Subreport: Logs for com.simiacryptus.ref.lang.ReferenceCountingBase
Using Seed 8552693377147246592
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
[
[ [ 0.568, -1.108, 1.916, 1.912, -0.824, 0.408, -0.588, -1.212, ... ], [ -1.12, -1.052, 0.924, 0.024, -0.536, 0.004, -1.064, 1.58, ... ], [ -0.892, 0.288, -0.484, 1.192, -1.648, 1.064, -1.82, -0.452, ... ], [ -0.808, -1.94, -1.116, 1.188, -1.644, 1.032, -1.36, -1.388, ... ] ],
[ [ 1.584, -0.66, 0.868, -1.092, -0.076, -0.696, 0.836, 0.876, ... ], [ 1.308, -1.62, 0.792, -1.344, -1.772, -1.656, -1.368, 1.024, ... ], [ 0.52, -0.316, -1.524, 0.26, 0.408, 1.164, -0.844, 0.628, ... ], [ -0.068, -1.72, -0.372, 1.808, 1.032, -1.204, -0.064, 0.932, ... ] ],
[ [ -1.584, -1.156, 1.948, -1.644, -1.668, 0.164, 0.88, 0.432, ... ], [ 1.196, -1.728, -1.516, -1.364, 0.008, -1.312, -1.524, 0.696, ... ], [ 1.68, 1.184, -0.108, -0.948, -1.684, -0.228, -0.64, 1.992, ... ], [ 0.572, -0.42, 1.248, -0.588, 0.528, -0.888, -1.084, -0.072, ... ] ],
[ [ -0.704, -0.82, 1.164, -1.748, 0.576, -0.132, -0.532, 1.8, ... ], [ -1.112, -0.836, -1.428, 0.404, 1.492, -0.428, -0.248, 0.548, ... ], [ 1.88, 1.288, -1.292, 0.756, -0.996, 1.804, -1.736, -1.292, ... ], [ -1.328, 1.26, -1.128, -0.104, -0.904, 0.648, 0.772, 0.868, ... ] ]
]
[
[ [ 1.98, 1.716, 0.596, 0.888, 1.12, 1.852, 1.62, -1.112, ... ], [ 1.652, 1.016, 1.752, -0.28, -0.032, 0.352, -1.068, -0.904, ... ], [ -1.832, -1.24, -0.092, 0.236, 1.948, 0.168, 1.18, -1.204, ... ], [ -1.668, 1.248, -1.908, 0.296, -1.316, -0.688, -0.576, 0.796, ... ] ],
[ [ -0.6, 1.224, 0.7, -1.488, -1.38, 1.944, -0.268, 0.488, ... ], [ -0.804, -1.456, 1.948, 1.936, -0.496, 1.368, 0.292, -0.684, ... ], [ 1.116, 0.48, -1.808, -0.028, -1.868, 0.564, -1.232, 1.16, ... ], [ 1.444, 1.548, 0.876, 1.684, -0.028, -0.66, 1.956, 1.748, ... ] ],
[ [ -0.536, -1.868, 0.512, 1.14, -0.672, 0.588, 1.828, -0.596, ... ], [ 0.688, -0.852, -0.084, -1.74, 1.472, -0.7, 1.628, 1.444, ... ], [ -1.476, 0.688, 1.712, -1.424, -1.012, 1.88, -1.656, -1.316, ... ], [ -1.524, -1.4, 1.924, 1.048, 1.584, -1.9, 0.64, -0.4, ... ] ],
[ [ -0.892, -0.42, -0.344, 1.232, 1.38, 0.644, -1.884, -1.156, ... ], [ -0.636, 0.56, -0.476, -0.108, -0.8, 0.952, 0.544, -0.468, ... ], [ 0.952, -0.404, 0.604, 1.448, 1.316, -1.836, 0.192, 1.544, ... ], [ -0.78, -0.64, -1.604, 1.164, -1.492, -1.328, -0.592, 1.964, ... ] ]
]
[
[ [ -0.764, 0.18, 1.544, 1.24, -1.948, 1.56, 1.628, 1.804, ... ], [ -1.236, 1.308, -1.452, -2.0, -1.612, -1.316, 1.268, -0.4, ... ], [ 0.944, -1.308, -1.324, -0.36, 0.252, -1.684, 1.12, 1.076, ... ], [ -1.108, 1.392, 0.92, 0.452, 1.224, -1.436, -1.252, 0.596, ... ] ],
[ [ 0.56, -1.488, 1.372, -0.836, 1.192, -1.712, 0.348, 0.064, ... ], [ 0.34, 0.372, -0.084, -0.7, -1.156, 1.548, 1.76, -0.208, ... ], [ 1.708, 0.532, -1.588, 0.612, 1.564, 1.36, -0.364, -0.156, ... ], [ 1.82, -1.476, 1.088, -1.8, -1.984, -1.564, -1.156, 0.232, ... ] ],
[ [ 1.044, 1.476, 1.94, -0.788, -0.66, -1.536, -0.24, -1.232, ... ], [ 1.16, -0.696, -0.904, -0.664, -1.044, -0.212, 0.168, -1.616, ... ], [ 0.76, 1.02, 1.616, -0.412, 1.464, 1.544, -1.208, -1.784, ... ], [ 0.952, 1.752, -1.632, 1.94, 1.0, 0.972, 0.832, -1.616, ... ] ],
[ [ 0.048, -1.168, 1.248, 1.196, -1.048, -0.584, 1.348, 1.844, ... ], [ -1.42, 1.284, -0.664, -0.548, 0.248, 0.66, 1.968, 0.068, ... ], [ -0.256, 1.04, 1.12, -0.072, -1.788, -0.796, -1.88, 1.944, ... ], [ 1.868, 0.788, 1.844, 0.332, -1.424, 1.392, 0.156, -0.38, ... ] ]
]
[
[ [ -0.132, 1.24, 1.596, 1.952, 0.856, 0.312, -0.228, 0.868, ... ], [ -1.532, -0.4, -0.568, 1.904, -0.216, -1.084, -1.632, -0.092, ... ], [ 0.224, 0.616, 1.88, -0.628, -1.548, -0.536, -1.272, 0.88, ... ], [ 1.38, -1.24, -1.624, -1.056, 0.008, -1.768, -1.564, -1.28, ... ] ],
[ [ 1.98, 0.78, -0.604, 1.748, -1.652, 0.968, -0.716, 0.764, ... ], [ -1.932, 1.444, -1.86, 1.472, 0.424, -0.024, -1.924, 0.244, ... ], [ 1.7, -1.088, 0.324, -0.536, 1.056, 1.444, 1.012, -0.196, ... ], [ 0.58, 0.556, -0.928, 0.4, 1.716, -1.908, 0.16, -0.876, ... ] ],
[ [ 0.516, -1.76, 0.72, -0.996, 1.844, -0.316, -0.952, -0.504, ... ], [ 1.3, -1.316, -0.696, -0.92, -1.988, -1.948, 0.844, -0.212, ... ], [ 1.896, 0.104, -0.948, -0.816, 1.884, 1.976, 0.38, 0.688, ... ], [ 0.404, -1.24, 0.64, -1.98, 1.684, -1.54, -1.98, -1.452, ... ] ],
[ [ 0.7, 0.628, 1.744, -1.108, 1.62, 1.212, -0.38, -1.44, ... ], [ -0.872, 1.468, 1.716, 1.168, -1.176, -1.812, -1.3, 1.304, ... ], [ -0.992, 1.26, -1.648, 0.408, -0.56, -1.588, 0.148, 0.92, ... ], [ 0.172, -0.86, 1.844, -1.892, -0.056, -0.704, 1.48, 1.664, ... ] ]
]
[
[ [ 0.488, -1.164, -1.86, -1.564, -0.452, 0.484, 0.576, -1.14, ... ], [ -1.764, -1.804, -1.044, 1.524, 1.296, 1.12, -0.452, 0.6, ... ], [ 0.468, 1.936, 0.12, -1.524, -1.256, -0.232, -0.492, -1.7, ... ], [ 1.704, -1.308, 1.424, 1.024, -0.1, 0.196, -1.076, 0.216, ... ] ],
[ [ -0.696, -1.664, -1.168, -1.58, -1.06, 0.756, -0.636, -0.816, ... ], [ 1.52, 0.24, -0.132, 0.468, 1.688, 0.02, -0.572, -0.548, ... ], [ 0.092, 0.9, 0.26, 0.844, -1.116, -1.168, 1.872, 0.992, ... ], [ 0.064, 0.128, -0.304, -1.184, -1.228, 1.968, 0.512, -1.908, ... ] ],
[ [ 1.536, 1.556, -1.444, -1.332, 0.876, -1.74, 0.528, -0.624, ... ], [ 0.036, -0.872, -0.804, 1.144, -1.004, -0.208, 1.484, -1.328, ... ], [ -1.92, -0.916, 1.272, 0.136, -1.108, -1.096, 0.232, -1.256, ... ], [ 0.352, -1.548, 0.264, -1.424, 1.98, 0.404, 0.164, -1.62, ... ] ],
[ [ -1.008, 0.052, -1.784, 1.244, -0.716, 0.364, 0.552, 1.708, ... ], [ -1.384, -0.364, 1.312, 1.748, 1.508, 0.58, 0.896, 1.364, ... ], [ -1.944, -0.82, 0.004, 1.952, -0.684, 0.264, -0.572, 0.9, ... ], [ 1.172, -0.4, 0.784, -1.124, -1.34, -1.12, -0.544, 1.0, ... ] ]
]
First, we train using basic gradient descent method apply weak line search conditions.
TrainingTester.java:480 executed in 2.85 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();
}
Reset training subject: 1374255782196
BACKPROP_AGG_SIZE = 3
THREADS = 64
SINGLE_THREADED = false
Initialized CoreSettings = {
"backpropAggregationSize" : 3,
"jvmThreads" : 64,
"singleThreaded" : false
}
Reset training subject: 1374308267989
Constructing line search parameters: GD
th(0)=29497.60546245923;dx=-2.8392876096000002E26
New Minimum: 29497.60546245923 > 4.284088968628055
Armijo: th(2.154434690031884)=4.284088968628055; dx=-2.839287609608877E14 evalInputDelta=29493.321373490602
Armijo: th(1.077217345015942)=17.783043235431407; dx=-2.8392876096089244E14 evalInputDelta=29479.822419223798
Armijo: th(0.3590724483386473)=126.25723802030423; dx=-2.8392876096099375E14 evalInputDelta=29371.348224438923
Armijo: th(0.08976811208466183)=291.2565544886799; dx=-2.839287609614603E14 evalInputDelta=29206.34890797055
Armijo: th(0.017953622416932366)=401.9191991977268; dx=-2.8392876096281325E14 evalInputDelta=29095.6862632615
Armijo: th(0.002992270402822061)=456.68209736642075; dx=-2.8392876096628956E14 evalInputDelta=29040.923365092807
Armijo: th(4.2746720040315154E-4)=478.1924112851331; dx=-2.839287609738015E14 evalInputDelta=29019.413051174095
Armijo: th(5.343340005039394E-5)=484.9674637694302; dx=-2.839287609886871E14 evalInputDelta=29012.6379986898
Armijo: th(5.9370444500437714E-6)=486.5696353217759; dx=-2.83928761001437E14 evalInputDelta=29011.035827137453
Armijo: th(5.937044450043771E-7)=486.7975034307559; dx=-2.8392876100423375E14 evalInputDelta=29010.807959028472
Armijo: th(5.397313136403428E-8)=486.8213835333578; dx=-2.839287610045477E14 evalInputDelta=29010.78407892587
Armijo: th(4.4977609470028565E-9)=486.82358105631937; dx=-2.839287610045768E14 evalInputDelta=29010.78188140291
Armijo: th(3.4598161130791205E-10)=489.19831431190397; dx=-2.5216028393052786E20 evalInputDelta=29008.407148147326
Armijo: th(2.4712972236279432E-11)=891.2453578517179; dx=-5.699718402902214E23 evalInputDelta=28606.360104607513
Armijo: th(1.6475314824186289E-12)=23907.300427620346; dx=-2.181169280014673E26 evalInputDelta=5590.305034838882
Armijo: th(1.029707176511643E-13)=29200.436894377915; dx=-2.7969759488074517E26 evalInputDelta=297.1685680813134
Armijo: th(6.057101038303783E-15)=29474.310232930402; dx=-2.8358923584018097E26 evalInputDelta=23.295229528826894
MIN ALPHA (3.3650561323909904E-16): th(2.154434690031884)=4.284088968628055
Fitness changed from 29497.60546245923 to 4.284088968628055
Iteration 1 complete. Error: 4.284088968628055 Total: 0.6035; Orientation: 0.0112; Line Search: 0.5058
th(0)=4.284088968628055;dx=-564.668192838964
New Minimum: 4.284088968628055 > 4.284088968628049
WOLFE (weak): th(2.154434690031884E-15)=4.284088968628049; dx=-564.668192838964 evalInputDelta=5.329070518200751E-15
New Minimum: 4.284088968628049 > 4.284088968628044
WOLFE (weak): th(4.308869380063768E-15)=4.284088968628044; dx=-564.668192838964 evalInputDelta=1.0658141036401503E-14
New Minimum: 4.284088968628044 > 4.284088968628026
WOLFE (weak): th(1.2926608140191303E-14)=4.284088968628026; dx=-564.668192838964 evalInputDelta=2.8421709430404007E-14
New Minimum: 4.284088968628026 > 4.284088968627939
WOLFE (weak): th(5.1706432560765214E-14)=4.284088968627939; dx=-564.668192838964 evalInputDelta=1.1546319456101628E-13
New Minimum: 4.284088968627939 > 4.284088968627481
WOLFE (weak): th(2.5853216280382605E-13)=4.284088968627481; dx=-564.668192838964 evalInputDelta=5.737632591262809E-13
New Minimum: 4.284088968627481 > 4.2840889686246175
WOLFE (weak): th(1.5511929768229563E-12)=4.2840889686246175; dx=-564.6681928389637 evalInputDelta=3.4372504842394846E-12
New Minimum: 4.2840889686246175 > 4.284088968603995
WOLFE (weak): th(1.0858350837760695E-11)=4.284088968603995; dx=-564.6681928389617 evalInputDelta=2.4059865211256692E-11
New Minimum: 4.284088968603995 > 4.284088968435578
WOLFE (weak): th(8.686680670208556E-11)=4.284088968435578; dx=-564.6681928389445 evalInputDelta=1.9247714533321414E-10
New Minimum: 4.284088968435578 > 4.284088966895752
WOLFE (weak): th(7.8180126031877E-10)=4.284088966895752; dx=-564.6681928387881 evalInputDelta=1.7323031897831243E-9
New Minimum: 4.284088966895752 > 4.284088951305028
WOLFE (weak): th(7.818012603187701E-9)=4.284088951305028; dx=-564.6681928372041 evalInputDelta=1.7323026568760724E-8
New Minimum: 4.284088951305028 > 4.284088778074767
WOLFE (weak): th(8.599813863506471E-8)=4.284088778074767; dx=-564.6681928196036 evalInputDelta=1.9055328781547587E-7
New Minimum: 4.284088778074767 > 4.284086681988706
WOLFE (weak): th(1.0319776636207765E-6)=4.284086681988706; dx=-564.6681926066387 evalInputDelta=2.286639348980657E-6
New Minimum: 4.284086681988706 > 4.284059242335198
WOLFE (weak): th(1.3415709627070094E-5)=4.284059242335198; dx=-564.6681898187392 evalInputDelta=2.9726292856580017E-5
New Minimum: 4.284059242335198 > 4.283672804215186
WOLFE (weak): th(1.878199347789813E-4)=4.283672804215186; dx=-564.6681505565879 evalInputDelta=4.161644128686248E-4
New Minimum: 4.283672804215186 > 4.2778473361312646
WOLFE (weak): th(0.0028172990216847197)=4.2778473361312646; dx=-564.6675587778399 evalInputDelta=0.006241632496790217
New Minimum: 4.2778473361312646 > 4.185751609729415
WOLFE (weak): th(0.045076784346955515)=4.185751609729415; dx=-564.6582400470259 evalInputDelta=0.09833735889863959
New Minimum: 4.185751609729415 > 2.980967081219452
WOLFE (weak): th(0.7663053338982437)=2.980967081219452; dx=-564.5419195383957 evalInputDelta=1.3031218874086026
New Minimum: 2.980967081219452 > 0.13211177957657042
WOLFE (weak): th(13.793496010168386)=0.13211177957657042; dx=-564.3343245235089 evalInputDelta=4.151977189051484
New Minimum: 0.13211177957657042 > 6.480671230069525E-5
WOLFE (weak): th(262.07642419319933)=6.480671230069525E-5; dx=-564.3311274635787 evalInputDelta=4.284024161915754
New Minimum: 6.480671230069525E-5 > 0.0
WOLFE (weak): th(5241.528483863986)=0.0; dx=-564.3311273587926 evalInputDelta=4.284088968628055
Armijo: th(110072.09816114372)=0.0; dx=-564.3311273587926 evalInputDelta=4.284088968628055
Armijo: th(57656.813322503855)=0.0; dx=-564.3311273587926 evalInputDelta=4.284088968628055
Armijo: th(31449.17090318392)=0.0; dx=-564.3311273587926 evalInputDelta=4.284088968628055
Armijo: th(18345.349693523953)=0.0; dx=-564.3311273587926 evalInputDelta=4.284088968628055
Armijo: th(11793.43908869397)=0.0; dx=-564.3311273587926 evalInputDelta=4.284088968628055
Armijo: th(8517.483786278977)=0.0; dx=-564.3311273587926 evalInputDelta=4.284088968628055
WOLFE (weak): th(6879.506135071482)=0.0; dx=-564.3311273587926 evalInputDelta=4.284088968628055
Armijo: th(7698.49496067523)=0.0; dx=-564.3311273587926 evalInputDelta=4.284088968628055
WOLFE (weak): th(7289.0005478733565)=0.0; dx=-564.3311273587926 evalInputDelta=4.284088968628055
WOLFE (weak): th(7493.747754274293)=0.0; dx=-564.3311273587926 evalInputDelta=4.284088968628055
Armijo: th(7596.121357474762)=0.0; dx=-564.3311273587926 evalInputDelta=4.284088968628055
WOLFE (weak): th(7544.934555874527)=0.0; dx=-564.3311273587926 evalInputDelta=4.284088968628055
mu ~= nu (7544.934555874527): th(5241.528483863986)=0.0
Fitness changed from 4.284088968628055 to 0.0
Iteration 2 complete. Error: 0.0 Total: 1.9686; Orientation: 0.0031; Line Search: 1.9561
th(0)=0.0;dx=-564.0867232
Armijo: th(16310.208051716048)=0.0; dx=-564.0867232 evalInputDelta=0.0
Armijo: th(8155.104025858024)=0.0; dx=-564.0867232 evalInputDelta=0.0
Armijo: th(2718.3680086193413)=0.0; dx=-564.0867232 evalInputDelta=0.0
Armijo: th(679.5920021548353)=0.0; dx=-564.0867232 evalInputDelta=0.0
Armijo: th(135.91840043096707)=0.0; dx=-564.0867232 evalInputDelta=0.0
Armijo: th(22.653066738494513)=0.0; dx=-564.0867232 evalInputDelta=0.0
Armijo: th(3.2361523912135017)=0.0; dx=-564.0867232 evalInputDelta=0.0
Armijo: th(0.4045190489016877)=0.0; dx=-564.0867232 evalInputDelta=0.0
Armijo: th(0.044946560989076415)=0.0; dx=-564.0867232 evalInputDelta=0.0
Armijo: th(0.004494656098907641)=0.0; dx=-564.0867232 evalInputDelta=0.0
Armijo: th(4.086050999006947E-4)=0.0; dx=-564.0867232 evalInputDelta=0.0
Armijo: th(3.405042499172456E-5)=0.0; dx=-564.0867232 evalInputDelta=0.0
Armijo: th(2.619263460901889E-6)=0.0; dx=-564.0867232 evalInputDelta=0.0
Armijo: th(1.870902472072778E-7)=0.0; dx=-564.0867232 evalInputDelta=0.0
Armijo: th(1.2472683147151853E-8)=0.0; dx=-564.0867232 evalInputDelta=0.0
Armijo: th(7.795426966969908E-10)=0.0; dx=-564.0867232 evalInputDelta=0.0
Armijo: th(4.585545274688181E-11)=0.0; dx=-564.0867232 evalInputDelta=0.0
Armijo: th(2.547525152604545E-12)=0.0; dx=-564.0867232 evalInputDelta=0.0
Armijo: th(1.340802711897129E-13)=0.0; dx=-564.0867232 evalInputDelta=0.0
Armijo: th(6.7040135594856445E-15)=0.0; dx=-564.0867232 evalInputDelta=0.0
MIN ALPHA (3.1923874092788784E-16): th(0.0)=0.0
Fitness changed from 0.0 to 0.0
Static Iteration Total: 0.2648; Orientation: 0.0563; Line Search: 0.2032
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 2.838s (< 30.000s)
Returns
0.0
First, we use a conjugate gradient descent method, which converges the fastest for purely linear functions.
TrainingTester.java:452 executed in 2.49 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();
}
Reset training subject: 1377101378975
Reset training subject: 1377106202747
Constructing line search parameters: GD
F(0.0) = LineSearchPoint{point=PointSample{avg=29497.60546245923}, derivative=-2.8392876096000002E26}
New Minimum: 29497.60546245923 > 511.61906650076133
F(1.0E-10) = LineSearchPoint{point=PointSample{avg=511.61906650076133}, derivative=-8.001920284002099E21}, evalInputDelta = -28985.98639595847
New Minimum: 511.61906650076133 > 487.4724757129249
F(7.000000000000001E-10) = LineSearchPoint{point=PointSample{avg=487.4724757129249}, derivative=-2.9440283928891113E19}, evalInputDelta = -29010.132986746303
New Minimum: 487.4724757129249 > 486.82356318445875
F(4.900000000000001E-9) = LineSearchPoint{point=PointSample{avg=486.82356318445875}, derivative=-2.8392876100457656E14}, evalInputDelta = -29010.78189927477
New Minimum: 486.82356318445875 > 486.82225717261935
F(3.430000000000001E-8) = LineSearchPoint{point=PointSample{avg=486.82225717261935}, derivative=-2.8392876100455925E14}, evalInputDelta = -29010.78320528661
New Minimum: 486.82225717261935 > 486.81312929288816
F(2.4010000000000004E-7) = LineSearchPoint{point=PointSample{avg=486.81312929288816}, derivative=-2.839287610044387E14}, evalInputDelta = -29010.79233316634
New Minimum: 486.81312929288816 > 486.7499156148626
F(1.6807000000000003E-6) = LineSearchPoint{point=PointSample{avg=486.7499156148626}, derivative=-2.839287610036204E14}, evalInputDelta = -29010.855546844366
New Minimum: 486.7499156148626 > 486.3365885829397
F(1.1764900000000001E-5) = LineSearchPoint{point=PointSample{avg=486.3365885829397}, derivative=-2.839287609989096E14}, evalInputDelta = -29011.268873876288
New Minimum: 486.3365885829397 > 484.19761003244423
F(8.235430000000001E-5) = LineSearchPoint{point=PointSample{avg=484.19761003244423}, derivative=-2.839287609851029E14}, evalInputDelta = -29013.407852426786
New Minimum: 484.19761003244423 > 476.27586987160464
F(5.764801000000001E-4) = LineSearchPoint{point=PointSample{avg=476.27586987160464}, derivative=-2.8392876097231275E14}, evalInputDelta = -29021.329592587623
New Minimum: 476.27586987160464 > 450.70650846601467
F(0.004035360700000001) = LineSearchPoint{point=PointSample{avg=450.70650846601467}, derivative=-2.8392876096551256E14}, evalInputDelta = -29046.898953993215
New Minimum: 450.70650846601467 > 378.56893290950165
F(0.028247524900000005) = LineSearchPoint{point=PointSample{avg=378.56893290950165}, derivative=-2.839287609623162E14}, evalInputDelta = -29119.03652954973
New Minimum: 378.56893290950165 > 204.91012423230055
F(0.19773267430000002) = LineSearchPoint{point=PointSample{avg=204.91012423230055}, derivative=-2.839287609611411E14}, evalInputDelta = -29292.695338226928
New Minimum: 204.91012423230055 > 10.714844774798445
F(1.3841287201) = LineSearchPoint{point=PointSample{avg=10.714844774798445}, derivative=-2.8392876096088956E14}, evalInputDelta = -29486.890617684432
New Minimum: 10.714844774798445 > 0.22912107149782393
F(9.688901040700001) = LineSearchPoint{point=PointSample{avg=0.22912107149782393}, derivative=-2.8392876096088706E14}, evalInputDelta = -29497.37634138773
New Minimum: 0.22912107149782393 > 0.002162117929530509
F(67.8223072849) = LineSearchPoint{point=PointSample{avg=0.002162117929530509}, derivative=-2.8392876096088706E14}, evalInputDelta = -29497.6033003413
New Minimum: 0.002162117929530509 > 0.0
F(474.7561509943) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-2.8392876096088706E14}, evalInputDelta = -29497.60546245923
F(3323.2930569601003) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-2.8392876096088706E14}, evalInputDelta = -29497.60546245923
F(23263.0513987207) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-2.8392876096088706E14}, evalInputDelta = -29497.60546245923
F(162841.3597910449) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-2.8392876096088706E14}, evalInputDelta = -29497.60546245923
F(1139889.5185373144) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-2.8392876096088706E14}, evalInputDelta = -29497.60546245923
F(7979226.6297612) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-2.8392876096088706E14}, evalInputDelta = -29497.60546245923
F(5.58545864083284E7) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-2.8392876096088706E14}, evalInputDelta = -29497.60546245923
F(3.909821048582988E8) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-2.8392876096088706E14}, evalInputDelta = -29497.60546245923
F(2.7368747340080914E9) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-2.8392876096088706E14}, evalInputDelta = -29497.60546245923
F(1.915812313805664E10) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-2.8392876096088706E14}, evalInputDelta = -29497.60546245923
0.0 <= 29497.60546245923
F(1.0E10) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-2.8392876096088706E14}, evalInputDelta = -29497.60546245923
Right bracket at 1.0E10
Converged to right
Fitness changed from 29497.60546245923 to 0.0
Iteration 1 complete. Error: 0.0 Total: 1.2940; Orientation: 0.0025; Line Search: 1.2771
F(0.0) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-564.0867232}
F(1.0E10) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-564.0867232}, evalInputDelta = 0.0
0.0 <= 0.0
F(5.0E9) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-564.0867232}, evalInputDelta = 0.0
Right bracket at 5.0E9
F(2.5E9) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-564.0867232}, evalInputDelta = 0.0
Right bracket at 2.5E9
F(1.25E9) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-564.0867232}, evalInputDelta = 0.0
Right bracket at 1.25E9
F(6.25E8) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-564.0867232}, evalInputDelta = 0.0
Right bracket at 6.25E8
F(3.125E8) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-564.0867232}, evalInputDelta = 0.0
Right bracket at 3.125E8
F(1.5625E8) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-564.0867232}, evalInputDelta = 0.0
Right bracket at 1.5625E8
F(7.8125E7) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-564.0867232}, evalInputDelta = 0.0
Right bracket at 7.8125E7
F(3.90625E7) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-564.0867232}, evalInputDelta = 0.0
Right bracket at 3.90625E7
F(1.953125E7) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-564.0867232}, evalInputDelta = 0.0
Right bracket at 1.953125E7
F(9765625.0) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-564.0867232}, evalInputDelta = 0.0
Right bracket at 9765625.0
F(4882812.5) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-564.0867232}, evalInputDelta = 0.0
Right bracket at 4882812.5
F(2441406.25) = LineSearchPoint{point=PointSample{avg=0.0}, derivative=-564.0867232}, evalInputDelta = 0.0
Loops = 12
Fitness changed from 0.0 to 0.0
Static Iteration Total: 1.1903; Orientation: 0.0019; Line Search: 1.1847
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 2.484s (< 30.000s)
Returns
0.0
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 48.66 seconds (0.083 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();
}
Reset training subject: 1379591547678
Reset training subject: 1379595916164
Adding measurement 3f9c14e5 to history. Total: 0
LBFGS Accumulation History: 1 points
Constructing line search parameters: GD
Non-optimal measurement 29497.60546245923 < 29497.60546245923. Total: 1
th(0)=29497.60546245923;dx=-2.8392876096000002E26
Adding measurement 3b8d8bae to history. Total: 1
New Minimum: 29497.60546245923 > 4.284088968628055
Armijo: th(2.154434690031884)=4.284088968628055; dx=-2.839287609608877E14 evalInputDelta=29493.321373490602
Non-optimal measurement 17.783043235431407 < 4.284088968628055. Total: 2
Armijo: th(1.077217345015942)=17.783043235431407; dx=-2.8392876096089244E14 evalInputDelta=29479.822419223798
Non-optimal measurement 126.25723802030423 < 4.284088968628055. Total: 2
Armijo: th(0.3590724483386473)=126.25723802030423; dx=-2.8392876096099375E14 evalInputDelta=29371.348224438923
Non-optimal measurement 291.2565544886799 < 4.284088968628055. Total: 2
Armijo: th(0.08976811208466183)=291.2565544886799; dx=-2.839287609614603E14 evalInputDelta=29206.34890797055
Non-optimal measurement 401.9191991977268 < 4.284088968628055. Total: 2
Armijo: th(0.017953622416932366)=401.9191991977268; dx=-2.8392876096281325E14 evalInputDelta=29095.6862632615
Non-optimal measurement 456.68209736642075 < 4.284088968628055. Total: 2
Armijo: th(0.002992270402822061)=456.68209736642075; dx=-2.8392876096628956E14 evalInputDelta=29040.923365092807
Non-optimal measurement 478.1924112851331 < 4.284088968628055. Total: 2
Armijo: th(4.2746720040315154E-4)=478.1924112851331; dx=-2.839287609738015E14 evalInputDelta=29019.413051174095
Non-optimal measurement 484.9674637694302 < 4.284088968628055. Total: 2
Armijo: th(5.343340005039394E-5)=484.9674637694302; dx=-2.839287609886871E14 evalInputDelta=29012.6379986898
Non-optimal measurement 486.5696353217759 < 4.284088968628055. Total: 2
Armijo: th(5.9370444500437714E-6)=486.5696353217759; dx=-2.83928761001437E14 evalInputDelta=29011.035827137453
Non-optimal measurement 486.7975034307559 < 4.284088968628055. Total: 2
Armijo: th(5.937044450043771E-7)=486.7975034307559; dx=-2.8392876100423375E14 evalInputDelta=29010.807959028472
Non-optimal measurement 486.8213835333578 < 4.284088968628055. Total: 2
Armijo: th(5.397313136403428E-8)=486.8213835333578; dx=-2.839287610045477E14 evalInputDelta=29010.78407892587
Non-optimal measurement 486.82358105631937 < 4.284088968628055. Total: 2
Armijo: th(4.4977609470028565E-9)=486.82358105631937; dx=-2.839287610045768E14 evalInputDelta=29010.78188140291
Non-optimal measurement 489.19831431190397 < 4.284088968628055. Total: 2
Armijo: th(3.4598161130791205E-10)=489.19831431190397; dx=-2.5216028393052786E20 evalInputDelta=29008.407148147326
Non-optimal measurement 891.2453578517179 < 4.284088968628055. Total: 2
Armijo: th(2.4712972236279432E-11)=891.2453578517179; dx=-5.699718402902214E23 evalInputDelta=28606.360104607513
Non-optimal measurement 23907.300427620346 < 4.284088968628055. Total: 2
Armijo: th(1.6475314824186289E-12)=23907.300427620346; dx=-2.181169280014673E26 evalInputDelta=5590.305034838882
Non-optimal measurement 29200.436894377915 < 4.284088968628055. Total: 2
Armijo: th(1.029707176511643E-13)=29200.436894377915; dx=-2.7969759488074517E26 evalInputDelta=297.1685680813134
Non-optimal measurement 29474.310232930402 < 4.284088968628055. Total: 2
Armijo: th(6.057101038303783E-15)=29474.310232930402; dx=-2.8358923584018097E26 evalInputDelta=23.295229528826894
Non-optimal measurement 4.284088968628055 < 4.284088968628055. Total: 2
MIN ALPHA (3.3650561323909904E-16): th(2.154434690031884)=4.284088968628055
Fitness changed from 29497.60546245923 to 4.284088968628055
Iteration 1 complete. Error: 4.284088968628055 Total: 1.1350; Orientation: 0.0153; Line Search: 1.0902
Non-optimal measurement 4.284088968628055 < 4.284088968628055. Total: 2
LBFGS Accumulation History: 2 points
Non-optimal measurement 4.284088968628055 < 4.284088968628055. Total: 2
th(0)=4.284088968628055;dx=-564.668192838964
Adding measurement 761133e5 to history. Total: 2
New Minimum: 4.284088968628055 > 4.284088968628049
WOLFE (weak): th(2.154434690031884E-15)=4.284088968628049; dx=-564.668192838964 evalInputDelta=5.329070518200751E-15
Adding measurement 7557deeb to history. Total: 3
New Minimum: 4.284088968628049 > 4.284088968628044
WOLFE (weak): th(4.308869380063768E-15)=4.284088968628044; dx=-564.668192838964 evalInputDelta=1.0658141036401503E-14
Adding measurement 44779ba to history. Total: 4
New Minimum: 4.284088968628044 > 4.284088968628026
WOLFE (weak): th(1.2926608140191303E-14)=4.284088968628026; dx=-564.668192838964 evalInputDelta=2.8421709430404007E-14
Adding measurement 296234f0 to history. Total: 5
New Minimum: 4.284088968628026 > 4.284088968627939
WOLFE (weak): th(5.1706432560765214E-14)=4.284088968627939; dx=-564.668192838964 evalInputDelta=1.1546319456101628E-13
Adding measurement 71e954bd to history. Total: 6
New Minimum: 4.284088968627939 > 4.284088968627481
WOLFE (weak): th(2.5853216280382605E-13)=4.284088968627481; dx=-564.668192838964 evalInputDelta=5.737632591262809E-13
Adding measurement 5d2bce91 to history. Total: 7
New Minimum: 4.284088968627481 > 4.2840889686246175
WOLFE (weak): th(1.5511929768229563E-12)=4.2840889686246175; dx=-564.6681928389637 evalInputDelta=3.4372504842394846E-12
Adding measurement 5ccb1b9c to history. Total: 8
New Minimum: 4.2840889686246175 > 4.284088968603995
WOLFE (weak): th(1.0858350837760695E-11)=4.284088968603995; dx=-564.6681928389617 evalInputDelta=2.4059865211256692E-11
Adding measurement 42b6b638 to history. Total: 9
New Minimum: 4.284088968603995 > 4.284088968435578
WOLFE (weak): th(8.686680670208556E-11)=4.284088968435578; dx=-564.6681928389445 evalInputDelta=1.9247714533321414E-10
Adding measurement 187897f3 to history. Total: 10
New Minimum: 4.284088968435578 > 4.284088966895752
WOLFE (weak): th(7.8180126031877E-10)=4.284088966895752; dx=-564.6681928387881 evalInputDelta=1.7323031897831243E-9
Adding measurement 14bf1e73 to history. Total: 11
New Minimum: 4.284088966895752 > 4.284088951305028
WOLFE (weak): th(7.818012603187701E-9)=4.284088951305028; dx=-564.6681928372041 evalInputDelta=1.7323026568760724E-8
Adding measurement 6a8d90df to history. Total: 12
New Minimum: 4.284088951305028 > 4.284088778074767
WOLFE (weak): th(8.599813863506471E-8)=4.284088778074767; dx=-564.6681928196036 evalInputDelta=1.9055328781547587E-7
Adding measurement 66f68d67 to history. Total: 13
New Minimum: 4.284088778074767 > 4.284086681988706
WOLFE (weak): th(1.0319776636207765E-6)=4.284086681988706; dx=-564.6681926066387 evalInputDelta=2.286639348980657E-6
Adding measurement 5c7f5221 to history. Total: 14
New Minimum: 4.284086681988706 > 4.284059242335198
WOLFE (weak): th(1.3415709627070094E-5)=4.284059242335198; dx=-564.6681898187392 evalInputDelta=2.9726292856580017E-5
Adding measurement 7118414a to history. Total: 15
New Minimum: 4.284059242335198 > 4.283672804215186
WOLFE (weak): th(1.878199347789813E-4)=4.283672804215186; dx=-564.6681505565879 evalInputDelta=4.161644128686248E-4
Adding measurement 70c3915b to history. Total: 16
New Minimum: 4.283672804215186 > 4.2778473361312646
WOLFE (weak): th(0.0028172990216847197)=4.2778473361312646; dx=-564.6675587778399 evalInputDelta=0.006241632496790217
Adding measurement 4741f36c to history. Total: 17
New Minimum: 4.2778473361312646 > 4.185751609729415
WOLFE (weak): th(0.045076784346955515)=4.185751609729415; dx=-564.6582400470259 evalInputDelta=0.09833735889863959
Adding measurement 3ff189e1 to history. Total: 18
New Minimum: 4.185751609729415 > 2.980967081219452
WOLFE (weak): th(0.7663053338982437)=2.980967081219452; dx=-564.5419195383957 evalInputDelta=1.3031218874086026
Adding measurement 637f0e to history. Total: 19
New Minimum: 2.980967081219452 > 0.13211177957657042
WOLFE (weak): th(13.793496010168386)=0.13211177957657042; dx=-564.3343245235089 evalInputDelta=4.151977189051484
Adding measurement 42b663ce to history. Total: 20
New Minimum: 0.13211177957657042 > 6.480671230069525E-5
WOLFE (weak): th(262.07642419319933)=6.480671230069525E-5; dx=-564.3311274635787 evalInputDelta=4.284024161915754
Adding measurement 2743ea to hist
...skipping 10032 bytes...
88706, 4.284088778074767, 4.284088951305028, 4.284088966895752
Rejected: LBFGS Orientation magnitude: 1.039e+05, gradient 2.375e+01, dot -0.956; [9be6f212-03c0-42da-bb08-8a0f6d7670c9 = 1.000/1.000e+00, 1051bb21-b2d3-4352-b32d-b2d7d147b145 = 1.000/1.000e+00, fed809df-46fe-4d7f-a9f8-02092e8569d2 = 1.000/1.000e+00, b51f2234-3a32-4f0f-b0c1-218655fd9258 = 1.000/1.000e+00, dd01f57c-1400-4662-a923-53136657641b = 1.000/1.000e+00]
Orientation rejected. Popping history element from 0.0, 6.480671230069525E-5, 0.13211177957657042, 2.980967081219452, 4.185751609729415, 4.2778473361312646, 4.283672804215186, 4.284059242335198, 4.284086681988706, 4.284088778074767, 4.284088951305028
Rejected: LBFGS Orientation magnitude: 1.039e+05, gradient 2.375e+01, dot -0.956; [1051bb21-b2d3-4352-b32d-b2d7d147b145 = 1.000/1.000e+00, b51f2234-3a32-4f0f-b0c1-218655fd9258 = 1.000/1.000e+00, dd01f57c-1400-4662-a923-53136657641b = 1.000/1.000e+00, 9be6f212-03c0-42da-bb08-8a0f6d7670c9 = 1.000/1.000e+00, fed809df-46fe-4d7f-a9f8-02092e8569d2 = 1.000/1.000e+00]
Orientation rejected. Popping history element from 0.0, 6.480671230069525E-5, 0.13211177957657042, 2.980967081219452, 4.185751609729415, 4.2778473361312646, 4.283672804215186, 4.284059242335198, 4.284086681988706, 4.284088778074767
Rejected: LBFGS Orientation magnitude: 1.039e+05, gradient 2.375e+01, dot -0.956; [dd01f57c-1400-4662-a923-53136657641b = 1.000/1.000e+00, 9be6f212-03c0-42da-bb08-8a0f6d7670c9 = 1.000/1.000e+00, fed809df-46fe-4d7f-a9f8-02092e8569d2 = 1.000/1.000e+00, b51f2234-3a32-4f0f-b0c1-218655fd9258 = 1.000/1.000e+00, 1051bb21-b2d3-4352-b32d-b2d7d147b145 = 1.000/1.000e+00]
Orientation rejected. Popping history element from 0.0, 6.480671230069525E-5, 0.13211177957657042, 2.980967081219452, 4.185751609729415, 4.2778473361312646, 4.283672804215186, 4.284059242335198, 4.284086681988706
Rejected: LBFGS Orientation magnitude: 1.451e+05, gradient 2.375e+01, dot -0.944; [dd01f57c-1400-4662-a923-53136657641b = 1.000/1.000e+00, fed809df-46fe-4d7f-a9f8-02092e8569d2 = 1.000/1.000e+00, 9be6f212-03c0-42da-bb08-8a0f6d7670c9 = 1.000/1.000e+00, 1051bb21-b2d3-4352-b32d-b2d7d147b145 = 1.000/1.000e+00, b51f2234-3a32-4f0f-b0c1-218655fd9258 = 1.000/1.000e+00]
Orientation rejected. Popping history element from 0.0, 6.480671230069525E-5, 0.13211177957657042, 2.980967081219452, 4.185751609729415, 4.2778473361312646, 4.283672804215186, 4.284059242335198
Rejected: LBFGS Orientation magnitude: 1.181e+05, gradient 2.375e+01, dot -1.000; [fed809df-46fe-4d7f-a9f8-02092e8569d2 = 1.000/1.000e+00, 9be6f212-03c0-42da-bb08-8a0f6d7670c9 = 1.000/1.000e+00, dd01f57c-1400-4662-a923-53136657641b = 1.000/1.000e+00, 1051bb21-b2d3-4352-b32d-b2d7d147b145 = 1.000/1.000e+00, b51f2234-3a32-4f0f-b0c1-218655fd9258 = 1.000/1.000e+00]
Orientation rejected. Popping history element from 0.0, 6.480671230069525E-5, 0.13211177957657042, 2.980967081219452, 4.185751609729415, 4.2778473361312646, 4.283672804215186
Rejected: LBFGS Orientation magnitude: 1.215e+05, gradient 2.375e+01, dot -1.000; [9be6f212-03c0-42da-bb08-8a0f6d7670c9 = 1.000/1.000e+00, dd01f57c-1400-4662-a923-53136657641b = 1.000/1.000e+00, 1051bb21-b2d3-4352-b32d-b2d7d147b145 = 1.000/1.000e+00, b51f2234-3a32-4f0f-b0c1-218655fd9258 = 1.000/1.000e+00, fed809df-46fe-4d7f-a9f8-02092e8569d2 = 1.000/1.000e+00]
Orientation rejected. Popping history element from 0.0, 6.480671230069525E-5, 0.13211177957657042, 2.980967081219452, 4.185751609729415, 4.2778473361312646
Rejected: LBFGS Orientation magnitude: 1.661e+05, gradient 2.375e+01, dot -1.000; [dd01f57c-1400-4662-a923-53136657641b = 1.000/1.000e+00, b51f2234-3a32-4f0f-b0c1-218655fd9258 = 1.000/1.000e+00, 9be6f212-03c0-42da-bb08-8a0f6d7670c9 = 1.000/1.000e+00, fed809df-46fe-4d7f-a9f8-02092e8569d2 = 1.000/1.000e+00, 1051bb21-b2d3-4352-b32d-b2d7d147b145 = 1.000/1.000e+00]
Orientation rejected. Popping history element from 0.0, 6.480671230069525E-5, 0.13211177957657042, 2.980967081219452, 4.185751609729415
Rejected: LBFGS Orientation magnitude: 1.683e+06, gradient 2.375e+01, dot -1.000; [dd01f57c-1400-4662-a923-53136657641b = 1.000/1.000e+00, 1051bb21-b2d3-4352-b32d-b2d7d147b145 = 1.000/1.000e+00, fed809df-46fe-4d7f-a9f8-02092e8569d2 = 1.000/1.000e+00, b51f2234-3a32-4f0f-b0c1-218655fd9258 = 1.000/1.000e+00, 9be6f212-03c0-42da-bb08-8a0f6d7670c9 = 1.000/1.000e+00]
Orientation rejected. Popping history element from 0.0, 6.480671230069525E-5, 0.13211177957657042, 2.980967081219452
LBFGS Accumulation History: 3 points
Removed measurement 2743ea to history. Total: 21
Removed measurement 42b663ce to history. Total: 20
Removed measurement 637f0e to history. Total: 19
Removed measurement 3ff189e1 to history. Total: 18
Removed measurement 4741f36c to history. Total: 17
Removed measurement 70c3915b to history. Total: 16
Removed measurement 7118414a to history. Total: 15
Removed measurement 5c7f5221 to history. Total: 14
Removed measurement 66f68d67 to history. Total: 13
Removed measurement 6a8d90df to history. Total: 12
Removed measurement 14bf1e73 to history. Total: 11
Removed measurement 187897f3 to history. Total: 10
Removed measurement 42b6b638 to history. Total: 9
Removed measurement 5ccb1b9c to history. Total: 8
Removed measurement 5d2bce91 to history. Total: 7
Removed measurement 71e954bd to history. Total: 6
Removed measurement 296234f0 to history. Total: 5
Removed measurement 44779ba to history. Total: 4
Removed measurement 7557deeb to history. Total: 3
Adding measurement 4acf52c9 to history. Total: 3
th(0)=0.0;dx=-564.0867232
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(16310.208051716048)=0.0; dx=-564.0867232 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(8155.104025858024)=0.0; dx=-564.0867232 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(2718.3680086193413)=0.0; dx=-564.0867232 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(679.5920021548353)=0.0; dx=-564.0867232 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(135.91840043096707)=0.0; dx=-564.0867232 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(22.653066738494513)=0.0; dx=-564.0867232 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(3.2361523912135017)=0.0; dx=-564.0867232 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(0.4045190489016877)=0.0; dx=-564.0867232 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(0.044946560989076415)=0.0; dx=-564.0867232 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(0.004494656098907641)=0.0; dx=-564.0867232 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(4.086050999006947E-4)=0.0; dx=-564.0867232 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(3.405042499172456E-5)=0.0; dx=-564.0867232 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(2.619263460901889E-6)=0.0; dx=-564.0867232 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(1.870902472072778E-7)=0.0; dx=-564.0867232 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(1.2472683147151853E-8)=0.0; dx=-564.0867232 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(7.795426966969908E-10)=0.0; dx=-564.0867232 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(4.585545274688181E-11)=0.0; dx=-564.0867232 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(2.547525152604545E-12)=0.0; dx=-564.0867232 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(1.340802711897129E-13)=0.0; dx=-564.0867232 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
Armijo: th(6.7040135594856445E-15)=0.0; dx=-564.0867232 evalInputDelta=0.0
Non-optimal measurement 0.0 < 0.0. Total: 4
MIN ALPHA (3.1923874092788784E-16): th(0.0)=0.0
Fitness changed from 0.0 to 0.0
Static Iteration Total: 45.7148; Orientation: 44.4437; Line Search: 1.2671
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 48.663s (< 30.000s)
Returns
0.0
TrainingTester.java:432 executed in 0.11 seconds (0.000 gc):
return TestUtil.compare(title + " vs Iteration", runs);
Plotting range=[1.0, -0.3681415186564715], [2.0, 1.6318584813435284]; valueStats=DoubleSummaryStatistics{count=2, sum=8.568178, min=4.284089, average=4.284089, max=4.284089}
Plotting 2 points for GD
Only 1 points for CjGD
Plotting 2 points for LBFGS
Returns
TrainingTester.java:435 executed in 0.01 seconds (0.000 gc):
return TestUtil.compareTime(title + " vs Time", runs);
Plotting range=[0.0, -0.3681415186564715], [1.968, 1.6318584813435284]; valueStats=DoubleSummaryStatistics{count=2, sum=8.568178, min=4.284089, average=4.284089, max=4.284089}
Plotting 2 points for GD
Only 1 points for CjGD
Plotting 2 points for LBFGS
Returns
TrainingTester.java:255 executed in 0.00 seconds (0.000 gc):
return grid(inputLearning, modelLearning, completeLearning);
Returns
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());
| details | result |
|---|---|
| {"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": "54.648",
"gc_time": "0.308"
},
"created_on": 1586735962036,
"file_name": "trainingTest",
"report": {
"simpleName": "Big1",
"canonicalName": "com.simiacryptus.mindseye.layers.java.ReshapeLayerTest.Big1",
"link": "https://github.com/SimiaCryptus/mindseye-java/tree/93db34cedee48c0202777a2b25deddf1dfaf5731/src/test/java/com/simiacryptus/mindseye/layers/java/ReshapeLayerTest.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/ReshapeLayer/Big1/trainingTest/202004125922",
"id": "85909190-f180-47cb-b562-50ecfa568503",
"report_type": "Components",
"display_name": "Comparative Training",
"target": {
"simpleName": "ReshapeLayer",
"canonicalName": "com.simiacryptus.mindseye.layers.java.ReshapeLayer",
"link": "https://github.com/SimiaCryptus/mindseye-java/tree/93db34cedee48c0202777a2b25deddf1dfaf5731/src/main/java/com/simiacryptus/mindseye/layers/java/ReshapeLayer.java",
"javaDoc": ""
}
}