1. Test Modules
  2. Training Characteristics
    1. Input Learning
      1. Gradient Descent
      2. Conjugate Gradient Descent
      3. Limited-Memory BFGS
    2. Results
  3. Results

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

Test Modules

Using Seed 3606190369892070400

Training Characteristics

Input Learning

In this apply, we use a network to learn this target input, given it's pre-evaluated output:

TrainingTester.java:332 executed in 0.01 seconds (0.000 gc):

    return RefArrays.stream(RefUtil.addRef(input_target)).flatMap(RefArrays::stream).map(x -> {
      try {
        return x.prettyPrint();
      } finally {
        x.freeRef();
      }
    }).reduce((a, b) -> a + "\n" + b).orElse("");

Returns

    [
    	[ [ 0.496 ], [ -1.228 ], [ 0.3 ], [ 1.048 ], [ -1.54 ], [ -0.012 ], [ -0.128 ], [ 1.368 ] ],
    	[ [ 1.556 ], [ -0.176 ], [ -0.768 ], [ -0.408 ], [ -0.892 ], [ 0.788 ], [ 1.356 ], [ 1.704 ] ],
    	[ [ -0.804 ], [ 1.912 ], [ 0.392 ], [ -0.384 ], [ 1.764 ], [ -1.492 ], [ -0.876 ], [ 1.108 ] ],
    	[ [ 0.048 ], [ -0.636 ], [ 1.64 ], [ 1.556 ], [ 1.032 ], [ -1.028 ], [ -1.832 ], [ 1.552 ] ],
    	[ [ -1.72 ], [ 1.512 ], [ 0.08 ], [ 1.876 ], [ -0.068 ], [ -0.384 ], [ 0.148 ], [ 0.996 ] ],
    	[ [ -1.616 ], [ -1.688 ], [ 0.048 ], [ 1.524 ], [ -0.712 ], [ -0.852 ], [ -1.424 ], [ 1.652 ] ],
    	[ [ 0.092 ], [ -0.556 ], [ -0.464 ], [ 1.208 ], [ -1.516 ], [ 1.62 ], [ -1.856 ], [ 0.636 ] ],
    	[ [ 0.7 ], [ 1.612 ], [ -0.032 ], [ -1.572 ], [ -0.804 ], [ 0.028 ], [ -1.476 ], [ -0.608 ] ]
    ]
    [
    	[ [ -0.408 ], [ 0.048 ], [ 1.356 ], [ -0.804 ], [ -0.068 ], [ -0.608 ], [ -1.572 ], [ 1.512 ] ],
    	[ [ 0.048 ], [ 0.08 ], [ -0.876 ], [ -0.804 ], [ -0.892 ], [ 1.524 ], [ -1.228 ], [ -0.012 ] ],
    	[ [ 0.392 ], [ -0.032 ], [ 1.556 ], [ 0.148 ], [ 0.028 ], [ -1.492 ], [ 0.7 ], [ -0.768 ] ],
    	[ [ 1.764 ], [ -0.384 ], [ -0.176 ], [ 0.092 ], [ 1.912 ], [ -1.476 ], [ -0.852 ], [ 0.3 ] ],
    	[ [ -1.424 ], [ -0.712 ], [ 1.032 ], [ -0.384 ], [ -1.54 ], [ 1.048 ], [ -1.616 ], [ -1.832 ] ],
    	[ [ -0.636 ], [ 1.704 ], [ 1.556 ], [ 1.652 ], [ -0.556 ], [ -1.72 ], [ 0.496 ], [ 1.64 ] ],
    	[ [ 0.996 ], [ -1.856 ], [ 1.552 ], [ -1.028 ], [ -1.688 ], [ 1.612 ], [ 1.876 ], [ 1.208 ] ],
    	[ [ 0.636 ], [ 0.788 ], [ 1.62 ], [ -0.464 ], [ 1.108 ], [ -1.516 ], [ -0.128 ], [ 1.368 ] ]
    ]
    [
    	[ [ -1.832 ], [ 1.64 ], [ -0.608 ], [ 0.092 ], [ -1.476 ], [ 0.148 ], [ -0.176 ], [ 1.108 ] ],
    	[ [ -0.636 ], [ -0.384 ], [ -1.492 ], [ -1.228 ], [ -0.032 ], [ -0.068 ], [ 0.496 ], [ 0.3 ] ],
    	[ [ 1.556 ], [ -0.464 ], [ -1.688 ], [ -0.408 ], [ 0.996 ], [ 1.556 ], [ -0.384 ], [ -0.556 ] ],
    	[ [ -1.856 ], [ 1.524 ], [ -0.804 ], [ -1.424 ], [ 1.912 ], [ 1.62 ], [ 1.876 ], [ -1.572 ] ],
    	[ [ 0.7 ], [ 0.788 ], [ -1.72 ], [ -1.54 ], [ -0.012 ], [ -0.876 ], [ -0.128 ], [ -0.852 ] ],
    	[ [ 0.392 ], [ 0.048 ], [ -0.712 ], [ 0.08 ], [ -1.516 ], [ 1.764 ], [ -0.892 ], [ -1.616 ] ],
    	[ [ -1.028 ], [ 1.032 ], [ 0.636 ], [ 1.368 ], [ 1.048 ], [ 1.208 ], [ 1.652 ], [ 1.612 ] ],
    	[ [ 0.028 ], [ 0.048 ], [ -0.768 ], [ 1.512 ], [ 1.552 ], [ 1.704 ], [ 1.356 ], [ -0.804 ] ]
    ]
    [
    	[ [ 0.996 ], [ 1.912 ], [ -1.424 ], [ 0.048 ], [ 0.496 ], [ -0.032 ], [ 0.7 ], [ -1.72 ] ],
    	[ [ 0.048 ], [ -1.476 ], [ -0.384 ], [ -0.408 ], [ -1.028 ], [ 1.62 ], [ -1.54 ], [ 1.652 ] ],
    	[ [ 0.092 ], [ -0.556 ], [ -0.636 ], [ 1.552 ], [ 1.032 ], [ 1.556 ], [ 0.3 ], [ -0.068 ] ],
    	[ [ -0.768 ], [ 0.392 ], [ 1.764 ], [ 1.512 ], [ -1.616 ], [ -0.384 ], [ -1.832 ], [ -0.876 ] ],
    	[ [ -0.712 ], [ -0.892 ], [ -0.128 ], [ 1.048 ], [ 0.788 ], [ 0.148 ], [ -0.804 ], [ 1.612 ] ],
    	[ [ -1.228 ], [ 1.368 ], [ -0.012 ], [ 1.556 ], [ -0.804 ], [ 1.356 ], [ 1.64 ], [ 1.524 ] ],
    	[ [ 1.876 ], [ -0.852 ], [ -1.492 ], [ 1.108 ], [ 0.028 ], [ -1.688 ], [ 0.08 ], [ 1.704 ] ],
    	[ [ -1.856 ], [ 0.636 ], [ -0.176 ], [ -0.464 ], [ -1.572 ], [ -1.516 ], [ 1.208 ], [ -0.608 ] ]
    ]
    [
    	[ [ 0.7 ], [ -0.012 ], [ -0.804 ], [ -0.408 ], [ 1.552 ], [ 0.028 ], [ 1.208 ], [ 0.092 ] ],
    	[ [ -0.384 ], [ -0.892 ], [ 0.996 ], [ -1.028 ], [ 1.652 ], [ -0.032 ], [ 1.556 ], [ -0.876 ] ],
    	[ [ 1.048 ], [ 1.764 ], [ 1.512 ], [ -0.556 ], [ 0.392 ], [ -1.856 ], [ 0.148 ], [ -1.832 ] ],
    	[ [ -0.712 ], [ 1.612 ], [ 0.048 ], [ -0.384 ], [ -0.608 ], [ -1.228 ], [ 0.048 ], [ 0.496 ] ],
    	[ [ -1.688 ], [ 1.912 ], [ 0.3 ], [ -0.176 ], [ -1.476 ], [ -1.516 ], [ 1.64 ], [ -1.424 ] ],
    	[ [ -1.54 ], [ -1.572 ], [ 0.08 ], [ 1.108 ], [ -1.72 ], [ 0.636 ], [ -0.068 ], [ -1.492 ] ],
    	[ [ -0.852 ], [ 1.62 ], [ -0.464 ], [ 1.524 ], [ -0.636 ], [ 1.556 ], [ -0.804 ], [ 1.032 ] ],
    	[ [ -0.768 ], [ 0.788 ], [ 1.704 ], [ 1.368 ], [ -1.616 ], [ 1.876 ], [ -0.128 ], [ 1.356 ] ]
    ]

Gradient Descent

First, we train using basic gradient descent method apply weak line search conditions.

TrainingTester.java:480 executed in 1.00 seconds (0.000 gc):

    IterativeTrainer iterativeTrainer = new IterativeTrainer(trainable.addRef());
    try {
      iterativeTrainer.setLineSearchFactory(label -> new ArmijoWolfeSearch());
      iterativeTrainer.setOrientation(new GradientDescent());
      iterativeTrainer.setMonitor(TrainingTester.getMonitor(history));
      iterativeTrainer.setTimeout(30, TimeUnit.SECONDS);
      iterativeTrainer.setMaxIterations(250);
      iterativeTrainer.setTerminateThreshold(0);
      return iterativeTrainer.run();
    } finally {
      iterativeTrainer.freeRef();
    }
Logging
Reset training subject: 891547765105
BACKPROP_AGG_SIZE = 3
THREADS = 64
SINGLE_THREADED = false
Initialized CoreSettings = {
"backpropAggregationSize" : 3,
"jvmThreads" : 64,
"singleThreaded" : false
}
Reset training subject: 891585107994
Constructing line search parameters: GD
th(0)=528.8293711151897;dx=-5.43112448E24
New Minimum: 528.8293711151897 > 0.04544361193472045
Armijo: th(2.154434690031884)=0.04544361193472045; dx=-5.431124480018978E12 evalInputDelta=528.783927503255
Armijo: th(1.077217345015942)=0.13207064317807687; dx=-5.431124480019014E12 evalInputDelta=528.6973004720116
Armijo: th(0.3590724483386473)=1.3220433701289842; dx=-5.431124480020144E12 evalInputDelta=527.5073277450607
Armijo: th(0.08976811208466183)=4.747368291706443; dx=-5.431124480035212E12 evalInputDelta=524.0820028234832
Armijo: th(0.017953622416932366)=8.541434805177051; dx=-5.43112448009962E12 evalInputDelta=520.2879363100127
Armijo: th(0.002992270402822061)=10.607012226712174; dx=-5.431124480214101E12 evalInputDelta=518.2223588884775
Armijo: th(4.2746720040315154E-4)=11.231022614624631; dx=-5.431124480284546E12 evalInputDelta=517.5983485005651
Armijo: th(5.343340005039394E-5)=11.339602215220847; dx=-5.43112448030026E12 evalInputDelta=517.4897688999689
Armijo: th(5.9370444500437714E-6)=11.35382284632942; dx=-5.431124480302417E12 evalInputDelta=517.4755482688603
Armijo: th(5.937044450043771E-7)=11.355429102301345; dx=-5.431124480302662E12 evalInputDelta=517.4739420128883
Armijo: th(5.397313136403428E-8)=11.35559142324907; dx=-5.431124480302687E12 evalInputDelta=517.4737796919406
Armijo: th(4.4977609470028565E-9)=11.355606303338789; dx=-5.431124480302689E12 evalInputDelta=517.473764811851
Armijo: th(3.4598161130791205E-10)=11.359937918203105; dx=-5.431161073070346E12 evalInputDelta=517.4694331969866
Armijo: th(2.4712972236279432E-11)=18.069740830470938; dx=-6.972800005440112E21 evalInputDelta=510.75963028471875
Armijo: th(1.6475314824186289E-12)=392.0001701056701; dx=-3.714277760008232E24 evalInputDelta=136.82920100951958
Armijo: th(1.029707176511643E-13)=494.3126824546316; dx=-4.881745280049965E24 evalInputDelta=34.51668866055809
Armijo: th(6.057101038303783E-15)=528.8293711151879; dx=-5.43112448E24 evalInputDelta=1.8189894035458565E-12
MIN ALPHA (3.3650561323909904E-16): th(2.154434690031884)=0.04544361193472045
Fitness changed from 528.8293711151897 to 0.04544361193472045
Iteration 1 complete. Error: 0.04544361193472045 Total: 0.2343; Orientation: 0.0040; Line Search: 0.1729
th(0)=0.04544361193472045;dx=-9.462489633836965
New Minimum: 0.04544361193472045 > 0.04544361193472043
WOLFE (weak): th(2.154434690031884E-15)=0.04544361193472043; dx=-9.462489633836965 evalInputDelta=2.0816681711721685E-17
New Minimum: 0.04544361193472043 > 0.04544361193472041
WOLFE (weak): th(4.308869380063768E-15)=0.04544361193472041; dx=-9.462489633836965 evalInputDelta=4.163336342344337E-17
New Minimum: 0.04544361193472041 > 0.04544361193472034
WOLFE (weak): th(1.2926608140191303E-14)=0.04544361193472034; dx=-9.462489633836965 evalInputDelta=1.1102230246251565E-16
New Minimum: 0.04544361193472034 > 0.04544361193472
WOLFE (weak): th(5.1706432560765214E-14)=0.04544361193472; dx=-9.462489633836965 evalInputDelta=4.510281037539698E-16
New Minimum: 0.04544361193472 > 0.04544361193471817
WOLFE (weak): th(2.5853216280382605E-13)=0.04544361193471817; dx=-9.462489633836965 evalInputDelta=2.275957200481571E-15
New Minimum: 0.04544361193471817 > 0.04544361193470677
WOLFE (weak): th(1.5511929768229563E-12)=0.04544361193470677; dx=-9.462489633836963 evalInputDelta=1.3676559884601147E-14
New Minimum: 0.04544361193470677 > 0.04544361193462471
WOLFE (weak): th(1.0858350837760695E-11)=0.04544361193462471; dx=-9.462489633836956 evalInputDelta=9.573591919220803E-14
New Minimum: 0.04544361193462471 > 0.04544361193395451
WOLFE (weak): th(8.686680670208556E-11)=0.04544361193395451; dx=-9.462489633836901 evalInputDelta=7.659359257949916E-13
New Minimum: 0.04544361193395451 > 0.04544361192782706
WOLFE (weak): th(7.8180126031877E-10)=0.04544361192782706; dx=-9.462489633836391 evalInputDelta=6.893388637685405E-12
New Minimum: 0.04544361192782706 > 0.045443611865786576
WOLFE (weak): th(7.818012603187701E-9)=0.045443611865786576; dx=-9.462489633831225 evalInputDelta=6.893387249906624E-11
New Minimum: 0.045443611865786576 > 0.04544361117644787
WOLFE (weak): th(8.599813863506471E-8)=0.04544361117644787; dx=-9.462489633773824 evalInputDelta=7.582725766730469E-10
New Minimum: 0.04544361117644787 > 0.04544360283544991
WOLFE (weak): th(1.0319776636207765E-6)=0.04544360283544991; dx=-9.462489633079281 evalInputDelta=9.099270538437398E-9
New Minimum: 0.04544360283544991 > 0.045443493644264386
WOLFE (weak): th(1.3415709627070094E-5)=0.045443493644264386; dx=-9.462489623987086 evalInputDelta=1.1829045606231992E-7
New Minimum: 0.045443493644264386 > 0.04544195588036056
WOLFE (weak): th(1.878199347789813E-4)=0.04544195588036056; dx=-9.462489495940819 evalInputDelta=1.6560543598873045E-6
New Minimum: 0.04544195588036056 > 0.0454187738383449
WOLFE (weak): th(0.0028172990216847197)=0.0454187738383449; dx=-9.462487565879371 evalInputDelta=2.4838096375545682E-5
New Minimum: 0.0454187738383449 > 0.04504689971110515
WOLFE (weak): th(0.045076784346955515)=0.04504689971110515; dx=-9.462456670613303 evalInputDelta=3.9671222361530034E-4
New Minimum: 0.04504689971110515 > 0.038893106300604316
WOLFE (weak): th(0.7663053338982437)=0.038893106300604316; dx=-9.461962897278429 evalInputDelta=0.006550505634116133
New Minimum: 0.038893106300604316 > 0.0036735237306089906
WOLFE (weak): th(13.793496010168386)=0.0036735237306089906; dx=-9.45990705356817 evalInputDelta=0.04177008820411146
New Minimum: 0.0036735237306089906 > 0.0
WOLFE (weak): th(262.07642419319933)=0.0; dx=-9.459778418829957 evalInputDelta=0.04544361193472045
Armijo: th(5241.528483863986)=0.0; dx=-9.459778418829957 evalInputDelta=0.04544361193472045
WOLFE (weak): th(2751.802454028593)=0.0; dx=-9.459778418829957 evalInputDelta=0.04544361193472045
WOLFE (weak): th(3996.6654689462894)=0.0; dx=-9.459778418829957 evalInputDelta=0.04544361193472045
WOLFE (weak): th(4619.096976405138)=0.0; dx=-9.459778418829957 evalInputDelta=0.04544361193472045
Armijo: th(4930.312730134562)=0.0; dx=-9.459778418829957 evalInputDelta=0.04544361193472045
WOLFE (weak): th(4774.70485326985)=0.0; dx=-9.459778418829957 evalInputDelta=0.04544361193472045
Armijo: th(4852.5087917022065)=0.0; dx=-9.459778418829957 evalInputDelta=0.04544361193472045
Armijo: th(4813.606822486028)=0.0; dx=-9.459778418829957 evalInputDelta=0.04544361193472045
mu ~= nu (4774.70485326985): th(262.07642419319933)=0.0
Fitness changed from 0.04544361193472045 to 0.0
Iteration 2 complete. Error: 0.0 Total: 0.1832; Orientation: 0.0014; Line Search: 0.1759
th(0)=0.0;dx=-9.458297600000002
Armijo: th(10328.695646543105)=0.0; dx=-9.458297600000002 evalInputDelta=0.0
Armijo: th(5164.347823271552)=0.0; dx=-9.458297600000002 evalInputDelta=0.0
Armijo: th(1721.4492744238507)=0.0; dx=-9.458297600000002 evalInputDelta=0.0
Armijo: th(430.3623186059627)=0.0; dx=-9.458297600000002 evalInputDelta=0.0
Armijo: th(86.07246372119253)=0.0; dx=-9.458297600000002 evalInputDelta=0.0
Armijo: th(14.345410620198756)=0.0; dx=-9.458297600000002 evalInputDelta=0.0
Armijo: th(2.0493443743141078)=0.0; dx=-9.458297600000002 evalInputDelta=0.0
Armijo: th(0.25616804678926347)=0.0; dx=-9.458297600000002 evalInputDelta=0.0
Armijo: th(0.028463116309918164)=0.0; dx=-9.458297600000002 evalInputDelta=0.0
Armijo: th(0.0028463116309918163)=0.0; dx=-9.458297600000002 evalInputDelta=0.0
Armijo: th(2.5875560281743784E-4)=0.0; dx=-9.458297600000002 evalInputDelta=0.0
Armijo: th(2.1562966901453154E-5)=0.0; dx=-9.458297600000002 evalInputDelta=0.0
Armijo: th(1.6586897616502425E-6)=0.0; dx=-9.458297600000002 evalInputDelta=0.0
Armijo: th(1.1847784011787447E-7)=0.0; dx=-9.458297600000002 evalInputDelta=0.0
Armijo: th(7.898522674524965E-9)=0.0; dx=-9.458297600000002 evalInputDelta=0.0
Armijo: th(4.936576671578103E-10)=0.0; dx=-9.458297600000002 evalInputDelta=0.0
Armijo: th(2.9038686303400606E-11)=0.0; dx=-9.458297600000002 evalInputDelta=0.0
Armijo: th(1.6132603501889225E-12)=0.0; dx=-9.458297600000002 evalInputDelta=0.0
Armijo: th(8.49084394836275E-14)=0.0; dx=-9.458297600000002 evalInputDelta=0.0
Armijo: th(4.2454219741813745E-15)=0.0; dx=-9.458297600000002 evalInputDelta=0.0
MIN ALPHA (2.0216295115149403E-16): th(0.0)=0.0
Fitness changed from 0.0 to 0.0
Static Iteration Total: 0.5709; Orientation: 0.0016; Line Search: 0.5662
Iteration 3 failed. Error: 0.0
Previous Error: 0.0 -> 0.0
Optimization terminated 3
Final threshold in iteration 3: 0.0 (> 0.0) after 0.989s (< 30.000s)

Returns

    0.0

Training Converged

Conjugate Gradient Descent

First, we use a conjugate gradient descent method, which converges the fastest for purely linear functions.

TrainingTester.java:452 executed in 1.11 seconds (0.000 gc):

    IterativeTrainer iterativeTrainer = new IterativeTrainer(trainable.addRef());
    try {
      iterativeTrainer.setLineSearchFactory(label -> new QuadraticSearch());
      iterativeTrainer.setOrientation(new GradientDescent());
      iterativeTrainer.setMonitor(TrainingTester.getMonitor(history));
      iterativeTrainer.setTimeout(30, TimeUnit.SECONDS);
      iterativeTrainer.setMaxIterations(250);
      iterativeTrainer.setTerminateThreshold(0);
      return iterativeTrainer.run();
    } finally {
      iterativeTrainer.freeRef();
    }
Logging
Reset training subject: 892543895555
Reset training subject: 892547581082
Constructing line search parameters: GD
F(0.0) = LineSearchPoint{point=PointSample{avg=528.8293711151897}, derivative=-5.43112448E24}
New Minimum: 528.8293711151897 > 12.412350451453152
F(1.0E-10) = LineSearchPoint{point=PointSample{avg=12.412350451453152}, derivative=-4.1088000543172054E20}, evalInputDelta = -516.4170206637366
New Minimum: 12.412350451453152 > 11.3556074455486
F(7.000000000000001E-10) = LineSearchPoint{point=PointSample{avg=11.3556074455486}, derivative=-5.431124480302689E12}, evalInputDelta = -517.4737636696411
New Minimum: 11.3556074455486 > 11.355606182361914
F(4.900000000000001E-9) = LineSearchPoint{point=PointSample{avg=11.355606182361914}, derivative=-5.431124480302689E12}, evalInputDelta = -517.4737649328277
New Minimum: 11.355606182361914 > 11.355597340077809
F(3.430000000000001E-8) = LineSearchPoint{point=PointSample{avg=11.355597340077809}, derivative=-5.4311244803026875E12}, evalInputDelta = -517.4737737751119
New Minimum: 11.355597340077809 > 11.3555354452008
F(2.4010000000000004E-7) = LineSearchPoint{point=PointSample{avg=11.3555354452008}, derivative=-5.431124480302678E12}, evalInputDelta = -517.473835669989
New Minimum: 11.3555354452008 > 11.355102235525198
F(1.6807000000000003E-6) = LineSearchPoint{point=PointSample{avg=11.355102235525198}, derivative=-5.431124480302612E12}, evalInputDelta = -517.4742688796645
New Minimum: 11.355102235525198 > 11.352072432449727
F(1.1764900000000001E-5) = LineSearchPoint{point=PointSample{avg=11.352072432449727}, derivative=-5.43112448030215E12}, evalInputDelta = -517.47729868274
New Minimum: 11.352072432449727 > 11.330993006849264
F(8.235430000000001E-5) = LineSearchPoint{point=PointSample{avg=11.330993006849264}, derivative=-5.431124480298967E12}, evalInputDelta = -517.4983781083405
New Minimum: 11.330993006849264 > 11.189337628914668
F(5.764801000000001E-4) = LineSearchPoint{point=PointSample{avg=11.189337628914668}, derivative=-5.431124480278846E12}, evalInputDelta = -517.640033486275
New Minimum: 11.189337628914668 > 10.395547921566154
F(0.004035360700000001) = LineSearchPoint{point=PointSample{avg=10.395547921566154}, derivative=-5.43112448019587E12}, evalInputDelta = -518.4338231936235
New Minimum: 10.395547921566154 > 7.671363612571414
F(0.028247524900000005) = LineSearchPoint{point=PointSample{avg=7.671363612571414}, derivative=-5.431124480075977E12}, evalInputDelta = -521.1580075026183
New Minimum: 7.671363612571414 > 2.5977084936444568
F(0.19773267430000002) = LineSearchPoint{point=PointSample{avg=2.5977084936444568}, derivative=-5.431124480023134E12}, evalInputDelta = -526.2316626215453
New Minimum: 2.5977084936444568 > 0.08290809653053569
F(1.3841287201) = LineSearchPoint{point=PointSample{avg=0.08290809653053569}, derivative=-5.43112448001899E12}, evalInputDelta = -528.7464630186591
New Minimum: 0.08290809653053569 > 0.0016082136004099798
F(9.688901040700001) = LineSearchPoint{point=PointSample{avg=0.0016082136004099798}, derivative=-5.431124480018971E12}, evalInputDelta = -528.8277629015893
New Minimum: 0.0016082136004099798 > 0.0
F(67.8223072849) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-5.431124480018971E12}, evalInputDelta = -528.8293711151897
F(474.7561509943) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-5.431124480018971E12}, evalInputDelta = -528.8293711151897
F(3323.2930569601003) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-5.431124480018971E12}, evalInputDelta = -528.8293711151897
F(23263.0513987207) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-5.431124480018971E12}, evalInputDelta = -528.8293711151897
F(162841.3597910449) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-5.431124480018971E12}, evalInputDelta = -528.8293711151897
F(1139889.5185373144) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-5.431124480018971E12}, evalInputDelta = -528.8293711151897
F(7979226.6297612) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-5.431124480018971E12}, evalInputDelta = -528.8293711151897
F(5.58545864083284E7) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-5.431124480018971E12}, evalInputDelta = -528.8293711151897
F(3.909821048582988E8) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-5.431124480018971E12}, evalInputDelta = -528.8293711151897
F(2.7368747340080914E9) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-5.431124480018971E12}, evalInputDelta = -528.8293711151897
F(1.915812313805664E10) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-5.431124480018971E12}, evalInputDelta = -528.8293711151897
0.0 <= 528.8293711151897
F(1.0E10) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-5.431124480018971E12}, evalInputDelta = -528.8293711151897
Right bracket at 1.0E10
Converged to right
Fitness changed from 528.8293711151897 to 0.0
Iteration 1 complete. Error: 0.0 Total: 0.9123; Orientation: 0.0015; Line Search: 0.9012
F(0.0) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-9.458297600000002}
F(1.0E10) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-9.458297600000002}, evalInputDelta = 0.0
0.0 <= 0.0
F(5.0E9) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-9.458297600000002}, evalInputDelta = 0.0
Right bracket at 5.0E9
F(2.5E9) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-9.458297600000002}, evalInputDelta = 0.0
Right bracket at 2.5E9
F(1.25E9) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-9.458297600000002}, evalInputDelta = 0.0
Right bracket at 1.25E9
F(6.25E8) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-9.458297600000002}, evalInputDelta = 0.0
Right bracket at 6.25E8
F(3.125E8) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-9.458297600000002}, evalInputDelta = 0.0
Right bracket at 3.125E8
F(1.5625E8) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-9.458297600000002}, evalInputDelta = 0.0
Right bracket at 1.5625E8
F(7.8125E7) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-9.458297600000002}, evalInputDelta = 0.0
Right bracket at 7.8125E7
F(3.90625E7) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-9.458297600000002}, evalInputDelta = 0.0
Right bracket at 3.90625E7
F(1.953125E7) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-9.458297600000002}, evalInputDelta = 0.0
Right bracket at 1.953125E7
F(9765625.0) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-9.458297600000002}, evalInputDelta = 0.0
Right bracket at 9765625.0
F(4882812.5) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-9.458297600000002}, evalInputDelta = 0.0
Right bracket at 4882812.5
F(2441406.25) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-9.458297600000002}, evalInputDelta = 0.0
Loops = 12
Fitness changed from 0.0 to 0.0
Static Iteration Total: 0.1925; Orientation: 0.0012; Line Search: 0.1885
Iteration 2 failed. Error: 0.0
Previous Error: 0.0 -> 0.0
Optimization terminated 2
Final threshold in iteration 2: 0.0 (> 0.0) after 1.105s (< 30.000s)

Returns

    0.0

Training Converged

Limited-Memory BFGS

Next, we apply the same optimization using L-BFGS, which is nearly ideal for purely second-order or quadratic functions.

TrainingTester.java:509 executed in 11.38 seconds (0.059 gc):

    IterativeTrainer iterativeTrainer = new IterativeTrainer(trainable.addRef());
    try {
      iterativeTrainer.setLineSearchFactory(label -> new ArmijoWolfeSearch());
      iterativeTrainer.setOrientation(new LBFGS());
      iterativeTrainer.setMonitor(TrainingTester.getMonitor(history));
      iterativeTrainer.setTimeout(30, TimeUnit.SECONDS);
      iterativeTrainer.setIterationsPerSample(100);
      iterativeTrainer.setMaxIterations(250);
      iterativeTrainer.setTerminateThreshold(0);
      return iterativeTrainer.run();
    } finally {
      iterativeTrainer.freeRef();
    }
Logging
Reset training subject: 893653733162
Reset training subject: 893656437635
Adding measurement 5b2f4e89 to history. Total: 0
LBFGS Accumulation History: 1 points
Constructing line search parameters: GD
Non-optimal measurement 528.8293711151897 < 528.8293711151897. Total: 1
th(0)=528.8293711151897;dx=-5.43112448E24
Adding measurement 7776074e to history. Total: 1
New Minimum: 528.8293711151897 > 0.04544361193472045
Armijo: th(2.154434690031884)=0.04544361193472045; dx=-5.431124480018979E12 evalInputDelta=528.783927503255
Non-optimal measurement 0.13207064317807687 < 0.04544361193472045. Total: 2
Armijo: th(1.077217345015942)=0.13207064317807687; dx=-5.431124480019014E12 evalInputDelta=528.6973004720116
Non-optimal measurement 1.3220433701289842 < 0.04544361193472045. Total: 2
Armijo: th(0.3590724483386473)=1.3220433701289842; dx=-5.431124480020143E12 evalInputDelta=527.5073277450607
Non-optimal measurement 4.747368291706443 < 0.04544361193472045. Total: 2
Armijo: th(0.08976811208466183)=4.747368291706443; dx=-5.431124480035212E12 evalInputDelta=524.0820028234832
Non-optimal measurement 8.541434805177051 < 0.04544361193472045. Total: 2
Armijo: th(0.017953622416932366)=8.541434805177051; dx=-5.43112448009962E12 evalInputDelta=520.2879363100127
Non-optimal measurement 10.607012226712174 < 0.04544361193472045. Total: 2
Armijo: th(0.002992270402822061)=10.607012226712174; dx=-5.431124480214101E12 evalInputDelta=518.2223588884775
Non-optimal measurement 11.231022614624631 < 0.04544361193472045. Total: 2
Armijo: th(4.2746720040315154E-4)=11.231022614624631; dx=-5.431124480284546E12 evalInputDelta=517.5983485005651
Non-optimal measurement 11.339602215220847 < 0.04544361193472045. Total: 2
Armijo: th(5.343340005039394E-5)=11.339602215220847; dx=-5.43112448030026E12 evalInputDelta=517.4897688999689
Non-optimal measurement 11.35382284632942 < 0.04544361193472045. Total: 2
Armijo: th(5.9370444500437714E-6)=11.35382284632942; dx=-5.431124480302417E12 evalInputDelta=517.4755482688603
Non-optimal measurement 11.355429102301345 < 0.04544361193472045. Total: 2
Armijo: th(5.937044450043771E-7)=11.355429102301345; dx=-5.431124480302662E12 evalInputDelta=517.4739420128883
Non-optimal measurement 11.35559142324907 < 0.04544361193472045. Total: 2
Armijo: th(5.397313136403428E-8)=11.35559142324907; dx=-5.4311244803026875E12 evalInputDelta=517.4737796919406
Non-optimal measurement 11.355606303338789 < 0.04544361193472045. Total: 2
Armijo: th(4.4977609470028565E-9)=11.355606303338789; dx=-5.431124480302689E12 evalInputDelta=517.473764811851
Non-optimal measurement 11.359937918203105 < 0.04544361193472045. Total: 2
Armijo: th(3.4598161130791205E-10)=11.359937918203105; dx=-5.431161073070346E12 evalInputDelta=517.4694331969866
Non-optimal measurement 18.069740830470938 < 0.04544361193472045. Total: 2
Armijo: th(2.4712972236279432E-11)=18.069740830470938; dx=-6.972800005440111E21 evalInputDelta=510.75963028471875
Non-optimal measurement 392.0001701056701 < 0.04544361193472045. Total: 2
Armijo: th(1.6475314824186289E-12)=392.0001701056701; dx=-3.7142777600082325E24 evalInputDelta=136.82920100951958
Non-optimal measurement 494.3126824546316 < 0.04544361193472045. Total: 2
Armijo: th(1.029707176511643E-13)=494.3126824546316; dx=-4.881745280049965E24 evalInputDelta=34.51668866055809
Non-optimal measurement 528.8293711151879 < 0.04544361193472045. Total: 2
Armijo: th(6.057101038303783E-15)=528.8293711151879; dx=-5.43112448E24 evalInputDelta=1.8189894035458565E-12
Non-optimal measurement 0.04544361193472045 < 0.04544361193472045. Total: 2
MIN ALPHA (3.3650561323909904E-16): th(2.154434690031884)=0.04544361193472045
Fitness changed from 528.8293711151897 to 0.04544361193472045
Iteration 1 complete. Error: 0.04544361193472045 Total: 0.1253; Orientation: 0.0040; Line Search: 0.1141
Non-optimal measurement 0.04544361193472045 < 0.04544361193472045. Total: 2
LBFGS Accumulation History: 2 points
Non-optimal measurement 0.04544361193472045 < 0.04544361193472045. Total: 2
th(0)=0.04544361193472045;dx=-9.462489633836965
Adding measurement 4090f961 to history. Total: 2
New Minimum: 0.04544361193472045 > 0.04544361193472043
WOLFE (weak): th(2.154434690031884E-15)=0.04544361193472043; dx=-9.462489633836965 evalInputDelta=2.0816681711721685E-17
Adding measurement 6990c655 to history. Total: 3
New Minimum: 0.04544361193472043 > 0.04544361193472041
WOLFE (weak): th(4.308869380063768E-15)=0.04544361193472041; dx=-9.462489633836965 evalInputDelta=4.163336342344337E-17
Adding measurement 4eadd1e4 to history. Total: 4
New Minimum: 0.04544361193472041 > 0.04544361193472034
WOLFE (weak): th(1.2926608140191303E-14)=0.04544361193472034; dx=-9.462489633836965 evalInputDelta=1.1102230246251565E-16
Adding measurement 68abefb7 to history. Total: 5
New Minimum: 0.04544361193472034 > 0.04544361193472
WOLFE (weak): th(5.1706432560765214E-14)=0.04544361193472; dx=-9.462489633836965 evalInputDelta=4.510281037539698E-16
Adding measurement 185743e2 to history. Total: 6
New Minimum: 0.04544361193472 > 0.04544361193471817
WOLFE (weak): th(2.5853216280382605E-13)=0.04544361193471817; dx=-9.462489633836965 evalInputDelta=2.275957200481571E-15
Adding measurement 5432616 to history. Total: 7
New Minimum: 0.04544361193471817 > 0.04544361193470677
WOLFE (weak): th(1.5511929768229563E-12)=0.04544361193470677; dx=-9.462489633836963 evalInputDelta=1.3676559884601147E-14
Adding measurement 5084d297 to history. Total: 8
New Minimum: 0.04544361193470677 > 0.04544361193462471
WOLFE (weak): th(1.0858350837760695E-11)=0.04544361193462471; dx=-9.462489633836956 evalInputDelta=9.573591919220803E-14
Adding measurement 6d7d8f54 to history. Total: 9
New Minimum: 0.04544361193462471 > 0.04544361193395451
WOLFE (weak): th(8.686680670208556E-11)=0.04544361193395451; dx=-9.462489633836901 evalInputDelta=7.659359257949916E-13
Adding measurement 6b53cff2 to history. Total: 10
New Minimum: 0.04544361193395451 > 0.04544361192782706
WOLFE (weak): th(7.8180126031877E-10)=0.04544361192782706; dx=-9.462489633836391 evalInputDelta=6.893388637685405E-12
Adding measurement 48f31b43 to history. Total: 11
New Minimum: 0.04544361192782706 > 0.045443611865786576
WOLFE (weak): th(7.818012603187701E-9)=0.045443611865786576; dx=-9.462489633831225 evalInputDelta=6.893387249906624E-11
Adding measurement 212605f9 to history. Total: 12
New Minimum: 0.045443611865786576 > 0.04544361117644787
WOLFE (weak): th(8.599813863506471E-8)=0.04544361117644787; dx=-9.462489633773824 evalInputDelta=7.582725766730469E-10
Adding measurement 3270bb4c to history. Total: 13
New Minimum: 0.04544361117644787 > 0.04544360283544991
WOLFE (weak): th(1.0319776636207765E-6)=0.04544360283544991; dx=-9.462489633079281 evalInputDelta=9.099270538437398E-9
Adding measurement 7163a429 to history. Total: 14
New Minimum: 0.04544360283544991 > 0.045443493644264386
WOLFE (weak): th(1.3415709627070094E-5)=0.045443493644264386; dx=-9.462489623987086 evalInputDelta=1.1829045606231992E-7
Adding measurement 103f668a to history. Total: 15
New Minimum: 0.045443493644264386 > 0.04544195588036056
WOLFE (weak): th(1.878199347789813E-4)=0.04544195588036056; dx=-9.462489495940819 evalInputDelta=1.6560543598873045E-6
Adding measurement 4e04252 to history. Total: 16
New Minimum: 0.04544195588036056 > 0.0454187738383449
WOLFE (weak): th(0.0028172990216847197)=0.0454187738383449; dx=-9.462487565879371 evalInputDelta=2.4838096375545682E-5
Adding measurement 520af119 to history. Total: 17
New Minimum: 0.0454187738383449 > 0.04504689971110515
WOLFE (weak): th(0.045076784346955515)=0.04504689971110515; dx=-9.462456670613303 evalInputDelta=3.9671222361530034E-4
Adding measurement f92085f to history. Total: 18
New Minimum: 0.04504689971110515 > 0.038893106300604316
WOLFE (weak): th(0.7663053338982437)=0.038893106300604316; dx=-9.461962897278429 evalInputDelta=0.006550505634116133
Adding measurement 6a7ee60c to history. Total: 19
New Minimum: 0.038893106300604316 > 0.0036735237306089906
WOLFE (weak): th(13.793496010168386)=0.0036735237306089906; dx=-9.45990705356817 evalInputDelta=0.04177008820411146
Adding measurement 7449365a to history. Total: 20
New Minimum: 0.0036735237306089906 > 0.0
WOLFE (weak): th(262

...skipping 9039 bytes...

-93a4-79fe72795755 = 1.000/1.000e+00, d75036a5-caba-48fc-af34-526d0d7d3fbb = 1.000/1.000e+00, 10f9c165-d432-4518-b5d9-ef01fdbc3039 = 1.000/1.000e+00, a7e21048-8e0b-4bd6-81ef-d1b96523edda = 1.000/1.000e+00, ca7da342-ac8d-45e7-9234-aeee76881c49 = 1.000/1.000e+00]
Orientation rejected. Popping history element from 0.0, 0.0036735237306089906, 0.038893106300604316, 0.04504689971110515, 0.0454187738383449, 0.04544195588036056, 0.045443493644264386, 0.04544360283544991, 0.04544361117644787, 0.045443611865786576, 0.04544361192782706
Rejected: LBFGS Orientation magnitude: 6.504e+04, gradient 3.075e+00, dot -0.954; [a7e21048-8e0b-4bd6-81ef-d1b96523edda = 1.000/1.000e+00, ca7da342-ac8d-45e7-9234-aeee76881c49 = 1.000/1.000e+00, d75036a5-caba-48fc-af34-526d0d7d3fbb = 1.000/1.000e+00, 52617d35-c501-4cfc-93a4-79fe72795755 = 1.000/1.000e+00, 10f9c165-d432-4518-b5d9-ef01fdbc3039 = 1.000/1.000e+00]
Orientation rejected. Popping history element from 0.0, 0.0036735237306089906, 0.038893106300604316, 0.04504689971110515, 0.0454187738383449, 0.04544195588036056, 0.045443493644264386, 0.04544360283544991, 0.04544361117644787, 0.045443611865786576
Rejected: LBFGS Orientation magnitude: 6.504e+04, gradient 3.075e+00, dot -0.954; [d75036a5-caba-48fc-af34-526d0d7d3fbb = 1.000/1.000e+00, ca7da342-ac8d-45e7-9234-aeee76881c49 = 1.000/1.000e+00, 10f9c165-d432-4518-b5d9-ef01fdbc3039 = 1.000/1.000e+00, 52617d35-c501-4cfc-93a4-79fe72795755 = 1.000/1.000e+00, a7e21048-8e0b-4bd6-81ef-d1b96523edda = 1.000/1.000e+00]
Orientation rejected. Popping history element from 0.0, 0.0036735237306089906, 0.038893106300604316, 0.04504689971110515, 0.0454187738383449, 0.04544195588036056, 0.045443493644264386, 0.04544360283544991, 0.04544361117644787
Rejected: LBFGS Orientation magnitude: 6.504e+04, gradient 3.075e+00, dot -0.954; [ca7da342-ac8d-45e7-9234-aeee76881c49 = 1.000/1.000e+00, 52617d35-c501-4cfc-93a4-79fe72795755 = 1.000/1.000e+00, a7e21048-8e0b-4bd6-81ef-d1b96523edda = 1.000/1.000e+00, d75036a5-caba-48fc-af34-526d0d7d3fbb = 1.000/1.000e+00, 10f9c165-d432-4518-b5d9-ef01fdbc3039 = 1.000/1.000e+00]
Orientation rejected. Popping history element from 0.0, 0.0036735237306089906, 0.038893106300604316, 0.04504689971110515, 0.0454187738383449, 0.04544195588036056, 0.045443493644264386, 0.04544360283544991
Rejected: LBFGS Orientation magnitude: 8.854e+04, gradient 3.075e+00, dot -0.936; [d75036a5-caba-48fc-af34-526d0d7d3fbb = 1.000/1.000e+00, 52617d35-c501-4cfc-93a4-79fe72795755 = 1.000/1.000e+00, 10f9c165-d432-4518-b5d9-ef01fdbc3039 = 1.000/1.000e+00, ca7da342-ac8d-45e7-9234-aeee76881c49 = 1.000/1.000e+00, a7e21048-8e0b-4bd6-81ef-d1b96523edda = 1.000/1.000e+00]
Orientation rejected. Popping history element from 0.0, 0.0036735237306089906, 0.038893106300604316, 0.04504689971110515, 0.0454187738383449, 0.04544195588036056, 0.045443493644264386
Rejected: LBFGS Orientation magnitude: 7.916e+04, gradient 3.075e+00, dot -1.000; [52617d35-c501-4cfc-93a4-79fe72795755 = 1.000/1.000e+00, a7e21048-8e0b-4bd6-81ef-d1b96523edda = 1.000/1.000e+00, 10f9c165-d432-4518-b5d9-ef01fdbc3039 = 1.000/1.000e+00, ca7da342-ac8d-45e7-9234-aeee76881c49 = 1.000/1.000e+00, d75036a5-caba-48fc-af34-526d0d7d3fbb = 1.000/1.000e+00]
Orientation rejected. Popping history element from 0.0, 0.0036735237306089906, 0.038893106300604316, 0.04504689971110515, 0.0454187738383449, 0.04544195588036056
Rejected: LBFGS Orientation magnitude: 7.961e+04, gradient 3.075e+00, dot -1.000; [ca7da342-ac8d-45e7-9234-aeee76881c49 = 1.000/1.000e+00, 52617d35-c501-4cfc-93a4-79fe72795755 = 1.000/1.000e+00, 10f9c165-d432-4518-b5d9-ef01fdbc3039 = 1.000/1.000e+00, d75036a5-caba-48fc-af34-526d0d7d3fbb = 1.000/1.000e+00, a7e21048-8e0b-4bd6-81ef-d1b96523edda = 1.000/1.000e+00]
Orientation rejected. Popping history element from 0.0, 0.0036735237306089906, 0.038893106300604316, 0.04504689971110515, 0.0454187738383449
Rejected: LBFGS Orientation magnitude: 8.499e+04, gradient 3.075e+00, dot -1.000; [a7e21048-8e0b-4bd6-81ef-d1b96523edda = 1.000/1.000e+00, 10f9c165-d432-4518-b5d9-ef01fdbc3039 = 1.000/1.000e+00, 52617d35-c501-4cfc-93a4-79fe72795755 = 1.000/1.000e+00, ca7da342-ac8d-45e7-9234-aeee76881c49 = 1.000/1.000e+00, d75036a5-caba-48fc-af34-526d0d7d3fbb = 1.000/1.000e+00]
Orientation rejected. Popping history element from 0.0, 0.0036735237306089906, 0.038893106300604316, 0.04504689971110515
LBFGS Accumulation History: 3 points
Removed measurement 7449365a to history. Total: 20
Removed measurement 6a7ee60c to history. Total: 19
Removed measurement f92085f to history. Total: 18
Removed measurement 520af119 to history. Total: 17
Removed measurement 4e04252 to history. Total: 16
Removed measurement 103f668a to history. Total: 15
Removed measurement 7163a429 to history. Total: 14
Removed measurement 3270bb4c to history. Total: 13
Removed measurement 212605f9 to history. Total: 12
Removed measurement 48f31b43 to history. Total: 11
Removed measurement 6b53cff2 to history. Total: 10
Removed measurement 6d7d8f54 to history. Total: 9
Removed measurement 5084d297 to history. Total: 8
Removed measurement 5432616 to history. Total: 7
Removed measurement 185743e2 to history. Total: 6
Removed measurement 68abefb7 to history. Total: 5
Removed measurement 4eadd1e4 to history. Total: 4
Removed measurement 6990c655 to history. Total: 3
Adding measurement 7c29a99d to history. Total: 3
th(0)=0.0;dx=-9.458297600000002
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(10328.695646543105)=0.0; dx=-9.458297600000002 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(5164.347823271552)=0.0; dx=-9.458297600000002 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(1721.4492744238507)=0.0; dx=-9.458297600000002 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(430.3623186059627)=0.0; dx=-9.458297600000002 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(86.07246372119253)=0.0; dx=-9.458297600000002 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(14.345410620198756)=0.0; dx=-9.458297600000002 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(2.0493443743141078)=0.0; dx=-9.458297600000002 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(0.25616804678926347)=0.0; dx=-9.458297600000002 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(0.028463116309918164)=0.0; dx=-9.458297600000002 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(0.0028463116309918163)=0.0; dx=-9.458297600000002 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(2.5875560281743784E-4)=0.0; dx=-9.458297600000002 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(2.1562966901453154E-5)=0.0; dx=-9.458297600000002 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(1.6586897616502425E-6)=0.0; dx=-9.458297600000002 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(1.1847784011787447E-7)=0.0; dx=-9.458297600000002 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(7.898522674524965E-9)=0.0; dx=-9.458297600000002 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(4.936576671578103E-10)=0.0; dx=-9.458297600000002 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(2.9038686303400606E-11)=0.0; dx=-9.458297600000002 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(1.6132603501889225E-12)=0.0; dx=-9.458297600000002 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(8.49084394836275E-14)=0.0; dx=-9.458297600000002 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(4.2454219741813745E-15)=0.0; dx=-9.458297600000002 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
MIN ALPHA (2.0216295115149403E-16): th(0.0)=0.0
Fitness changed from 0.0 to 0.0
Static Iteration Total: 10.8065; Orientation: 10.4994; Line Search: 0.3041
Iteration 3 failed. Error: 0.0
Previous Error: 0.0 -> 0.0
Optimization terminated 3
Final threshold in iteration 3: 0.0 (> 0.0) after 11.382s (< 30.000s)

Returns

    0.0

Training Converged

TrainingTester.java:432 executed in 0.10 seconds (0.000 gc):

    return TestUtil.compare(title + " vs Iteration", runs);
Logging
Plotting range=[1.0, -2.3425271574691937], [2.0, -0.3425271574691935]; valueStats=DoubleSummaryStatistics{count=2, sum=0.090887, min=0.045444, average=0.045444, max=0.045444}
Plotting 2 points for GD
Only 1 points for CjGD
Plotting 2 points for LBFGS

Returns

Result

TrainingTester.java:435 executed in 0.01 seconds (0.000 gc):

    return TestUtil.compareTime(title + " vs Time", runs);
Logging
Plotting range=[0.0, -2.3425271574691937], [0.45, -0.3425271574691935]; valueStats=DoubleSummaryStatistics{count=2, sum=0.090887, min=0.045444, average=0.045444, max=0.045444}
Plotting 2 points for GD
Only 1 points for CjGD
Plotting 2 points for LBFGS

Returns

Result

Results

TrainingTester.java:255 executed in 0.00 seconds (0.000 gc):

    return grid(inputLearning, modelLearning, completeLearning);

Returns

Result

TrainingTester.java:258 executed in 0.00 seconds (0.000 gc):

    return new ComponentResult(null == inputLearning ? null : inputLearning.value,
        null == modelLearning ? null : modelLearning.value, null == completeLearning ? null : completeLearning.value);

Returns

    {"input":{ "LBFGS": { "type": "Converged", "value": 0.0 }, "CjGD": { "type": "Converged", "value": 0.0 }, "GD": { "type": "Converged", "value": 0.0 } }, "model":null, "complete":null}

LayerTests.java:425 executed in 0.00 seconds (0.000 gc):

    throwException(exceptions.addRef());

Results

detailsresult
{"input":{ "LBFGS": { "type": "Converged", "value": 0.0 }, "CjGD": { "type": "Converged", "value": 0.0 }, "GD": { "type": "Converged", "value": 0.0 } }, "model":null, "complete":null}OK
  {
    "result": "OK",
    "performance": {
      "execution_time": "14.264",
      "gc_time": "0.358"
    },
    "created_on": 1586735479341,
    "file_name": "trainingTest",
    "report": {
      "simpleName": "Basic",
      "canonicalName": "com.simiacryptus.mindseye.layers.java.ImgZeroPaddingLayerTest.Basic",
      "link": "https://github.com/SimiaCryptus/mindseye-java/tree/93db34cedee48c0202777a2b25deddf1dfaf5731/src/test/java/com/simiacryptus/mindseye/layers/java/ImgZeroPaddingLayerTest.java",
      "javaDoc": ""
    },
    "training_analysis": {
      "input": {
        "LBFGS": {
          "type": "Converged",
          "value": 0.0
        },
        "CjGD": {
          "type": "Converged",
          "value": 0.0
        },
        "GD": {
          "type": "Converged",
          "value": 0.0
        }
      }
    },
    "archive": "s3://code.simiacrypt.us/tests/com/simiacryptus/mindseye/layers/java/ImgZeroPaddingLayer/Basic/trainingTest/202004125119",
    "id": "9e651b74-8ef8-42fa-a34d-307c6225cf6d",
    "report_type": "Components",
    "display_name": "Comparative Training",
    "target": {
      "simpleName": "ImgZeroPaddingLayer",
      "canonicalName": "com.simiacryptus.mindseye.layers.java.ImgZeroPaddingLayer",
      "link": "https://github.com/SimiaCryptus/mindseye-java/tree/93db34cedee48c0202777a2b25deddf1dfaf5731/src/main/java/com/simiacryptus/mindseye/layers/java/ImgZeroPaddingLayer.java",
      "javaDoc": ""
    }
  }