpleName": "Basic",
      "canonicalName": "com.simiacryptus.mindseye.layers.java.ReLuActivationLayerTest.Basic",
      "link": "https://github.com/SimiaCryptus/mindseye-java/tree/c9a1867488dc7e77a975f095285b5882c0486db6/src/test/java/com/simiacryptus/mindseye/layers/java/ReLuActivationLayerTest.java",
      "javaDoc": "The type Basic."
    },
    "archive": "s3://code.simiacrypt.us/tests/com/simiacryptus/mindseye/layers/java/ReLuActivationLayer/Basic/derivativeTest/202004163911",
    "id": "ac74607d-699b-433f-a8eb-ef45cab79318",
    "report_type": "Components",
    "display_name": "Derivative Validation",
    "target": {
      "simpleName": "ReLuActivationLayer",
      "canonicalName": "com.simiacryptus.mindseye.layers.java.ReLuActivationLayer",
      "link": "https://github.com/SimiaCryptus/mindseye-java/tree/c9a1867488dc7e77a975f095285b5882c0486db6/src/main/java/com/simiacryptus/mindseye/layers/java/ReLuActivationLayer.java",
      "javaDoc": "The type Re lu activation layer."
    }
  }