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 3451881638442288128

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

    [
    	[ [ -1.688 ], [ 0.3 ], [ 0.048 ], [ -1.616 ], [ -0.032 ], [ -0.176 ], [ 1.704 ], [ 1.368 ] ],
    	[ [ 1.108 ], [ 0.08 ], [ -0.876 ], [ -1.476 ], [ 0.392 ], [ -1.856 ], [ 1.912 ], [ -1.516 ] ],
    	[ [ 0.148 ], [ -0.408 ], [ 1.556 ], [ -0.464 ], [ -0.804 ], [ -0.384 ], [ -1.572 ], [ 0.788 ] ],
    	[ [ -1.54 ], [ 1.552 ], [ -0.636 ], [ -0.556 ], [ -1.832 ], [ -0.128 ], [ -0.068 ], [ 1.048 ] ],
    	[ [ 1.032 ], [ 0.092 ], [ 1.356 ], [ 0.636 ], [ -0.712 ], [ 1.208 ], [ -0.768 ], [ -0.384 ] ],
    	[ [ 0.496 ], [ -0.012 ], [ 0.028 ], [ 1.764 ], [ 1.556 ], [ 1.652 ], [ -1.72 ], [ 1.612 ] ],
    	[ [ -0.892 ], [ -0.852 ], [ -1.028 ], [ -1.424 ], [ -0.608 ], [ -1.228 ], [ -0.804 ], [ 1.524 ] ],
    	[ [ 0.048 ], [ 1.62 ], [ 1.64 ], [ -1.492 ], [ 1.876 ], [ 0.7 ], [ 1.512 ], [ 0.996 ] ]
    ]
    [
    	[ [ 0.82 ], [ -0.368 ], [ 0.644 ], [ -1.664 ], [ -1.176 ], [ 0.012 ], [ -0.968 ], [ 1.628 ] ],
    	[ [ -1.456 ], [ 1.956 ], [ -1.248 ], [ -1.16 ], [ 1.352 ], [ -1.256 ], [ -1.76 ], [ -0.784 ] ],
    	[ [ -0.316 ], [ 0.688 ], [ -0.312 ], [ -1.16 ], [ 1.324 ], [ 1.916 ], [ -1.724 ], [ -0.472 ] ],
    	[ [ -1.1 ], [ -0.124 ], [ 0.344 ], [ -0.628 ], [ -1.564 ], [ 1.24 ], [ 0.52 ], [ -0.808 ] ],
    	[ [ -1.656 ], [ 1.288 ], [ -1.484 ], [ 1.98 ], [ 0.972 ], [ 0.692 ], [ 1.776 ], [ 1.444 ] ],
    	[ [ -1.808 ], [ -2.0 ], [ 0.016 ], [ 0.812 ], [ -1.552 ], [ -1.58 ], [ -0.504 ], [ 1.42 ] ],
    	[ [ -1.256 ], [ -0.892 ], [ 0.66 ], [ 0.52 ], [ 0.82 ], [ 1.628 ], [ -0.888 ], [ -1.116 ] ],
    	[ [ -1.156 ], [ -1.764 ], [ 0.56 ], [ 0.184 ], [ 1.156 ], [ -0.968 ], [ -1.34 ], [ -0.856 ] ]
    ]
    [
    	[ [ -1.832 ], [ -0.712 ], [ -1.616 ], [ 1.524 ], [ 0.048 ], [ 0.08 ], [ 1.64 ], [ 1.368 ] ],
    	[ [ -0.636 ], [ -1.856 ], [ -0.852 ], [ -1.476 ], [ 0.7 ], [ -0.804 ], [ -0.384 ], [ 1.62 ] ],
    	[ [ -0.892 ], [ -0.408 ], [ -1.688 ], [ 1.612 ], [ 1.876 ], [ -0.384 ], [ -1.516 ], [ 1.912 ] ],
    	[ [ -1.54 ], [ 0.496 ], [ -1.492 ], [ -0.128 ], [ 1.556 ], [ 1.512 ], [ 0.028 ], [ 1.048 ] ],
    	[ [ -0.068 ], [ -1.72 ], [ -0.556 ], [ 0.788 ], [ -0.032 ], [ 1.764 ], [ -1.028 ], [ 1.652 ] ],
    	[ [ -0.176 ], [ 0.3 ], [ -0.464 ], [ 1.032 ], [ 1.552 ], [ -0.608 ], [ -0.768 ], [ -0.012 ] ],
    	[ [ 1.108 ], [ 0.996 ], [ 1.208 ], [ 0.148 ], [ -1.424 ], [ -0.804 ], [ -1.572 ], [ -1.228 ] ],
    	[ [ 0.392 ], [ 0.048 ], [ 0.636 ], [ 1.556 ], [ 0.092 ], [ 1.356 ], [ 1.704 ], [ -0.876 ] ]
    ]
    [
    	[ [ -0.968 ], [ 1.916 ], [ 0.56 ], [ -1.764 ], [ -1.16 ], [ 1.956 ], [ 1.324 ], [ 0.82 ] ],
    	[ [ 0.972 ], [ -0.316 ], [ 1.628 ], [ 1.352 ], [ 0.52 ], [ 0.644 ], [ 1.156 ], [ -1.76 ] ],
    	[ [ 1.98 ], [ 1.24 ], [ -1.564 ], [ -1.156 ], [ 1.444 ], [ -0.368 ], [ -1.456 ], [ 0.82 ] ],
    	[ [ -1.16 ], [ -1.552 ], [ 1.288 ], [ -1.256 ], [ -1.656 ], [ 0.66 ], [ 0.016 ], [ 0.688 ] ],
    	[ [ 0.692 ], [ -0.628 ], [ -0.312 ], [ -1.664 ], [ 0.184 ], [ -1.724 ], [ -1.176 ], [ -0.784 ] ],
    	[ [ 0.52 ], [ 0.344 ], [ -1.248 ], [ -1.484 ], [ -1.256 ], [ -1.116 ], [ 1.42 ], [ -0.856 ] ],
    	[ [ -0.808 ], [ -0.124 ], [ -0.968 ], [ 0.012 ], [ 1.776 ], [ -1.34 ], [ -0.888 ], [ -0.504 ] ],
    	[ [ -0.892 ], [ 1.628 ], [ -1.808 ], [ -0.472 ], [ -2.0 ], [ 0.812 ], [ -1.1 ], [ -1.58 ] ]
    ]
    [
    	[ [ -0.852 ], [ 0.636 ], [ -0.636 ], [ -0.804 ], [ -0.128 ], [ -1.832 ], [ 1.032 ], [ 1.368 ] ],
    	[ [ -0.068 ], [ 1.64 ], [ -1.72 ], [ -1.424 ], [ 1.652 ], [ 0.7 ], [ 1.512 ], [ 1.704 ] ],
    	[ [ -0.892 ], [ -0.176 ], [ 1.764 ], [ -1.028 ], [ -0.384 ], [ 0.496 ], [ 0.08 ], [ -0.464 ] ],
    	[ [ 1.108 ], [ -1.476 ], [ 1.048 ], [ 0.3 ], [ 1.556 ], [ 0.028 ], [ -1.54 ], [ -0.712 ] ],
    	[ [ 1.552 ], [ 1.612 ], [ 0.392 ], [ 0.048 ], [ -0.032 ], [ -1.856 ], [ 1.876 ], [ 0.092 ] ],
    	[ [ -0.556 ], [ 1.524 ], [ 1.208 ], [ -1.616 ], [ -0.408 ], [ 1.912 ], [ 0.996 ], [ -0.768 ] ],
    	[ [ -1.516 ], [ -1.572 ], [ -1.492 ], [ 0.788 ], [ -0.012 ], [ 0.048 ], [ -0.804 ], [ 1.62 ] ],
    	[ [ 1.556 ], [ -1.228 ], [ 1.356 ], [ -0.876 ], [ 0.148 ], [ -0.384 ], [ -0.608 ], [ -1.688 ] ]
    ]
    [
    	[ [ -1.456 ], [ 0.344 ], [ -0.124 ], [ -1.116 ], [ 0.692 ], [ 1.24 ], [ -1.248 ], [ -1.256 ] ],
    	[ [ 1.956 ], [ 0.184 ], [ 1.628 ], [ -1.16 ], [ -0.628 ], [ -0.968 ], [ -0.808 ], [ 0.012 ] ],
    	[ [ 1.628 ], [ 1.288 ], [ -1.34 ], [ 1.324 ], [ 1.776 ], [ 1.916 ], [ 0.016 ], [ -1.156 ] ],
    	[ [ -1.724 ], [ -1.1 ], [ -0.316 ], [ -0.312 ], [ -1.256 ], [ 0.52 ], [ 0.644 ], [ -1.656 ] ],
    	[ [ 1.352 ], [ 1.156 ], [ -0.368 ], [ -1.58 ], [ -1.808 ], [ 1.98 ], [ -0.784 ], [ -0.504 ] ],
    	[ [ -1.664 ], [ -0.856 ], [ -0.892 ], [ 0.688 ], [ -1.16 ], [ -2.0 ], [ 0.52 ], [ -1.552 ] ],
    	[ [ -1.564 ], [ -0.968 ], [ 0.82 ], [ 0.66 ], [ 1.42 ], [ -1.764 ], [ -0.472 ], [ 0.812 ] ],
    	[ [ 0.56 ], [ -1.484 ], [ 1.444 ], [ -0.888 ], [ -1.76 ], [ 0.82 ], [ -1.176 ], [ 0.972 ] ]
    ]
    [
    	[ [ 1.612 ], [ 1.64 ], [ 0.636 ], [ 1.556 ], [ -0.876 ], [ 1.556 ], [ 1.876 ], [ -1.028 ] ],
    	[ [ -1.72 ], [ 1.912 ], [ -0.464 ], [ -1.616 ], [ 0.148 ], [ -0.128 ], [ -1.832 ], [ -1.856 ] ],
    	[ [ 1.764 ], [ -0.804 ], [ -0.768 ], [ -0.032 ], [ -1.424 ], [ -0.892 ], [ 0.048 ], [ 1.62 ] ],
    	[ [ -0.852 ], [ -0.176 ], [ -0.012 ], [ 0.3 ], [ -0.384 ], [ 1.512 ], [ 0.7 ], [ 0.788 ] ],
    	[ [ 1.356 ], [ 1.524 ], [ -0.384 ], [ 0.092 ], [ 1.048 ], [ 0.496 ], [ 0.996 ], [ 1.208 ] ],
    	[ [ 0.08 ], [ 1.368 ], [ -1.492 ], [ 0.392 ], [ -1.572 ], [ 1.704 ], [ -1.688 ], [ -0.556 ] ],
    	[ [ -1.54 ], [ 0.048 ], [ -0.068 ], [ 1.108 ], [ -0.408 ], [ -1.228 ], [ 1.032 ], [ 1.552 ] ],
    	[ [ -0.636 ], [ -0.608 ], [ -1.476 ], [ -1.516 ], [ -0.712 ], [ 1.652 ], [ 0.028 ], [ -0.804 ] ]
    ]
    [
    	[ [ 0.56 ], [ -1.1 ], [ -1.484 ], [ 0.344 ], [ -1.552 ], [ -1.456 ], [ -0.968 ], [ -1.76 ] ],
    	[ [ 0.66 ], [ -0.808 ], [ 1.156 ], [ 1.352 ], [ -0.888 ], [ -1.256 ], [ -1.248 ], [ -2.0 ] ],
    	[ [ -1.176 ], [ 1.776 ], [ -1.116 ], [ 0.016 ], [ 0.812 ], [ -1.808 ], [ -0.472 ], [ 1.288 ] ],
    	[ [ 0.82 ], [ 0.688 ], [ -0.784 ], [ -1.724 ], [ -0.124 ], [ -0.312 ], [ -0.504 ], [ -0.316 ] ],
    	[ [ 1.628 ], [ 0.972 ], [ 0.012 ], [ 1.444 ], [ 1.24 ], [ -0.892 ], [ 1.42 ], [ 0.52 ] ],
    	[ [ 1.956 ], [ -1.58 ], [ -1.16 ], [ 0.184 ], [ 1.916 ], [ -0.368 ], [ -1.16 ], [ -1.256 ] ],
    	[ [ -1.664 ], [ 1.98 ], [ -1.656 ], [ 0.644 ], [ -1.764 ], [ -1.34 ], [ -0.968 ], [ 0.52 ] ],
    	[ [ -1.156 ], [ 0.82 ], [ 1.628 ], [ -0.856 ], [ 1.324 ], [ 0.692 ], [ -0.628 ], [ -1.564 ] ]
    ]
    [
    	[ [ 1.704 ], [ -1.228 ], [ -1.832 ], [ -0.608 ], [ 0.092 ], [ -1.616 ], [ 0.392 ], [ -0.032 ] ],
    	[ [ 0.3 ], [ -0.892 ], [ -0.384 ], [ 1.556 ], [ 0.148 ], [ -0.556 ], [ -0.712 ], [ 1.912 ] ],
    	[ [ -1.028 ], [ -1.72 ], [ 1.032 ], [ 1.764 ], [ 0.028 ], [ 0.788 ], [ 0.048 ], [ -0.068 ] ],
    	[ [ 0.496 ], [ -1.54 ], [ 1.64 ], [ -0.852 ], [ 0.996 ], [ -1.516 ], [ -0.128 ], [ 1.208 ] ],
    	[ [ 1.048 ], [ 0.048 ], [ 1.108 ], [ -1.424 ], [ -0.804 ], [ -0.464 ], [ -1.856 ], [ -0.384 ] ],
    	[ [ -1.476 ], [ -0.176 ], [ 1.652 ], [ 1.512 ], [ 0.636 ], [ 1.62 ], [ -0.876 ], [ -0.804 ] ],
    	[ [ 1.524 ], [ 1.552 ], [ -1.572 ], [ 1.876 ], [ -0.636 ], [ 0.08 ], [ 1.612 ], [ -1.688 ] ],
    	[ [ -0.012 ], [ 1.356 ], [ -0.408 ], [ 1.556 ], [ 1.368 ], [ -1.492 ], [ -0.768 ], [ 0.7 ] ]
    ]
    [
    	[ [ 1.628 ], [ -1.724 ], [ 1.916 ], [ -1.16 ], [ 0.688 ], [ -1.552 ], [ 1.444 ], [ -0.808 ] ],
    	[ [ -0.312 ], [ 0.52 ], [ 1.42 ], [ -0.368 ], [ 1.776 ], [ -1.1 ], [ -0.892 ], [ -1.808 ] ],
    	[ [ -1.664 ], [ 1.352 ], [ -0.472 ], [ 0.344 ], [ 0.66 ], [ 0.016 ], [ -0.124 ], [ -1.656 ] ],
    	[ [ -0.856 ], [ 0.56 ], [ -0.968 ], [ 1.288 ], [ 0.812 ], [ 1.956 ], [ -1.76 ], [ 1.24 ] ],
    	[ [ -1.116 ], [ 0.972 ], [ -0.504 ], [ -1.248 ], [ -1.764 ], [ -1.564 ], [ -2.0 ], [ 1.98 ] ],
    	[ [ 0.82 ], [ -0.968 ], [ -0.784 ], [ -0.888 ], [ -1.456 ], [ 0.012 ], [ -1.156 ], [ 1.324 ] ],
    	[ [ 0.184 ], [ 0.644 ], [ 1.628 ], [ -1.176 ], [ -1.484 ], [ 0.692 ], [ -1.16 ], [ 0.82 ] ],
    	[ [ -1.34 ], [ -1.256 ], [ -1.58 ], [ 0.52 ], [ -0.628 ], [ -1.256 ], [ 1.156 ], [ -0.316 ] ]
    ]

Gradient Descent

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

TrainingTester.java:480 executed in 1.12 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: 2516829331339
BACKPROP_AGG_SIZE = 3
THREADS = 64
SINGLE_THREADED = false
Initialized CoreSettings = {
"backpropAggregationSize" : 3,
"jvmThreads" : 64,
"singleThreaded" : false
}
Reset training subject: 2516870419568
Constructing line search parameters: GD
th(0)=917.3778512961144;dx=-9.1101792E24
New Minimum: 917.3778512961144 > 0.06185868047315239
Armijo: th(2.154434690031884)=0.06185868047315239; dx=-9.110179200021508E12 evalInputDelta=917.3159926156412
Armijo: th(1.077217345015942)=0.3475817674561618; dx=-9.110179200021598E12 evalInputDelta=917.0302695286582
Armijo: th(0.3590724483386473)=2.1421277620337946; dx=-9.110179200023053E12 evalInputDelta=915.2357235340805
Armijo: th(0.08976811208466183)=6.6303280548113985; dx=-9.11017920004115E12 evalInputDelta=910.747523241303
Armijo: th(0.017953622416932366)=10.542443179667108; dx=-9.110179200099678E12 evalInputDelta=906.8354081164473
Armijo: th(0.002992270402822061)=12.461549544515188; dx=-9.11017920019532E12 evalInputDelta=904.9163017515992
Armijo: th(4.2746720040315154E-4)=13.021980988981587; dx=-9.110179200262186E12 evalInputDelta=904.3558703071328
Armijo: th(5.343340005039394E-5)=13.121339630409974; dx=-9.110179200280037E12 evalInputDelta=904.2565116657045
Armijo: th(5.9370444500437714E-6)=13.13445813495864; dx=-9.110179200282629E12 evalInputDelta=904.2433931611557
Armijo: th(5.937044450043771E-7)=13.135941723459954; dx=-9.110179200282928E12 evalInputDelta=904.2419095726544
Armijo: th(5.397313136403428E-8)=13.13609166921485; dx=-9.110179200282957E12 evalInputDelta=904.2417596268996
Armijo: th(4.4977609470028565E-9)=13.136105415054455; dx=-9.110179200282959E12 evalInputDelta=904.24174588106
Armijo: th(3.4598161130791205E-10)=13.456328048917607; dx=-3.2000009110213595E19 evalInputDelta=903.9215232471968
Armijo: th(2.4712972236279432E-11)=23.521548880849803; dx=-1.6080640009156675E22 evalInputDelta=893.8563024152646
Armijo: th(1.6475314824186289E-12)=765.4623363106555; dx=-7.322439680050642E24 evalInputDelta=151.9155149854589
Armijo: th(1.029707176511643E-13)=896.3443646211654; dx=-8.845573760014252E24 evalInputDelta=21.033486674949017
Armijo: th(6.057101038303783E-15)=917.3778512961128; dx=-9.1101792E24 evalInputDelta=1.5916157281026244E-12
MIN ALPHA (3.3650561323909904E-16): th(2.154434690031884)=0.06185868047315239
Fitness changed from 917.3778512961144 to 0.06185868047315239
Iteration 1 complete. Error: 0.06185868047315239 Total: 0.6694; Orientation: 0.0050; Line Search: 0.6095
th(0)=0.06185868047315239;dx=-17.033619801514888
New Minimum: 0.06185868047315239 > 0.06185868047315231
WOLFE (weak): th(2.154434690031884E-15)=0.06185868047315231; dx=-17.033619801514888 evalInputDelta=8.326672684688674E-17
New Minimum: 0.06185868047315231 > 0.06185868047315224
WOLFE (weak): th(4.308869380063768E-15)=0.06185868047315224; dx=-17.033619801514888 evalInputDelta=1.5265566588595902E-16
New Minimum: 0.06185868047315224 > 0.06185868047315192
WOLFE (weak): th(1.2926608140191303E-14)=0.06185868047315192; dx=-17.033619801514888 evalInputDelta=4.718447854656915E-16
New Minimum: 0.06185868047315192 > 0.06185868047315046
WOLFE (weak): th(5.1706432560765214E-14)=0.06185868047315046; dx=-17.033619801514888 evalInputDelta=1.9290125052862095E-15
New Minimum: 0.06185868047315046 > 0.06185868047314278
WOLFE (weak): th(2.5853216280382605E-13)=0.06185868047314278; dx=-17.033619801514885 evalInputDelta=9.610368056911511E-15
New Minimum: 0.06185868047314278 > 0.06185868047309471
WOLFE (weak): th(1.5511929768229563E-12)=0.06185868047309471; dx=-17.03361980151488 evalInputDelta=5.768302502318079E-14
New Minimum: 0.06185868047309471 > 0.061858680472748596
WOLFE (weak): th(1.0858350837760695E-11)=0.061858680472748596; dx=-17.03361980151485 evalInputDelta=4.0379505295007334E-13
New Minimum: 0.061858680472748596 > 0.061858680469921955
WOLFE (weak): th(8.686680670208556E-11)=0.061858680469921955; dx=-17.033619801514575 evalInputDelta=3.2304367514335297E-12
New Minimum: 0.061858680469921955 > 0.06185868044407842
WOLFE (weak): th(7.8180126031877E-10)=0.06185868044407842; dx=-17.03361980151209 evalInputDelta=2.907397239626519E-11
New Minimum: 0.06185868044407842 > 0.06185868018241271
WOLFE (weak): th(7.818012603187701E-9)=0.06185868018241271; dx=-17.033619801486907 evalInputDelta=2.907396823292885E-10
New Minimum: 0.06185868018241271 > 0.06185867727501565
WOLFE (weak): th(8.599813863506471E-8)=0.06185867727501565; dx=-17.03361980120708 evalInputDelta=3.198136741544566E-9
New Minimum: 0.06185867727501565 > 0.061858642095513304
WOLFE (weak): th(1.0319776636207765E-6)=0.061858642095513304; dx=-17.03361979782122 evalInputDelta=3.8377639087483484E-8
New Minimum: 0.061858642095513304 > 0.061858181564141634
WOLFE (weak): th(1.3415709627070094E-5)=0.061858181564141634; dx=-17.033619753497298 evalInputDelta=4.989090107571093E-7
New Minimum: 0.061858181564141634 > 0.061851695805622456
WOLFE (weak): th(1.878199347789813E-4)=0.061851695805622456; dx=-17.033619129280147 evalInputDelta=6.984667529935651E-6
New Minimum: 0.061851695805622456 > 0.06175392371504239
WOLFE (weak): th(0.0028172990216847197)=0.06175392371504239; dx=-17.033609720598857 evalInputDelta=1.0475675811000285E-4
New Minimum: 0.06175392371504239 > 0.060196505818371014
WOLFE (weak): th(0.045076784346955515)=0.060196505818371014; dx=-17.033460274506794 evalInputDelta=0.0016621746547813773
New Minimum: 0.060196505818371014 > 0.04988440414374552
WOLFE (weak): th(0.7663053338982437)=0.04988440414374552; dx=-17.03262621173083 evalInputDelta=0.011974276329406869
New Minimum: 0.04988440414374552 > 0.005626526959256698
WOLFE (weak): th(13.793496010168386)=0.005626526959256698; dx=-17.029972975783085 evalInputDelta=0.0562321535138957
New Minimum: 0.005626526959256698 > 0.0
WOLFE (weak): th(262.07642419319933)=0.0; dx=-17.029861324858636 evalInputDelta=0.06185868047315239
Armijo: th(5241.528483863986)=0.0; dx=-17.029861324858636 evalInputDelta=0.06185868047315239
WOLFE (weak): th(2751.802454028593)=0.0; dx=-17.029861324858636 evalInputDelta=0.06185868047315239
Armijo: th(3996.6654689462894)=0.0; dx=-17.029861324858636 evalInputDelta=0.06185868047315239
WOLFE (weak): th(3374.233961487441)=0.0; dx=-17.029861324858636 evalInputDelta=0.06185868047315239
Armijo: th(3685.449715216865)=0.0; dx=-17.029861324858636 evalInputDelta=0.06185868047315239
WOLFE (weak): th(3529.841838352153)=0.0; dx=-17.029861324858636 evalInputDelta=0.06185868047315239
WOLFE (weak): th(3607.645776784509)=0.0; dx=-17.029861324858636 evalInputDelta=0.06185868047315239
Armijo: th(3646.547746000687)=0.0; dx=-17.029861324858636 evalInputDelta=0.06185868047315239
WOLFE (weak): th(3627.0967613925977)=0.0; dx=-17.029861324858636 evalInputDelta=0.06185868047315239
mu ~= nu (3627.0967613925977): th(262.07642419319933)=0.0
Fitness changed from 0.06185868047315239 to 0.0
Iteration 2 complete. Error: 0.0 Total: 0.2735; Orientation: 0.0019; Line Search: 0.2679
th(0)=0.0;dx=-17.0275168
Armijo: th(7835.296024843983)=0.0; dx=-17.0275168 evalInputDelta=0.0
Armijo: th(3917.6480124219916)=0.0; dx=-17.0275168 evalInputDelta=0.0
Armijo: th(1305.8826708073304)=0.0; dx=-17.0275168 evalInputDelta=0.0
Armijo: th(326.4706677018326)=0.0; dx=-17.0275168 evalInputDelta=0.0
Armijo: th(65.29413354036652)=0.0; dx=-17.0275168 evalInputDelta=0.0
Armijo: th(10.882355590061087)=0.0; dx=-17.0275168 evalInputDelta=0.0
Armijo: th(1.5546222271515837)=0.0; dx=-17.0275168 evalInputDelta=0.0
Armijo: th(0.19432777839394796)=0.0; dx=-17.0275168 evalInputDelta=0.0
Armijo: th(0.02159197537710533)=0.0; dx=-17.0275168 evalInputDelta=0.0
Armijo: th(0.002159197537710533)=0.0; dx=-17.0275168 evalInputDelta=0.0
Armijo: th(1.962906852464121E-4)=0.0; dx=-17.0275168 evalInputDelta=0.0
Armijo: th(1.6357557103867676E-5)=0.0; dx=-17.0275168 evalInputDelta=0.0
Armijo: th(1.2582736233744366E-6)=0.0; dx=-17.0275168 evalInputDelta=0.0
Armijo: th(8.987668738388834E-8)=0.0; dx=-17.0275168 evalInputDelta=0.0
Armijo: th(5.991779158925889E-9)=0.0; dx=-17.0275168 evalInputDelta=0.0
Armijo: th(3.7448619743286807E-10)=0.0; dx=-17.0275168 evalInputDelta=0.0
Armijo: th(2.202859984899224E-11)=0.0; dx=-17.0275168 evalInputDelta=0.0
Armijo: th(1.2238111027217912E-12)=0.0; dx=-17.0275168 evalInputDelta=0.0
Armijo: th(6.441111066956796E-14)=0.0; dx=-17.0275168 evalInputDelta=0.0
Armijo: th(3.220555533478398E-15)=0.0; dx=-17.0275168 evalInputDelta=0.0
MIN ALPHA (1.5335978730849515E-16): th(0.0)=0.0
Fitness changed from 0.0 to 0.0
Static Iteration Total: 0.1689; Orientation: 0.0021; Line Search: 0.1638
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 1.112s (< 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 1.07 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: 2517946933962
Reset training subject: 2517949679588
Constructing line search parameters: GD
F(0.0) = LineSearchPoint{point=PointSample{avg=917.3778512961144}, derivative=-9.1101792E24}
New Minimum: 917.3778512961144 > 14.103400084660572
F(1.0E-10) = LineSearchPoint{point=PointSample{avg=14.103400084660572}, derivative=-1.209600091135841E20}, evalInputDelta = -903.2744512114538
New Minimum: 14.103400084660572 > 13.143547720596711
F(7.000000000000001E-10) = LineSearchPoint{point=PointSample{avg=13.143547720596711}, derivative=-9.110221781742928E12}, evalInputDelta = -904.2343035755176
New Minimum: 13.143547720596711 > 13.136105303299036
F(4.900000000000001E-9) = LineSearchPoint{point=PointSample{avg=13.136105303299036}, derivative=-9.110179200282959E12}, evalInputDelta = -904.2417459928154
New Minimum: 13.136105303299036 > 13.136097135023345
F(3.430000000000001E-8) = LineSearchPoint{point=PointSample{avg=13.136097135023345}, derivative=-9.110179200282959E12}, evalInputDelta = -904.2417541610911
New Minimum: 13.136097135023345 > 13.136039958442055
F(2.4010000000000004E-7) = LineSearchPoint{point=PointSample{avg=13.136039958442055}, derivative=-9.110179200282947E12}, evalInputDelta = -904.2418113376723
New Minimum: 13.136039958442055 > 13.135639788424566
F(1.6807000000000003E-6) = LineSearchPoint{point=PointSample{avg=13.135639788424566}, derivative=-9.110179200282867E12}, evalInputDelta = -904.2422115076898
New Minimum: 13.135639788424566 > 13.132841825876742
F(1.1764900000000001E-5) = LineSearchPoint{point=PointSample{avg=13.132841825876742}, derivative=-9.110179200282307E12}, evalInputDelta = -904.2450094702376
New Minimum: 13.132841825876742 > 13.113411250421013
F(8.235430000000001E-5) = LineSearchPoint{point=PointSample{avg=13.113411250421013}, derivative=-9.1101792002785E12}, evalInputDelta = -904.2644400456934
New Minimum: 13.113411250421013 > 12.984136095825713
F(5.764801000000001E-4) = LineSearchPoint{point=PointSample{avg=12.984136095825713}, derivative=-9.11017920025609E12}, evalInputDelta = -904.3937152002886
New Minimum: 12.984136095825713 > 12.271583081166629
F(0.004035360700000001) = LineSearchPoint{point=PointSample{avg=12.271583081166629}, derivative=-9.11017920017979E12}, evalInputDelta = -905.1062682149477
New Minimum: 12.271583081166629 > 9.686177150958441
F(0.028247524900000005) = LineSearchPoint{point=PointSample{avg=9.686177150958441}, derivative=-9.11017920007916E12}, evalInputDelta = -907.691674145156
New Minimum: 9.686177150958441 > 3.998478835377359
F(0.19773267430000002) = LineSearchPoint{point=PointSample{avg=3.998478835377359}, derivative=-9.110179200027467E12}, evalInputDelta = -913.379372460737
New Minimum: 3.998478835377359 > 0.20174429140861724
F(1.3841287201) = LineSearchPoint{point=PointSample{avg=0.20174429140861724}, derivative=-9.11017920002155E12}, evalInputDelta = -917.1761070047057
New Minimum: 0.20174429140861724 > 0.006941818964854687
F(9.688901040700001) = LineSearchPoint{point=PointSample{avg=0.006941818964854687}, derivative=-9.1101792000215E12}, evalInputDelta = -917.3709094771496
New Minimum: 0.006941818964854687 > 3.881627871223817E-4
F(67.8223072849) = LineSearchPoint{point=PointSample{avg=3.881627871223817E-4}, derivative=-9.1101792000215E12}, evalInputDelta = -917.3774631333273
New Minimum: 3.881627871223817E-4 > 0.0
F(474.7561509943) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-9.1101792000215E12}, evalInputDelta = -917.3778512961144
F(3323.2930569601003) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-9.1101792000215E12}, evalInputDelta = -917.3778512961144
F(23263.0513987207) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-9.1101792000215E12}, evalInputDelta = -917.3778512961144
F(162841.3597910449) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-9.1101792000215E12}, evalInputDelta = -917.3778512961144
F(1139889.5185373144) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-9.1101792000215E12}, evalInputDelta = -917.3778512961144
F(7979226.6297612) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-9.1101792000215E12}, evalInputDelta = -917.3778512961144
F(5.58545864083284E7) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-9.1101792000215E12}, evalInputDelta = -917.3778512961144
F(3.909821048582988E8) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-9.1101792000215E12}, evalInputDelta = -917.3778512961144
F(2.7368747340080914E9) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-9.1101792000215E12}, evalInputDelta = -917.3778512961144
F(1.915812313805664E10) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-9.1101792000215E12}, evalInputDelta = -917.3778512961144
0.0 <= 917.3778512961144
F(1.0E10) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-9.1101792000215E12}, evalInputDelta = -917.3778512961144
Right bracket at 1.0E10
Converged to right
Fitness changed from 917.3778512961144 to 0.0
Iteration 1 complete. Error: 0.0 Total: 0.7414; Orientation: 0.0016; Line Search: 0.7318
F(0.0) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-17.0275168}
F(1.0E10) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-17.0275168}, evalInputDelta = 0.0
0.0 <= 0.0
F(5.0E9) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-17.0275168}, evalInputDelta = 0.0
Right bracket at 5.0E9
F(2.5E9) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-17.0275168}, evalInputDelta = 0.0
Right bracket at 2.5E9
F(1.25E9) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-17.0275168}, evalInputDelta = 0.0
Right bracket at 1.25E9
F(6.25E8) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-17.0275168}, evalInputDelta = 0.0
Right bracket at 6.25E8
F(3.125E8) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-17.0275168}, evalInputDelta = 0.0
Right bracket at 3.125E8
F(1.5625E8) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-17.0275168}, evalInputDelta = 0.0
Right bracket at 1.5625E8
F(7.8125E7) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-17.0275168}, evalInputDelta = 0.0
Right bracket at 7.8125E7
F(3.90625E7) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-17.0275168}, evalInputDelta = 0.0
Right bracket at 3.90625E7
F(1.953125E7) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-17.0275168}, evalInputDelta = 0.0
Right bracket at 1.953125E7
F(9765625.0) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-17.0275168}, evalInputDelta = 0.0
Right bracket at 9765625.0
F(4882812.5) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-17.0275168}, evalInputDelta = 0.0
Right bracket at 4882812.5
F(2441406.25) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-17.0275168}, evalInputDelta = 0.0
Loops = 12
Fitness changed from 0.0 to 0.0
Static Iteration Total: 0.3325; Orientation: 0.0018; Line Search: 0.3287
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 1.074s (< 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 23.82 seconds (0.069 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: 2519025183984
Reset training subject: 2519027125845
Adding measurement 3c726357 to history. Total: 0
LBFGS Accumulation History: 1 points
Constructing line search parameters: GD
Non-optimal measurement 917.3778512961144 < 917.3778512961144. Total: 1
th(0)=917.3778512961144;dx=-9.1101792E24
Adding measurement 7408b762 to history. Total: 1
New Minimum: 917.3778512961144 > 0.06185868047315239
Armijo: th(2.154434690031884)=0.06185868047315239; dx=-9.110179200021508E12 evalInputDelta=917.3159926156412
Non-optimal measurement 0.3475817674561618 < 0.06185868047315239. Total: 2
Armijo: th(1.077217345015942)=0.3475817674561618; dx=-9.110179200021598E12 evalInputDelta=917.0302695286582
Non-optimal measurement 2.1421277620337946 < 0.06185868047315239. Total: 2
Armijo: th(0.3590724483386473)=2.1421277620337946; dx=-9.110179200023053E12 evalInputDelta=915.2357235340805
Non-optimal measurement 6.6303280548113985 < 0.06185868047315239. Total: 2
Armijo: th(0.08976811208466183)=6.6303280548113985; dx=-9.11017920004115E12 evalInputDelta=910.747523241303
Non-optimal measurement 10.542443179667108 < 0.06185868047315239. Total: 2
Armijo: th(0.017953622416932366)=10.542443179667108; dx=-9.110179200099678E12 evalInputDelta=906.8354081164473
Non-optimal measurement 12.461549544515188 < 0.06185868047315239. Total: 2
Armijo: th(0.002992270402822061)=12.461549544515188; dx=-9.11017920019532E12 evalInputDelta=904.9163017515992
Non-optimal measurement 13.021980988981587 < 0.06185868047315239. Total: 2
Armijo: th(4.2746720040315154E-4)=13.021980988981587; dx=-9.110179200262186E12 evalInputDelta=904.3558703071328
Non-optimal measurement 13.121339630409974 < 0.06185868047315239. Total: 2
Armijo: th(5.343340005039394E-5)=13.121339630409974; dx=-9.110179200280037E12 evalInputDelta=904.2565116657045
Non-optimal measurement 13.13445813495864 < 0.06185868047315239. Total: 2
Armijo: th(5.9370444500437714E-6)=13.13445813495864; dx=-9.110179200282629E12 evalInputDelta=904.2433931611557
Non-optimal measurement 13.135941723459954 < 0.06185868047315239. Total: 2
Armijo: th(5.937044450043771E-7)=13.135941723459954; dx=-9.110179200282928E12 evalInputDelta=904.2419095726544
Non-optimal measurement 13.13609166921485 < 0.06185868047315239. Total: 2
Armijo: th(5.397313136403428E-8)=13.13609166921485; dx=-9.110179200282957E12 evalInputDelta=904.2417596268996
Non-optimal measurement 13.136105415054455 < 0.06185868047315239. Total: 2
Armijo: th(4.4977609470028565E-9)=13.136105415054455; dx=-9.110179200282959E12 evalInputDelta=904.24174588106
Non-optimal measurement 13.456328048917607 < 0.06185868047315239. Total: 2
Armijo: th(3.4598161130791205E-10)=13.456328048917607; dx=-3.2000009110213595E19 evalInputDelta=903.9215232471968
Non-optimal measurement 23.521548880849803 < 0.06185868047315239. Total: 2
Armijo: th(2.4712972236279432E-11)=23.521548880849803; dx=-1.6080640009156675E22 evalInputDelta=893.8563024152646
Non-optimal measurement 765.4623363106555 < 0.06185868047315239. Total: 2
Armijo: th(1.6475314824186289E-12)=765.4623363106555; dx=-7.322439680050642E24 evalInputDelta=151.9155149854589
Non-optimal measurement 896.3443646211654 < 0.06185868047315239. Total: 2
Armijo: th(1.029707176511643E-13)=896.3443646211654; dx=-8.845573760014252E24 evalInputDelta=21.033486674949017
Non-optimal measurement 917.3778512961128 < 0.06185868047315239. Total: 2
Armijo: th(6.057101038303783E-15)=917.3778512961128; dx=-9.1101792E24 evalInputDelta=1.5916157281026244E-12
Non-optimal measurement 0.06185868047315239 < 0.06185868047315239. Total: 2
MIN ALPHA (3.3650561323909904E-16): th(2.154434690031884)=0.06185868047315239
Fitness changed from 917.3778512961144 to 0.06185868047315239
Iteration 1 complete. Error: 0.06185868047315239 Total: 0.4004; Orientation: 0.0058; Line Search: 0.3881
Non-optimal measurement 0.06185868047315239 < 0.06185868047315239. Total: 2
LBFGS Accumulation History: 2 points
Non-optimal measurement 0.06185868047315239 < 0.06185868047315239. Total: 2
th(0)=0.06185868047315239;dx=-17.033619801514888
Adding measurement 417e6e05 to history. Total: 2
New Minimum: 0.06185868047315239 > 0.06185868047315231
WOLFE (weak): th(2.154434690031884E-15)=0.06185868047315231; dx=-17.033619801514888 evalInputDelta=8.326672684688674E-17
Adding measurement 1e60ec2c to history. Total: 3
New Minimum: 0.06185868047315231 > 0.06185868047315224
WOLFE (weak): th(4.308869380063768E-15)=0.06185868047315224; dx=-17.033619801514888 evalInputDelta=1.5265566588595902E-16
Adding measurement 3afc21f4 to history. Total: 4
New Minimum: 0.06185868047315224 > 0.06185868047315192
WOLFE (weak): th(1.2926608140191303E-14)=0.06185868047315192; dx=-17.033619801514888 evalInputDelta=4.718447854656915E-16
Adding measurement 66c37120 to history. Total: 5
New Minimum: 0.06185868047315192 > 0.06185868047315046
WOLFE (weak): th(5.1706432560765214E-14)=0.06185868047315046; dx=-17.033619801514888 evalInputDelta=1.9290125052862095E-15
Adding measurement a6f167b to history. Total: 6
New Minimum: 0.06185868047315046 > 0.06185868047314278
WOLFE (weak): th(2.5853216280382605E-13)=0.06185868047314278; dx=-17.033619801514885 evalInputDelta=9.610368056911511E-15
Adding measurement 4b3b3157 to history. Total: 7
New Minimum: 0.06185868047314278 > 0.06185868047309471
WOLFE (weak): th(1.5511929768229563E-12)=0.06185868047309471; dx=-17.03361980151488 evalInputDelta=5.768302502318079E-14
Adding measurement 37e5a200 to history. Total: 8
New Minimum: 0.06185868047309471 > 0.061858680472748596
WOLFE (weak): th(1.0858350837760695E-11)=0.061858680472748596; dx=-17.03361980151485 evalInputDelta=4.0379505295007334E-13
Adding measurement 6e5d13bb to history. Total: 9
New Minimum: 0.061858680472748596 > 0.061858680469921955
WOLFE (weak): th(8.686680670208556E-11)=0.061858680469921955; dx=-17.033619801514575 evalInputDelta=3.2304367514335297E-12
Adding measurement 556e3a70 to history. Total: 10
New Minimum: 0.061858680469921955 > 0.06185868044407842
WOLFE (weak): th(7.8180126031877E-10)=0.06185868044407842; dx=-17.03361980151209 evalInputDelta=2.907397239626519E-11
Adding measurement 5198fd01 to history. Total: 11
New Minimum: 0.06185868044407842 > 0.06185868018241271
WOLFE (weak): th(7.818012603187701E-9)=0.06185868018241271; dx=-17.033619801486907 evalInputDelta=2.907396823292885E-10
Adding measurement 4988588e to history. Total: 12
New Minimum: 0.06185868018241271 > 0.06185867727501565
WOLFE (weak): th(8.599813863506471E-8)=0.06185867727501565; dx=-17.03361980120708 evalInputDelta=3.198136741544566E-9
Adding measurement 49892c81 to history. Total: 13
New Minimum: 0.06185867727501565 > 0.061858642095513304
WOLFE (weak): th(1.0319776636207765E-6)=0.061858642095513304; dx=-17.03361979782122 evalInputDelta=3.8377639087483484E-8
Adding measurement 45b371ea to history. Total: 14
New Minimum: 0.061858642095513304 > 0.061858181564141634
WOLFE (weak): th(1.3415709627070094E-5)=0.061858181564141634; dx=-17.033619753497298 evalInputDelta=4.989090107571093E-7
Adding measurement 796b9eba to history. Total: 15
New Minimum: 0.061858181564141634 > 0.061851695805622456
WOLFE (weak): th(1.878199347789813E-4)=0.061851695805622456; dx=-17.033619129280147 evalInputDelta=6.984667529935651E-6
Adding measurement 7e5bcf3d to history. Total: 16
New Minimum: 0.061851695805622456 > 0.06175392371504239
WOLFE (weak): th(0.0028172990216847197)=0.06175392371504239; dx=-17.033609720598857 evalInputDelta=1.0475675811000285E-4
Adding measurement 3f3b5109 to history. Total: 17
New Minimum: 0.06175392371504239 > 0.060196505818371014
WOLFE (weak): th(0.045076784346955515)=0.060196505818371014; dx=-17.033460274506794 evalInputDelta=0.0016621746547813773
Adding measurement 4c646342 to history. Total: 18
New Minimum: 0.060196505818371014 > 0.04988440414374552
WOLFE (weak): th(0.7663053338982437)=0.04988440414374552; dx=-17.03262621173083 evalInputDelta=0.011974276329406869
Adding measurement 252f361d to history. Total: 19
New Minimum: 0.04988440414374552 > 0.005626526959256698
WOLFE (weak): th(13.793496010168386)=0.005626526959256698; dx=-17.029972975783085 evalInputDelta=0.0562321535138957
Adding measurement 41bcd646 to history. Total: 20
New Minimum: 0.0056265269592

...skipping 14172 bytes...

b1800 = 1.000/1.000e+00, a590eaa3-fde8-42ef-8b58-c717e3967814 = 1.000/1.000e+00, 9940a45e-e8e3-4de5-bd38-56221347f91a = 1.000/1.000e+00, d0fa58cb-bc9a-49d7-88b3-9c4a4b1c47ef = 1.000/1.000e+00, c75f81e5-558b-41ec-9451-89e63f30358b = 1.000/1.000e+00]
Orientation rejected. Popping history element from 0.0, 0.005626526959256698, 0.04988440414374552, 0.060196505818371014, 0.06175392371504239, 0.061851695805622456, 0.061858181564141634, 0.061858642095513304, 0.06185867727501565
Rejected: LBFGS Orientation magnitude: 3.166e+04, gradient 4.126e+00, dot -0.962; [27079941-0ae3-45b1-b88d-16dee131ac92 = 1.000/1.000e+00, c5a87a60-1c7e-4c4b-a162-51a6d4711569 = 1.000/1.000e+00, d6372f76-6bc4-4b58-bd5d-5c2ea5bb1800 = 1.000/1.000e+00, a590eaa3-fde8-42ef-8b58-c717e3967814 = 1.000/1.000e+00, c75f81e5-558b-41ec-9451-89e63f30358b = 1.000/1.000e+00, a2db4300-5b25-4041-849b-4a98a8b43109 = 1.000/1.000e+00, 38f8adf4-5500-4a4b-8934-196b7e7f3e71 = 1.000/1.000e+00, 071f3e1d-9459-4ace-8d94-ceed36dc2735 = 1.000/1.000e+00, d0fa58cb-bc9a-49d7-88b3-9c4a4b1c47ef = 1.000/1.000e+00, 9940a45e-e8e3-4de5-bd38-56221347f91a = 1.000/1.000e+00]
Orientation rejected. Popping history element from 0.0, 0.005626526959256698, 0.04988440414374552, 0.060196505818371014, 0.06175392371504239, 0.061851695805622456, 0.061858181564141634, 0.061858642095513304
Rejected: LBFGS Orientation magnitude: 4.359e+04, gradient 4.126e+00, dot -0.941; [9940a45e-e8e3-4de5-bd38-56221347f91a = 1.000/1.000e+00, c75f81e5-558b-41ec-9451-89e63f30358b = 1.000/1.000e+00, c5a87a60-1c7e-4c4b-a162-51a6d4711569 = 1.000/1.000e+00, 38f8adf4-5500-4a4b-8934-196b7e7f3e71 = 1.000/1.000e+00, a590eaa3-fde8-42ef-8b58-c717e3967814 = 1.000/1.000e+00, 071f3e1d-9459-4ace-8d94-ceed36dc2735 = 1.000/1.000e+00, a2db4300-5b25-4041-849b-4a98a8b43109 = 1.000/1.000e+00, 27079941-0ae3-45b1-b88d-16dee131ac92 = 1.000/1.000e+00, d6372f76-6bc4-4b58-bd5d-5c2ea5bb1800 = 1.000/1.000e+00, d0fa58cb-bc9a-49d7-88b3-9c4a4b1c47ef = 1.000/1.000e+00]
Orientation rejected. Popping history element from 0.0, 0.005626526959256698, 0.04988440414374552, 0.060196505818371014, 0.06175392371504239, 0.061851695805622456, 0.061858181564141634
Rejected: LBFGS Orientation magnitude: 3.921e+04, gradient 4.126e+00, dot -1.000; [a2db4300-5b25-4041-849b-4a98a8b43109 = 1.000/1.000e+00, c75f81e5-558b-41ec-9451-89e63f30358b = 1.000/1.000e+00, 27079941-0ae3-45b1-b88d-16dee131ac92 = 1.000/1.000e+00, d0fa58cb-bc9a-49d7-88b3-9c4a4b1c47ef = 1.000/1.000e+00, c5a87a60-1c7e-4c4b-a162-51a6d4711569 = 1.000/1.000e+00, 9940a45e-e8e3-4de5-bd38-56221347f91a = 1.000/1.000e+00, d6372f76-6bc4-4b58-bd5d-5c2ea5bb1800 = 1.000/1.000e+00, 071f3e1d-9459-4ace-8d94-ceed36dc2735 = 1.000/1.000e+00, a590eaa3-fde8-42ef-8b58-c717e3967814 = 1.000/1.000e+00, 38f8adf4-5500-4a4b-8934-196b7e7f3e71 = 1.000/1.000e+00]
Orientation rejected. Popping history element from 0.0, 0.005626526959256698, 0.04988440414374552, 0.060196505818371014, 0.06175392371504239, 0.061851695805622456
Rejected: LBFGS Orientation magnitude: 3.974e+04, gradient 4.126e+00, dot -1.000; [38f8adf4-5500-4a4b-8934-196b7e7f3e71 = 1.000/1.000e+00, c5a87a60-1c7e-4c4b-a162-51a6d4711569 = 1.000/1.000e+00, a590eaa3-fde8-42ef-8b58-c717e3967814 = 1.000/1.000e+00, a2db4300-5b25-4041-849b-4a98a8b43109 = 1.000/1.000e+00, 9940a45e-e8e3-4de5-bd38-56221347f91a = 1.000/1.000e+00, c75f81e5-558b-41ec-9451-89e63f30358b = 1.000/1.000e+00, 27079941-0ae3-45b1-b88d-16dee131ac92 = 1.000/1.000e+00, d6372f76-6bc4-4b58-bd5d-5c2ea5bb1800 = 1.000/1.000e+00, 071f3e1d-9459-4ace-8d94-ceed36dc2735 = 1.000/1.000e+00, d0fa58cb-bc9a-49d7-88b3-9c4a4b1c47ef = 1.000/1.000e+00]
Orientation rejected. Popping history element from 0.0, 0.005626526959256698, 0.04988440414374552, 0.060196505818371014, 0.06175392371504239
Rejected: LBFGS Orientation magnitude: 1.215e+05, gradient 4.126e+00, dot -1.000; [a590eaa3-fde8-42ef-8b58-c717e3967814 = 1.000/1.000e+00, a2db4300-5b25-4041-849b-4a98a8b43109 = 1.000/1.000e+00, d0fa58cb-bc9a-49d7-88b3-9c4a4b1c47ef = 1.000/1.000e+00, 9940a45e-e8e3-4de5-bd38-56221347f91a = 1.000/1.000e+00, 38f8adf4-5500-4a4b-8934-196b7e7f3e71 = 1.000/1.000e+00, c75f81e5-558b-41ec-9451-89e63f30358b = 1.000/1.000e+00, 27079941-0ae3-45b1-b88d-16dee131ac92 = 1.000/1.000e+00, d6372f76-6bc4-4b58-bd5d-5c2ea5bb1800 = 1.000/1.000e+00, 071f3e1d-9459-4ace-8d94-ceed36dc2735 = 1.000/1.000e+00, c5a87a60-1c7e-4c4b-a162-51a6d4711569 = 1.000/1.000e+00]
Orientation rejected. Popping history element from 0.0, 0.005626526959256698, 0.04988440414374552, 0.060196505818371014
LBFGS Accumulation History: 3 points
Removed measurement 41bcd646 to history. Total: 20
Removed measurement 252f361d to history. Total: 19
Removed measurement 4c646342 to history. Total: 18
Removed measurement 3f3b5109 to history. Total: 17
Removed measurement 7e5bcf3d to history. Total: 16
Removed measurement 796b9eba to history. Total: 15
Removed measurement 45b371ea to history. Total: 14
Removed measurement 49892c81 to history. Total: 13
Removed measurement 4988588e to history. Total: 12
Removed measurement 5198fd01 to history. Total: 11
Removed measurement 556e3a70 to history. Total: 10
Removed measurement 6e5d13bb to history. Total: 9
Removed measurement 37e5a200 to history. Total: 8
Removed measurement 4b3b3157 to history. Total: 7
Removed measurement a6f167b to history. Total: 6
Removed measurement 66c37120 to history. Total: 5
Removed measurement 3afc21f4 to history. Total: 4
Removed measurement 1e60ec2c to history. Total: 3
Adding measurement 730d9897 to history. Total: 3
th(0)=0.0;dx=-17.0275168
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(7835.296024843983)=0.0; dx=-17.0275168 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(3917.6480124219916)=0.0; dx=-17.0275168 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(1305.8826708073304)=0.0; dx=-17.0275168 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(326.4706677018326)=0.0; dx=-17.0275168 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(65.29413354036652)=0.0; dx=-17.0275168 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(10.882355590061087)=0.0; dx=-17.0275168 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(1.5546222271515837)=0.0; dx=-17.0275168 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(0.19432777839394796)=0.0; dx=-17.0275168 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(0.02159197537710533)=0.0; dx=-17.0275168 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(0.002159197537710533)=0.0; dx=-17.0275168 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(1.962906852464121E-4)=0.0; dx=-17.0275168 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(1.6357557103867676E-5)=0.0; dx=-17.0275168 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(1.2582736233744366E-6)=0.0; dx=-17.0275168 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(8.987668738388834E-8)=0.0; dx=-17.0275168 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(5.991779158925889E-9)=0.0; dx=-17.0275168 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(3.7448619743286807E-10)=0.0; dx=-17.0275168 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(2.202859984899224E-11)=0.0; dx=-17.0275168 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(1.2238111027217912E-12)=0.0; dx=-17.0275168 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(6.441111066956796E-14)=0.0; dx=-17.0275168 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(3.220555533478398E-15)=0.0; dx=-17.0275168 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
MIN ALPHA (1.5335978730849515E-16): th(0.0)=0.0
Fitness changed from 0.0 to 0.0
Static Iteration Total: 22.7606; Orientation: 22.5084; Line Search: 0.2493
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 23.819s (< 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, -2.20859934830537], [2.0, -0.20859934830536964]; valueStats=DoubleSummaryStatistics{count=2, sum=0.123717, min=0.061859, average=0.061859, max=0.061859}
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, -2.20859934830537], [0.657, -0.20859934830536964]; valueStats=DoubleSummaryStatistics{count=2, sum=0.123717, min=0.061859, average=0.061859, max=0.061859}
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": "26.747",
      "gc_time": "0.341"
    },
    "created_on": 1586737104617,
    "file_name": "trainingTest",
    "report": {
      "simpleName": "Basic",
      "canonicalName": "com.simiacryptus.mindseye.layers.java.ImgConcatLayerTest.Basic",
      "link": "https://github.com/SimiaCryptus/mindseye-java/tree/93db34cedee48c0202777a2b25deddf1dfaf5731/src/test/java/com/simiacryptus/mindseye/layers/java/ImgConcatLayerTest.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/ImgConcatLayer/Basic/trainingTest/202004131824",
    "id": "c3c097a5-2a32-4a98-a3e8-ea8ce7302048",
    "report_type": "Components",
    "display_name": "Comparative Training",
    "target": {
      "simpleName": "ImgConcatLayer",
      "canonicalName": "com.simiacryptus.mindseye.layers.java.ImgConcatLayer",
      "link": "https://github.com/SimiaCryptus/mindseye-java/tree/93db34cedee48c0202777a2b25deddf1dfaf5731/src/main/java/com/simiacryptus/mindseye/layers/java/ImgConcatLayer.java",
      "javaDoc": ""
    }
  }