Subreport: Logs for com.simiacryptus.ref.lang.ReferenceCountingBase
Using Seed 6104476664242658304
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, -0.608, 1.208 ], [ -0.128, 0.048, -1.028 ] ],
[ [ 0.7, 1.764, -1.72 ], [ 0.496, 1.524, -0.384 ] ]
]
Inputs Statistics: {meanExponent=-0.3033810201823816, negative=5, min=-1.72, max=1.764, mean=0.16266666666666665, count=12, sum=1.952, positive=7, stdDev=0.9945442283891763, zeros=0}
Output: [
[ [ -1.524326895743639, -2.2123268957436393, -0.3963268957436392 ], [ -0.9551297254091832, -0.7791297254091831, -1.8551297254091832 ] ],
[ [ -1.3830042577785624, -0.31900425777856234, -3.8030042577785625 ], [ -1.4375225540020196, -0.40952255400201953, -2.3175225540020197 ] ]
]
Outputs Statistics: {meanExponent=0.05214088725588086, negative=12, min=-3.8030042577785625, max=-0.31900425777856234, mean=-1.4493291915666846, count=12, sum=-17.391950298800214, positive=0, stdDev=0.9665917451781383, zeros=0}
We validate the agreement between the implemented derivative of the inputs apply finite difference estimations:
SingleDerivativeTester.java:117 executed in 0.17 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, -0.608, 1.208 ], [ -0.128, 0.048, -1.028 ] ],
[ [ 0.7, 1.764, -1.72 ], [ 0.496, 1.524, -0.384 ] ]
]
Value Statistics: {meanExponent=-0.3033810201823816, negative=5, min=-1.72, max=1.764, mean=0.16266666666666665, count=12, sum=1.952, positive=7, stdDev=0.9945442283891763, zeros=0}
Implemented Feedback: [ [ 0.7822324121653401, 0.0, 0.0, 0.0, -0.21776758783465983, 0.0, 0.0, 0.0, ... ], [ 0.0, 0.7491761195820035, 0.0, 0.0, 0.0, -0.2508238804179966, 0.0, 0.0, ... ], [ 0.0, 0.0, 0.6152377721221317, 0.0, 0.0, 0.0, -0.3847622278778683, 0.0, ... ], [ 0.0, 0.0, 0.0, 0.7624845378891645, 0.0, 0.0, 0.0, -0.23751546211083552, ... ], [ -0.10944568329718807, 0.0, 0.0, 0.0, 0.8905543167028119, 0.0, 0.0, 0.0, ... ], [ 0.0, -0.7268724544377956, 0.0, 0.0, 0.0, 0.2731275455622044, 0.0, 0.0, ... ], [ 0.0, 0.0, -0.4588051240527918, 0.0, 0.0, 0.0, 0.5411948759472083, 0.0, ... ], [ 0.0, 0.0, 0.0, -0.6639671829454358, 0.0, 0.0, 0.0, 0.3360328170545642, ... ], ... ]
Implemented Statistics: {meanExponent=-0.4898719169405722, negative=24, min=-0.7268724544377956, max=0.9776963348557921, mean=5.396917480816733E-18, count=144, sum=7.771561172376096E-16, positive=12, stdDev=0.26327730170635294, zeros=108}
Measured Feedback: [ [ 0.782223894761902, 0.0, 0.0, 0.0, -0.21777610523798785, 0.0, 0.0, 0.0, ... ], [ 0.0, 0.7491667238634925, 0.0, 0.0, 0.0, -0.2508332761363974, 0.0, 0.0, ... ], [ 0.0, 0.0, 0.6152259360203693, 0.0, 0.0, 0.0, -0.3847740639795205, 0.0, ... ], [ 0.0, 0.0, 0.0, 0.7624754826363045, 0.0, 0.0, 0.0, -0.23752451736358537, ... ], [ -0.10945055679112414, 0.0, 0.0, 0.0, 0.8905494432109862, 0.0, 0.0, 0.0, ... ], [ 0.0, -0.7268823807349278, 0.0, 0.0, 0.0, 0.27311761926496203, 0.0, 0.0, ... ], [ 0.0, 0.0, -0.458817539236156, 0.0, 0.0, 0.0, 0.5411824607637339, 0.0, ... ], [ 0.0, 0.0, 0.0, -0.6639783385620213, 0.0, 0.0, 0.0, 0.3360216614378686, ... ], ... ]
Measured Statistics: {meanExponent=-0.4898646787482104, negative=24, min=-0.7268823807349278, max=0.9776952445106346, mean=-2.0898217724231295E-6, count=144, sum=-3.0093433522893065E-4, positive=12, stdDev=0.2632778738709077, zeros=108}
Feedback Error: [ [ -8.517403438101923E-6, 0.0, 0.0, 0.0, -8.51740332802331E-6, 0.0, 0.0, 0.0, ... ], [ 0.0, -9.395718510996076E-6, 0.0, 0.0, 0.0, -9.395718400806441E-6, 0.0, 0.0, ... ], [ 0.0, 0.0, -1.1836101762363604E-5, 0.0, 0.0, 0.0, -1.183610165222948E-5, 0.0, ... ], [ 0.0, 0.0, 0.0, -9.055252859990759E-6, 0.0, 0.0, 0.0, -9.055252749856635E-6, ... ], [ -4.873493936075057E-6, 0.0, 0.0, 0.0, -4.873491825763132E-6, 0.0, 0.0, 0.0, ... ], [ 0.0, -9.926297132256678E-6, 0.0, 0.0, 0.0, -9.926297242390802E-6, 0.0, 0.0, ... ], [ 0.0, 0.0, -1.2415183364200555E-5, 0.0, 0.0, 0.0, -1.241518347439019E-5, 0.0, ... ], [ 0.0, 0.0, 0.0, -1.1155616585489803E-5, 0.0, 0.0, 0.0, -1.1155616695623927E-5, ... ], ... ]
Error Statistics: {meanExponent=-5.141205231194625, negative=36, min=-1.241518447442358E-5, max=0.0, mean=-2.0898217724291527E-6, count=144, sum=-3.00934335229798E-4, positive=0, stdDev=3.978462985612465E-6, zeros=108}
Returns
{
"absoluteTol" : {
"count" : 144,
"sum" : 3.00934335229798E-4,
"min" : 0.0,
"max" : 1.241518447442358E-5,
"sumOfSquare" : 2.908155278650662E-9,
"standardDeviation" : 3.978462985612465E-6,
"average" : 2.0898217724291527E-6
},
"relativeTol" : {
"count" : 36,
"sum" : 5.000004819939184E-4,
"min" : 5.576096210091851E-7,
"max" : 2.4442633301452194E-5,
"sumOfSquare" : 8.738370473281865E-9,
"standardDeviation" : 7.0591009816937945E-6,
"average" : 1.3888902277608845E-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" : 144,
"sum" : 3.00934335229798E-4,
"min" : 0.0,
"max" : 1.241518447442358E-5,
"sumOfSquare" : 2.908155278650662E-9,
"standardDeviation" : 3.978462985612465E-6,
"average" : 2.0898217724291527E-6
},
"relativeTol" : {
"count" : 36,
"sum" : 5.000004819939184E-4,
"min" : 5.576096210091851E-7,
"max" : 2.4442633301452194E-5,
"sumOfSquare" : 8.738370473281865E-9,
"standardDeviation" : 7.0591009816937945E-6,
"average" : 1.3888902277608845E-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: 2.0898e-06 +- 3.9785e-06 [0.0000e+00 - 1.2415e-05] (144#)
relativeTol: 1.3889e-05 +- 7.0591e-06 [5.5761e-07 - 2.4443e-05] (36#)
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=2.0898e-06 +- 3.9785e-06 [0.0000e+00 - 1.2415e-05] (144#), relativeTol=1.3889e-05 +- 7.0591e-06 [5.5761e-07 - 2.4443e-05] (36#)} | OK |
{
"result": "OK",
"performance": {
"execution_time": "0.352",
"gc_time": "0.128"
},
"created_on": 1586743156352,
"file_name": "derivativeTest",
"report": {
"simpleName": "PixelLog",
"canonicalName": "com.simiacryptus.mindseye.layers.cudnn.SoftmaxLayerTest.PixelLog",
"link": "https://github.com/SimiaCryptus/mindseye-cudnn/tree/59d5b3318556370acb2d83ee6ec123ce0fc6974f/src/test/java/com/simiacryptus/mindseye/layers/cudnn/SoftmaxLayerTest.java",
"javaDoc": ""
},
"archive": "s3://code.simiacrypt.us/tests/com/simiacryptus/mindseye/layers/cudnn/SoftmaxActivationLayer/PixelLog/derivativeTest/202004135916",
"id": "9f432eb2-9f4b-4dec-9556-555ba6b374cd",
"report_type": "Components",
"display_name": "Derivative Validation",
"target": {
"simpleName": "SoftmaxActivationLayer",
"canonicalName": "com.simiacryptus.mindseye.layers.cudnn.SoftmaxActivationLayer",
"link": "https://github.com/SimiaCryptus/mindseye-cudnn/tree/59d5b3318556370acb2d83ee6ec123ce0fc6974f/src/main/java/com/simiacryptus/mindseye/layers/cudnn/SoftmaxActivationLayer.java",
"javaDoc": ""
}
}