Subreport: Logs for com.simiacryptus.ref.lang.ReferenceCountingBase
Using Seed 1384131387623865344
In this apply, we use a network to learn this target input, given it's pre-evaluated output:
TrainingTester.java:332 executed in 0.04 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
[
[ [ -0.936 ], [ -0.676 ], [ 0.736 ], [ -1.992 ], [ -0.54 ], [ -0.888 ], [ -1.116 ], [ 1.46 ], ... ],
[ [ 0.04 ], [ -0.552 ], [ -0.004 ], [ 0.624 ], [ -0.036 ], [ 0.408 ], [ -0.788 ], [ -1.048 ], ... ],
[ [ 1.98 ], [ -0.424 ], [ -1.324 ], [ -1.72 ], [ 1.572 ], [ -0.08 ], [ -1.584 ], [ 1.64 ], ... ],
[ [ 0.956 ], [ -1.636 ], [ -1.456 ], [ 0.884 ], [ 0.876 ], [ -0.016 ], [ -0.148 ], [ -1.696 ], ... ],
[ [ -1.296 ], [ -1.668 ], [ -0.76 ], [ -0.56 ], [ -0.5 ], [ -1.492 ], [ -1.672 ], [ -1.968 ], ... ],
[ [ 1.912 ], [ -0.704 ], [ 0.972 ], [ 1.944 ], [ 1.628 ], [ -0.436 ], [ 1.02 ], [ 1.588 ], ... ],
[ [ -1.144 ], [ -0.288 ], [ -1.556 ], [ -0.676 ], [ 0.964 ], [ 1.22 ], [ -1.384 ], [ -0.416 ], ... ],
[ [ 1.412 ], [ 1.052 ], [ -0.208 ], [ 0.004 ], [ 1.18 ], [ -0.664 ], [ 1.224 ], [ 0.584 ], ... ],
...
]
[
[ [ 0.596 ], [ 0.732 ], [ 0.32 ], [ -0.224 ], [ -0.712 ], [ -0.412 ], [ -1.216 ], [ -1.356 ], ... ],
[ [ -0.36 ], [ 1.436 ], [ 0.788 ], [ 0.136 ], [ -0.328 ], [ -0.216 ], [ -0.32 ], [ 1.384 ], ... ],
[ [ 0.948 ], [ 1.412 ], [ -1.208 ], [ 0.86 ], [ 0.988 ], [ 1.78 ], [ -0.928 ], [ 1.5 ], ... ],
[ [ -1.748 ], [ 1.388 ], [ -1.38 ], [ 1.472 ], [ 0.784 ], [ 1.98 ], [ 0.66 ], [ 1.516 ], ... ],
[ [ -0.724 ], [ 0.424 ], [ 1.276 ], [ -1.164 ], [ -0.428 ], [ 0.228 ], [ -1.332 ], [ -1.004 ], ... ],
[ [ -1.392 ], [ 1.024 ], [ 0.444 ], [ 0.812 ], [ 1.82 ], [ 1.656 ], [ 0.24 ], [ -1.28 ], ... ],
[ [ 0.332 ], [ 1.584 ], [ 0.388 ], [ 0.036 ], [ -1.652 ], [ 0.128 ], [ -1.132 ], [ -0.672 ], ... ],
[ [ 1.0 ], [ -1.916 ], [ -1.752 ], [ 0.68 ], [ -1.484 ], [ -1.824 ], [ -1.46 ], [ 1.224 ], ... ],
...
]
[
[ [ 0.648 ], [ -0.316 ], [ -0.148 ], [ 0.284 ], [ -0.564 ], [ -0.792 ], [ 1.18 ], [ -1.728 ], ... ],
[ [ 1.684 ], [ -0.172 ], [ -1.128 ], [ -0.492 ], [ 0.168 ], [ -1.212 ], [ -0.48 ], [ -1.84 ], ... ],
[ [ 0.996 ], [ -0.228 ], [ 1.108 ], [ 1.96 ], [ -1.432 ], [ -1.324 ], [ -0.292 ], [ 1.16 ], ... ],
[ [ 1.764 ], [ 1.916 ], [ -0.496 ], [ 1.916 ], [ 1.396 ], [ -1.488 ], [ -1.052 ], [ -1.388 ], ... ],
[ [ 1.684 ], [ 1.24 ], [ 0.26 ], [ 0.78 ], [ -0.276 ], [ -1.376 ], [ 1.82 ], [ 0.004 ], ... ],
[ [ 1.352 ], [ 0.572 ], [ 1.0 ], [ 1.94 ], [ -0.76 ], [ 1.176 ], [ 1.508 ], [ 1.696 ], ... ],
[ [ -0.524 ], [ -1.424 ], [ -1.672 ], [ -1.656 ], [ 1.004 ], [ 0.112 ], [ 1.12 ], [ 1.592 ], ... ],
[ [ -0.26 ], [ -1.548 ], [ -1.896 ], [ -0.292 ], [ -1.36 ], [ 0.008 ], [ 0.432 ], [ -1.66 ], ... ],
...
]
[
[ [ 0.364 ], [ 0.936 ], [ 1.732 ], [ -0.812 ], [ 1.468 ], [ -1.136 ], [ 0.5 ], [ -0.104 ], ... ],
[ [ 1.16 ], [ -1.796 ], [ 1.828 ], [ -0.74 ], [ -0.972 ], [ -1.96 ], [ -1.08 ], [ -0.624 ], ... ],
[ [ 0.028 ], [ 1.496 ], [ 0.332 ], [ 0.012 ], [ -1.212 ], [ -0.464 ], [ -0.452 ], [ -0.608 ], ... ],
[ [ 0.788 ], [ -0.556 ], [ -1.068 ], [ -1.096 ], [ 1.94 ], [ 1.16 ], [ -0.512 ], [ 1.02 ], ... ],
[ [ 1.308 ], [ 1.808 ], [ 1.008 ], [ 1.396 ], [ -1.176 ], [ -0.652 ], [ -1.896 ], [ 0.484 ], ... ],
[ [ 1.276 ], [ -1.612 ], [ 1.572 ], [ -1.784 ], [ 1.632 ], [ 0.24 ], [ -0.856 ], [ 1.476 ], ... ],
[ [ 1.828 ], [ -1.616 ], [ -0.636 ], [ -0.428 ], [ -0.76 ], [ 0.728 ], [ 1.504 ], [ -0.3 ], ... ],
[ [ 1.66 ], [ -0.008 ], [ -0.184 ], [ 0.184 ], [ 0.004 ], [ 0.536 ], [ -1.652 ], [ 0.236 ], ... ],
...
]
[
[ [ 1.38 ], [ 0.896 ], [ -1.52 ], [ -1.664 ], [ 0.664 ], [ 0.46 ], [ 0.988 ], [ 0.512 ], ... ],
[ [ -1.728 ], [ 0.792 ], [ 1.504 ], [ -1.9 ], [ -0.448 ], [ -1.732 ], [ -1.696 ], [ 0.352 ], ... ],
[ [ 0.06 ], [ -0.324 ], [ -1.696 ], [ -1.224 ], [ 0.676 ], [ -0.84 ], [ -1.612 ], [ 1.788 ], ... ],
[ [ -1.348 ], [ 1.144 ], [ -1.748 ], [ -0.128 ], [ -0.88 ], [ -1.788 ], [ 1.344 ], [ 1.284 ], ... ],
[ [ -1.476 ], [ 0.296 ], [ -1.744 ], [ 1.548 ], [ -0.252 ], [ 1.308 ], [ 0.756 ], [ 0.484 ], ... ],
[ [ 1.312 ], [ -1.412 ], [ -1.864 ], [ -1.08 ], [ -0.116 ], [ -0.12 ], [ 1.304 ], [ -1.536 ], ... ],
[ [ -1.504 ], [ -1.316 ], [ -1.824 ], [ -0.832 ], [ -0.04 ], [ 0.16 ], [ -1.24 ], [ 1.76 ], ... ],
[ [ -0.428 ], [ -0.46 ], [ 1.656 ], [ 1.436 ], [ -0.98 ], [ 0.884 ], [ -0.8 ], [ -1.58 ], ... ],
...
]
[
[ [ -1.38 ], [ -1.428 ], [ -1.856 ], [ 1.468 ], [ -0.604 ], [ 0.592 ], [ 1.124 ], [ -0.692 ], ... ],
[ [ 0.372 ], [ -1.06 ], [ -0.816 ], [ 0.512 ], [ -0.144 ], [ 0.052 ], [ -1.948 ], [ -1.512 ], ... ],
[ [ 0.0 ], [ 1.06 ], [ 1.248 ], [ -0.816 ], [ -1.228 ], [ 0.528 ], [ -1.612 ], [ 1.52 ], ... ],
[ [ -0.016 ], [ -0.82 ], [ -0.196 ], [ -0.076 ], [ -1.98 ], [ -0.572 ], [ 0.3 ], [ 1.512 ], ... ],
[ [ 0.772 ], [ 0.596 ], [ -1.464 ], [ -1.828 ], [ 1.492 ], [ 0.36 ], [ -0.632 ], [ -1.72 ], ... ],
[ [ -0.516 ], [ 1.684 ], [ 1.968 ], [ -0.792 ], [ 1.892 ], [ 1.912 ], [ -0.272 ], [ -0.876 ], ... ],
[ [ 1.624 ], [ -0.048 ], [ 1.812 ], [ -0.296 ], [ 1.804 ], [ 0.052 ], [ 1.092 ], [ 1.892 ], ... ],
[ [ 1.82 ], [ 0.928 ], [ 1.98 ], [ 0.036 ], [ 1.92 ], [ -1.288 ], [ 0.78 ], [ -0.228 ], ... ],
...
]
[
[ [ 0.812 ], [ 0.316 ], [ -0.316 ], [ -1.56 ], [ 1.976 ], [ -0.16 ], [ 0.172 ], [ 0.536 ], ... ],
[ [ -0.3 ], [ 1.556 ], [ 1.792 ], [ -0.108 ], [ -0.316 ], [ 1.392 ], [ 1.644 ], [ 1.848 ], ... ],
[ [ -1.568 ], [ -0.5 ], [ -1.18 ], [ 1.864 ], [ 1.348 ], [ 1.208 ], [ 0.764 ], [ 0.184 ], ... ],
[ [ -0.948 ], [ -0.624 ], [ -1.036 ], [ -1.12 ], [ -1.36 ], [ 0.0 ], [ 0.504 ], [ 1.776 ], ... ],
[ [ 1.76 ], [ 1.468 ], [ 1.128 ], [ -0.708 ], [ 0.232 ], [ -1.336 ], [ -1.184 ], [ -0.932 ], ... ],
[ [ -0.66 ], [ -1.704 ], [ -0.756 ], [ 1.676 ], [ -0.048 ], [ 0.964 ], [ 1.22 ], [ 1.636 ], ... ],
[ [ 1.152 ], [ -1.12 ], [ 0.812 ], [ 0.94 ], [ -1.704 ], [ -0.008 ], [ -0.184 ], [ 1.952 ], ... ],
[ [ -1.656 ], [ 0.776 ], [ 0.64 ], [ 0.628 ], [ 0.164 ], [ 0.956 ], [ 1.536 ], [ -0.204 ], ... ],
...
]
[
[ [ 1.028 ], [ -1.508 ], [ -1.808 ], [ 1.536 ], [ 0.964 ], [ 0.196 ], [ -1.028 ], [ 1.716 ], ... ],
[ [ 0.384 ], [ 0.788 ], [ 0.408 ], [ -0.528 ], [ -0.236 ], [ 1.044 ], [ 0.832 ], [ -1.476 ], ... ],
[ [ -1.956 ], [ -0.692 ], [ 0.12 ], [ -0.66 ], [ 0.116 ], [ -1.932 ], [ 0.728 ], [ 0.916 ], ... ],
[ [ 1.372 ], [ -1.312 ], [ 0.488 ], [ -1.004 ], [ -1.764 ], [ 1.612 ], [ 0.856 ], [ -0.052 ], ... ],
[ [ -0.292 ], [ 1.252 ], [ -0.84 ], [ -0.536 ], [ -1.468 ], [ 1.068 ], [ 0.6 ], [ -1.336 ], ... ],
[ [ -1.26 ], [ -1.836 ], [ 0.856 ], [ -0.888 ], [ 1.5 ], [ 1.064 ], [ -1.868 ], [ 0.336 ], ... ],
[ [ -1.776 ], [ -1.88 ], [ 0.78 ], [ -0.304 ], [ -0.092 ], [ 0.176 ], [ 1.156 ], [ 0.128 ], ... ],
[ [ 0.016 ], [ 0.48 ], [ 1.58 ], [ -1.408 ], [ 0.516 ], [ -0.644 ], [ 1.004 ], [ -1.112 ], ... ],
...
]
[
[ [ -1.028 ], [ 0.168 ], [ -1.784 ], [ -0.248 ], [ -0.396 ], [ -1.892 ], [ -1.668 ], [ 1.776 ], ... ],
[ [ -1.772 ], [ 0.76 ], [ 0.984 ], [ 1.752 ], [ -1.728 ], [ -0.772 ], [ -0.352 ], [ 1.44 ], ... ],
[ [ 0.948 ], [ -1.1 ], [ -1.968 ], [ -1.836 ], [ 1.628 ], [ -0.836 ], [ 1.444 ], [ 1.948 ], ... ],
[ [ -0.072 ], [ -1.028 ], [ 0.116 ], [ -1.864 ], [ 1.8 ], [ -1.7 ], [ 0.652 ], [ 0.264 ], ... ],
[ [ -1.42 ], [ 1.636 ], [ -1.788 ], [ -1.288 ], [ 1.44 ], [ 0.988 ], [ 0.96 ], [ 1.46 ], ... ],
[ [ 1.0 ], [ -0.72 ], [ -1.68 ], [ 1.404 ], [ 1.684 ], [ -1.364 ], [ -1.948 ], [ 0.104 ], ... ],
[ [ -1.148 ], [ -0.136 ], [ 0.46 ], [ 1.272 ], [ -1.936 ], [ -0.9 ], [ 0.66 ], [ -0.964 ], ... ],
[ [ -0.3 ], [ -1.556 ], [ -1.952 ], [ -0.288 ], [ 1.748 ], [ -1.816 ], [ -1.044 ], [ -0.032 ], ... ],
...
]
[
[ [ -0.728 ], [ 1.232 ], [ 1.784 ], [ -1.236 ], [ 0.444 ], [ -0.388 ], [ -0.308 ], [ 0.42 ], ... ],
[ [ 1.048 ], [ -1.592 ], [ -1.66 ], [ -0.656 ], [ -1.16 ], [ -1.036 ], [ -0.564 ], [ 0.164 ], ... ],
[ [ 1.924 ], [ 1.912 ], [ -1.152 ], [ 0.228 ], [ -1.368 ], [ 1.684 ], [ -1.476 ], [ 0.144 ], ... ],
[ [ 1.652 ], [ 1.856 ], [ 1.64 ], [ -1.38 ], [ -1.724 ], [ -1.016 ], [ 1.096 ], [ 0.456 ], ... ],
[ [ 0.516 ], [ -0.848 ], [ 0.264 ], [ 1.876 ], [ -0.932 ], [ -0.892 ], [ -0.452 ], [ 0.336 ], ... ],
[ [ 1.632 ], [ 0.988 ], [ -1.964 ], [ -1.676 ], [ -0.304 ], [ 0.308 ], [ 1.944 ], [ 0.192 ], ... ],
[ [ 0.22 ], [ -1.956 ], [ -0.484 ], [ -1.484 ], [ 0.348 ], [ 0.808 ], [ -0.356 ], [ 0.296 ], ... ],
[ [ -0.104 ], [ -0.344 ], [ -1.2 ], [ -0.908 ], [ -0.232 ], [ 1.848 ], [ 0.9 ], [ -1.048 ], ... ],
...
]
First, we train using basic gradient descent method apply weak line search conditions.
TrainingTester.java:480 executed in 34.30 seconds (2.591 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: 4887289956052
Reset training subject: 4888314130135
Constructing line search parameters: GD
th(0)=2.6665319460607515;dx=-7.407033181102625E-7
New Minimum: 2.6665319460607515 > 2.6665303501296647
WOLFE (weak): th(2.154434690031884)=2.6665303501296647; dx=-7.407030964597175E-7 evalInputDelta=1.5959310868396415E-6
New Minimum: 2.6665303501296647 > 2.666528754363376
WOLFE (weak): th(4.308869380063768)=2.666528754363376; dx=-7.407028750969708E-7 evalInputDelta=3.191697375282132E-6
New Minimum: 2.666528754363376 > 2.666522371176858
WOLFE (weak): th(12.926608140191302)=2.666522371176858; dx=-7.407019884169422E-7 evalInputDelta=9.574883893304076E-6
New Minimum: 2.666522371176858 > 2.6664936469855203
WOLFE (weak): th(51.70643256076521)=2.6664936469855203; dx=-7.406979988431708E-7 evalInputDelta=3.829907523122955E-5
New Minimum: 2.6664936469855203 > 2.6663404537393696
WOLFE (weak): th(258.53216280382605)=2.6663404537393696; dx=-7.406767214576574E-7 evalInputDelta=1.914923213819364E-4
New Minimum: 2.6663404537393696 > 2.665383095922124
WOLFE (weak): th(1551.1929768229563)=2.665383095922124; dx=-7.405437383863488E-7 evalInputDelta=0.001148850138627644
New Minimum: 2.665383095922124 > 2.658495194160193
WOLFE (weak): th(10858.350837760694)=2.658495194160193; dx=-7.3958626054724E-7 evalInputDelta=0.008036751900558414
New Minimum: 2.658495194160193 > 2.602577555010995
WOLFE (weak): th(86866.80670208555)=2.602577555010995; dx=-7.317668553305018E-7 evalInputDelta=0.06395439104975642
New Minimum: 2.602577555010995 > 2.11888857859173
END: th(781801.26031877)=2.11888857859173; dx=-6.602751533668624E-7 evalInputDelta=0.5476433674690213
Fitness changed from 2.6665319460607515 to 2.11888857859173
Iteration 1 complete. Error: 2.11888857859173 Total: 15.4914; Orientation: 1.1154; Line Search: 11.8926
th(0)=2.11888857859173;dx=-5.885801609015235E-7
New Minimum: 2.11888857859173 > 1.2434780975500326
END: th(1684339.7559414052)=1.2434780975500326; dx=-4.508900267958028E-7 evalInputDelta=0.8754104810416976
Fitness changed from 2.11888857859173 to 1.2434780975500326
Iteration 2 complete. Error: 1.2434780975500326 Total: 2.9544; Orientation: 0.5256; Line Search: 1.8343
th(0)=1.2434780975500326;dx=-3.4541058256622693E-7
New Minimum: 1.2434780975500326 > 0.3059155076653606
END: th(3628800.0)=0.3059155076653606; dx=-1.71323648971085E-7 evalInputDelta=0.937562589884672
Fitness changed from 1.2434780975500326 to 0.3059155076653606
Iteration 3 complete. Error: 0.3059155076653606 Total: 3.7425; Orientation: 0.5682; Line Search: 2.5166
th(0)=0.3059155076653606;dx=-8.497652989870302E-8
New Minimum: 0.3059155076653606 > 0.002253881897828794
WOLF (strong): th(7818012.6031877)=0.002253881897828794; dx=7.293967497682364E-9 evalInputDelta=0.3036616257675318
END: th(3909006.30159385)=0.06391320578584184; dx=-3.884128116688185E-8 evalInputDelta=0.24200230187951874
Fitness changed from 0.3059155076653606 to 0.002253881897828794
Iteration 4 complete. Error: 0.002253881897828794 Total: 3.7699; Orientation: 0.5201; Line Search: 2.6716
th(0)=0.002253881897828794;dx=-6.260783056297943E-10
New Minimum: 0.002253881897828794 > 6.48924881529821E-5
WOLF (strong): th(8421698.779707026)=6.48924881529821E-5; dx=1.0623320654969053E-10 evalInputDelta=0.002188989409675812
END: th(4210849.389853513)=3.8847386037924313E-4; dx=-2.59922561281918E-10 evalInputDelta=0.0018654080374495509
Fitness changed from 0.002253881897828794 to 6.48924881529821E-5
Iteration 5 complete. Error: 6.48924881529821E-5 Total: 3.8911; Orientation: 0.5173; Line Search: 2.6976
Low gradient: 4.245667338797599E-6
th(0)=6.48924881529821E-5;dx=-1.8025691151732663E-11
New Minimum: 6.48924881529821E-5 > 4.386733097203415E-6
WOLF (strong): th(9072000.0)=4.386733097203415E-6; dx=4.686680177230997E-12 evalInputDelta=6.050575505577868E-5
END: th(4536000.0)=8.883779460513484E-6; dx=-6.669504911473468E-12 evalInputDelta=5.600870869246861E-5
Fitness changed from 6.48924881529821E-5 to 4.386733097203415E-6
Iteration 6 complete. Error: 4.386733097203415E-6 Total: 4.4368; Orientation: 0.5154; Line Search: 3.3744
Final threshold in iteration 6: 4.386733097203415E-6 (> 0.0) after 34.287s (< 30.000s)
Returns
4.386733097203415E-6
First, we use a conjugate gradient descent method, which converges the fastest for purely linear functions.
TrainingTester.java:452 executed in 31.64 seconds (0.848 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: 4921707102132
Reset training subject: 4922266706188
Constructing line search parameters: GD
F(0.0) = LineSearchPoint{point=PointSample{avg=2.6665319460607515}, derivative=-7.407033181102625E-7}
F(1.0E-10) = LineSearchPoint{point=PointSample{avg=2.6665319460607515}, derivative=-7.407033181102625E-7}, evalInputDelta = 0.0
F(7.000000000000001E-10) = LineSearchPoint{point=PointSample{avg=2.6665319460607515}, derivative=-7.407033181102625E-7}, evalInputDelta = 0.0
F(4.900000000000001E-9) = LineSearchPoint{point=PointSample{avg=2.6665319460607515}, derivative=-7.407033181102625E-7}, evalInputDelta = 0.0
F(3.430000000000001E-8) = LineSearchPoint{point=PointSample{avg=2.6665319460607515}, derivative=-7.407033181102625E-7}, evalInputDelta = 0.0
F(2.4010000000000004E-7) = LineSearchPoint{point=PointSample{avg=2.6665319460607515}, derivative=-7.407033181102625E-7}, evalInputDelta = 0.0
F(1.6807000000000003E-6) = LineSearchPoint{point=PointSample{avg=2.6665319460607515}, derivative=-7.407033181102625E-7}, evalInputDelta = 0.0
New Minimum: 2.6665319460607515 > 2.6665319460607475
F(1.1764900000000001E-5) = LineSearchPoint{point=PointSample{avg=2.6665319460607475}, derivative=-7.407033181102625E-7}, evalInputDelta = -3.9968028886505635E-15
New Minimum: 2.6665319460607475 > 2.6665319460607204
F(8.235430000000001E-5) = LineSearchPoint{point=PointSample{avg=2.6665319460607204}, derivative=-7.407033181102625E-7}, evalInputDelta = -3.108624468950438E-14
New Minimum: 2.6665319460607204 > 2.666531945963218
F(5.764801000000001E-4) = LineSearchPoint{point=PointSample{avg=2.666531945963218}, derivative=-7.407033180956233E-7}, evalInputDelta = -9.753353680252985E-11
New Minimum: 2.666531945963218 > 2.6665319431052166
F(0.004035360700000001) = LineSearchPoint{point=PointSample{avg=2.6665319431052166}, derivative=-7.407033177062354E-7}, evalInputDelta = -2.9555349279064558E-9
New Minimum: 2.6665319431052166 > 2.6665319249868666
F(0.028247524900000005) = LineSearchPoint{point=PointSample{avg=2.6665319249868666}, derivative=-7.407033152321416E-7}, evalInputDelta = -2.1073884859390546E-8
New Minimum: 2.6665319249868666 > 2.6665317995165543
F(0.19773267430000002) = LineSearchPoint{point=PointSample{avg=2.6665317995165543}, derivative=-7.40703297885925E-7}, evalInputDelta = -1.465441972214876E-7
New Minimum: 2.6665317995165543 > 2.6665309206750014
F(1.3841287201) = LineSearchPoint{point=PointSample{avg=2.6665309206750014}, derivative=-7.407031758709724E-7}, evalInputDelta = -1.0253857500863717E-6
New Minimum: 2.6665309206750014 > 2.666524769313392
F(9.688901040700001) = LineSearchPoint{point=PointSample{avg=2.666524769313392}, derivative=-7.407023213327709E-7}, evalInputDelta = -7.176747359505242E-6
New Minimum: 2.666524769313392 > 2.6664817100082465
F(67.8223072849) = LineSearchPoint{point=PointSample{avg=2.6664817100082465}, derivative=-7.40696340721863E-7}, evalInputDelta = -5.023605250498164E-5
New Minimum: 2.6664817100082465 > 2.6661803040747722
F(474.7561509943) = LineSearchPoint{point=PointSample{avg=2.6661803040747722}, derivative=-7.406544771648014E-7}, evalInputDelta = -3.5164198597925633E-4
New Minimum: 2.6661803040747722 > 2.6640709398300753
F(3323.2930569601003) = LineSearchPoint{point=PointSample{avg=2.6640709398300753}, derivative=-7.403614327340155E-7}, evalInputDelta = -0.0024610062306762437
New Minimum: 2.6640709398300753 > 2.649328763041108
F(23263.0513987207) = LineSearchPoint{point=PointSample{avg=2.649328763041108}, derivative=-7.383101210087045E-7}, evalInputDelta = -0.017203183019643298
New Minimum: 2.649328763041108 > 2.547278800546901
F(162841.3597910449) = LineSearchPoint{point=PointSample{avg=2.547278800546901}, derivative=-7.239509385375475E-7}, evalInputDelta = -0.11925314551385036
New Minimum: 2.547278800546901 > 1.8890475155452777
F(1139889.5185373144) = LineSearchPoint{point=PointSample{avg=1.8890475155452777}, derivative=-6.234366587673029E-7}, evalInputDelta = -0.7774844305154738
New Minimum: 1.8890475155452777 > 0.031232688438368662
F(7979226.6297612) = LineSearchPoint{point=PointSample{avg=0.031232688438368662}, derivative=8.016329848068257E-8}, evalInputDelta = -2.635299257622383
0.031232688438368662 <= 2.6665319460607515
New Minimum: 0.031232688438368662 > 2.7903006326416304E-15
F(7200000.0014168965) = LineSearchPoint{point=PointSample{avg=2.7903006326416304E-15}, derivative=6.527664518374189E-17}, evalInputDelta = -2.666531946060749
Right bracket at 7200000.0014168965
Converged to right
Fitness changed from 2.6665319460607515 to 2.7903006326416304E-15
Iteration 1 complete. Error: 2.7903006326416304E-15 Total: 25.5637; Orientation: 0.5234; Line Search: 23.3665
Zero gradient: 2.784032193295081E-11
F(0.0) = LineSearchPoint{point=PointSample{avg=2.7903006326416304E-15}, derivative=-7.750835253303391E-22}
New Minimum: 2.7903006326416304E-15 > 1.0794451752202722E-18
F(7200000.0014168965) = LineSearchPoint{point=PointSample{avg=1.0794451752202722E-18}, derivative=6.102114827546661E-25}, evalInputDelta = -2.78922118746641E-15
1.0794451752202722E-18 <= 2.7903006326416304E-15
Converged to right
Fitness changed from 2.7903006326416304E-15 to 1.0794451752202722E-18
Iteration 2 complete. Error: 1.0794451752202722E-18 Total: 3.0650; Orientation: 0.5906; Line Search: 1.8818
Zero gradient: 5.475818556826023E-13
F(0.0) = LineSearchPoint{point=PointSample{avg=1.0794451752202722E-18}, derivative=-2.998458886728023E-25}
New Minimum: 1.0794451752202722E-18 > 3.4522794915609565E-48
F(7200000.0014168965) = LineSearchPoint{point=PointSample{avg=3.4522794915609565E-48}, derivative=3.315603624005143E-48}, evalInputDelta = -1.0794451752202722E-18
3.4522794915609565E-48 <= 1.0794451752202722E-18
Converged to right
Fitness changed from 1.0794451752202722E-18 to 3.4522794915609565E-48
Iteration 3 complete. Error: 3.4522794915609565E-48 Total: 3.0057; Orientation: 0.5400; Line Search: 1.8466
Final threshold in iteration 3: 3.4522794915609565E-48 (> 0.0) after 31.635s (< 30.000s)
Returns
3.4522794915609565E-48
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 200.18 seconds (3.413 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: 4953476660160
Reset training subject: 4954035077260
Adding measurement 5b1a9d7a to history. Total: 0
LBFGS Accumulation History: 1 points
Constructing line search parameters: GD
Non-optimal measurement 2.6665319460607515 < 2.6665319460607515. Total: 1
th(0)=2.6665319460607515;dx=-7.407033181102625E-7
Adding measurement 4fdea972 to history. Total: 1
New Minimum: 2.6665319460607515 > 2.6665303501296647
WOLFE (weak): th(2.154434690031884)=2.6665303501296647; dx=-7.407030964597175E-7 evalInputDelta=1.5959310868396415E-6
Adding measurement 1d3b0651 to history. Total: 2
New Minimum: 2.6665303501296647 > 2.666528754363376
WOLFE (weak): th(4.308869380063768)=2.666528754363376; dx=-7.407028750969708E-7 evalInputDelta=3.191697375282132E-6
Adding measurement f42ff46 to history. Total: 3
New Minimum: 2.666528754363376 > 2.666522371176858
WOLFE (weak): th(12.926608140191302)=2.666522371176858; dx=-7.407019884169422E-7 evalInputDelta=9.574883893304076E-6
Adding measurement 392b6429 to history. Total: 4
New Minimum: 2.666522371176858 > 2.6664936469855203
WOLFE (weak): th(51.70643256076521)=2.6664936469855203; dx=-7.406979988431708E-7 evalInputDelta=3.829907523122955E-5
Adding measurement 7e29c56a to history. Total: 5
New Minimum: 2.6664936469855203 > 2.6663404537393696
WOLFE (weak): th(258.53216280382605)=2.6663404537393696; dx=-7.406767214576574E-7 evalInputDelta=1.914923213819364E-4
Adding measurement 67bcd753 to history. Total: 6
New Minimum: 2.6663404537393696 > 2.665383095922124
WOLFE (weak): th(1551.1929768229563)=2.665383095922124; dx=-7.405437383863488E-7 evalInputDelta=0.001148850138627644
Adding measurement 11627841 to history. Total: 7
New Minimum: 2.665383095922124 > 2.658495194160193
WOLFE (weak): th(10858.350837760694)=2.658495194160193; dx=-7.3958626054724E-7 evalInputDelta=0.008036751900558414
Adding measurement 7215c1c6 to history. Total: 8
New Minimum: 2.658495194160193 > 2.602577555010995
WOLFE (weak): th(86866.80670208555)=2.602577555010995; dx=-7.317668553305018E-7 evalInputDelta=0.06395439104975642
Adding measurement 4f78309e to history. Total: 9
New Minimum: 2.602577555010995 > 2.11888857859173
END: th(781801.26031877)=2.11888857859173; dx=-6.602751533668624E-7 evalInputDelta=0.5476433674690213
Fitness changed from 2.6665319460607515 to 2.11888857859173
Iteration 1 complete. Error: 2.11888857859173 Total: 15.7755; Orientation: 0.6173; Line Search: 13.4808
Non-optimal measurement 2.11888857859173 < 2.11888857859173. Total: 10
Rejected: LBFGS Orientation magnitude: 5.689e+03, gradient 7.672e-04, dot -0.991; [09c74526-9691-4672-86fb-8e5b12bb6783 = 1.000/1.000e+00, 3e3ab626-7fe5-431d-8eef-bbf31d2d3fd5 = 1.000/1.000e+00, adb845e8-9ce9-4b74-af98-cc8fc9c487b4 = 1.000/1.000e+00, 13e3d85c-032d-4b4e-81b1-074a957deb39 = 1.000/1.000e+00, dd76a339-fd1f-4788-a3e8-cdea342a8e05 = 1.000/1.000e+00, 9d2cfc1e-22ea-4cdc-aeb4-95ea28e20562 = 1.000/1.000e+00, c1f97b90-c2bc-43ec-b97f-6e75e13030d1 = 1.000/1.000e+00, 6c7d0ece-a5c3-49a5-b66b-c2bd1f701b84 = 1.000/1.000e+00, 54696d94-062b-43d4-adc8-d8d4a0426b51 = 1.000/1.000e+00, b6fcb78e-dff5-46bd-b83c-88f95bf1c682 = 1.000/1.000e+00]
Orientation rejected. Popping history element from 2.11888857859173, 2.602577555010995, 2.658495194160193, 2.665383095922124, 2.6663404537393696, 2.6664936469855203, 2.666522371176858, 2.666528754363376, 2.6665303501296647, 2.6665319460607515
Rejected: LBFGS Orientation magnitude: 5.696e+03, gradient 7.672e-04, dot -0.991; [dd76a339-fd1f-4788-a3e8-cdea342a8e05 = 1.000/1.000e+00, 3e3ab626-7fe5-431d-8eef-bbf31d2d3fd5 = 1.000/1.000e+00, 09c74526-9691-4672-86fb-8e5b12bb6783 = 1.000/1.000e+00, b6fcb78e-dff5-46bd-b83c-88f95bf1c682 = 1.000/1.000e+00, 9d2cfc1e-22ea-4cdc-aeb4-95ea28e20562 = 1.000/1.000e+00, 6c7d0ece-a5c3-49a5-b66b-c2bd1f701b84 = 1.000/1.000e+00, 13e3d85c-032d-4b4e-81b1-074a957deb39 = 1.000/1.000e+00, c1f97b90-c2bc-43ec-b97f-6e75e13030d1 = 1.000/1.000e+00, adb845e8-9ce9-4b74-af98-cc8fc9c487b4 = 1.000/1.000e+00, 54696d94-062b-43d4-adc8-d8d4a0426b51 = 1.000/1.000e+00]
Orientation rejected. Popping history element from 2.11888857859173, 2.602577555010995, 2.658495194160193, 2.665383095922124, 2.6663404537393696, 2.6664936469855203, 2.666522371176858, 2.666528754363376, 2.6665303501296647
Rejected: LBFGS Orientation magnitude: 5.534e+03, gradient 7.672e-04, dot -0.999; [3e3ab626-7fe5-431d-8eef-bbf31d2d3fd5 = 1.000/1.000e+00, dd76a339-fd1f-4788-a3e8-cdea342a8e05 = 1.000/1.000e+00, 9d2cfc1e-22ea-4cdc-aeb4-95ea28e20562 = 1.000/1.000e+00, c1f97b90-c2bc-43ec-b97f-6e75e13030d1 = 1.000/1.000e+00, 09c74526-9691-4672-86fb-8e5b12bb6783 = 1.000/1.000e+00, adb845e8-9ce9-4b74-af98-cc8fc9c487b4 = 1.000/1.000e+00, 13e3d85c-032d-4b4e-81b1-074a957deb39 = 1.000/1.000e+00, b6fcb78e-dff5-46bd-b83c-88f95bf1c682 = 1.000/1.000e+00, 54696d94-062b-43d4-adc8-d8d4a0426b51 = 1.000/1.000e+00, 6c7d0ece-a5c3-49a5-b66b-c2bd1f701b84 = 1.000/1.000e+00]
Orientation rejected. Popping history element from 2.11888857859173, 2.602577555010995, 2.658495194160193, 2.665383095922124, 2.6663404537393696, 2.6664936469855203, 2.666522371176858, 2.666528754363376
Rejected: LBFGS Orientation magnitude: 5.524e+03, gradient 7.672e-04, dot -1.000; [c1f97b90-c2bc-43ec-b97f-6e75e13030d1 = 1.000/1.000e+00, 54696d94-062b-43d4-adc8-d8d4a0426b51 = 1.000/1.000e+00, dd76a339-fd1f-4788-a3e8-cdea342a8e05 = 1.000/1.000e+00, adb845e8-9ce9-4b74-af98-cc8fc9c487b4 = 1.000/1.000e+00, 13e3d85c-032d-4b4e-81b1-074a957deb39 = 1.000/1.000e+00, 3e3ab626-7fe5-431d-8eef-bbf31d2d3fd5 = 1.000/1.000e+00, b6fcb78e-dff5-46bd-b83c-88f95bf1c682 = 1.000/1.000e+00, 6c7d0ece-a5c3-49a5-b66b-c2bd1f701b84 = 1.000/1.000e+00, 09c74526-9691-4672-86fb-8e5b12bb6783 = 1.000/1.000e+00, 9d2cfc1e-22ea-4cdc-aeb4-95ea28e20562 = 1.000/1.000e+00]
Orientation rejected. Popping history element from 2.11888857859173, 2.602577555010995, 2.658495194160193, 2.665383095922124, 2.6663404537393696, 2.6664936469855203, 2.666522371176858
Rejected: LBFGS Orientation magnitude: 5.524e+03, gradient 7.672e-04, dot -1.000; [09c74526-9691-4672-86fb-8e5b12bb6783 = 1.000/1.000e+00, 3e3ab626-7fe5-431d-8eef-bbf31d2d3fd5 = 1.000/1.000e+00, dd76a339-fd1f-4788-a3e8-cdea342a8e05 = 1.000/1.000e+00, 6c7d0ece-a5c3-49a5-b66b-c2bd1f701b84 = 1.000/1.000e+00, 54696d94-062b-43d4-adc8-d8d4a0426b51 = 1.000/1.000e+00, 9d2cfc1e-22ea-4cdc-aeb4-95ea28e20562 = 1.000/1.000e+00, c1f97b90-c2bc-43ec-b97f-6e75e13030d1 = 1.000/1.000e+00, adb845e8-9ce9-4b74-af98-cc8fc9c487b4 = 1.000/1.000e+00, b6fcb78e-dff5-46bd-b83c-88f95bf1c682 = 1.000/1.000e+00, 13e3d85c-032d-4b4e-81b1-074a957deb39 = 1.000/1.000e+00]
Orientation rejected. Popping history element from 2.11888857859173, 2.602577555010995, 2.658495194160193, 2.665383095922124, 2.6663404537393696, 2.6664936469855203
Rejected: LBFGS Orientation magnitude: 5.524e+03, gradient 7.672e-04, dot -1.000; [13e3d85c-032d-4b4e-81b1-074a957deb39 = 1.000/1.000e+00, 54696d94-062b-43d4-adc8-d8d4a0426b51 = 1.000/1.000e+00, 9d2cfc1e-22ea-4cdc-aeb4-95ea28e20562 = 1.000/1.000e+00, dd76a339-fd1f-4788-a3e8-cdea342a8e05 = 1.000/1.000e+00, adb845e8-9ce9-4b74-af98-cc8fc9c487b4 = 1.000/1.000e+00, b6fcb78e-dff5-46bd-b83c-88f95bf1c682 = 1.000/1.000e+00, 6c7d0ece-a5c3-49a5-b66b-c2bd1f701b84 = 1.000/1.000e+00, 09c74526-9691-4672-86fb-8e5b12bb6783 = 1.000/1.000e+00, c1f97b90-c2bc-43ec-b97f-6e75e13030d1 = 1.000/1.000e+00, 3e3ab626-7fe5-431d-8eef-bbf31d2d3fd5 = 1.000/1.000e+00]
Orientation rejected. Popping history element from 2.11888857859173, 2.602577555010995, 2.658495194160193, 2.665383095922124, 2.6663404537393696
Rejected: LBFGS Orientation magnitude: 5.524e+03, gradient 7.672e-04, dot -1.000; [54696d94-062b-43d4-adc8-d8d4a0426b51 = 1.000/1.000e+00, 3e3ab626-7fe5-431d-8eef-bbf31d2d3fd5 = 1.000/1.000e+00, c1f97b90-c2bc-43ec-b97f-6e75e13030d1 = 1.000/1.000e+00, 09c74526-9691-4672-86fb-8e5b12bb6783 = 1.000/1.000e+00, adb845e8-9ce9-4b74-af98-cc8fc9c487b4 = 1.000/1.000e+00, b6fcb78e-dff5-46bd-b83c-88f95bf1c682 = 1.000/1.000e+00, 6c7d0ece-a5c3-49a5-b66b-c2bd1f701b84 = 1.000/1.000e+00, 13e3d85c-032d-4b4e-81b1-074a957deb39 = 1.000/1.000e+00, 9d2cfc1e-22ea-4cdc-aeb4-95ea28e20562 = 1.000/1.000e+00, dd76a339-fd1f-4788-a3e8-cdea342a8e05 = 1.000/1.000e+00]
Orientation rejected. Popping history element from 2.11888857859173, 2.602577555010995, 2.658495194160193, 2.665383095922124
LBFGS Accumulation History: 3 points
Removed measurement 4f78309e to history. Total: 9
Removed measurement 7215c1c6 to history. Total: 8
Removed measurement 11627841 to history. Total: 7
Removed measurement 67bcd753 to history. Total: 6
Removed measurement 7e29c56a to history. Total: 5
Removed measurement 392b6429 to history. Total: 4
Removed measurement f42ff46 to history. Total: 3
Adding measurement 1f39cf16 to history. Total: 3
th(0)=2.11888857859173;dx=-5.885801609015235E-7
Adding measurement 6f27d6d1 to history. Total: 4
New Minimum: 2.11888857859173 > 1.2434780975500326
END: th(1684339.7559414052)=1.2434780975500326; dx=-4.508900267958028E-7 evalInputDelta=0.8754104810416976
Fitness changed from 2.11888857859173 to 1.2434780975500326
Iteration 2 complete. Error: 1.2434780975500326 Total: 184.3984; Orientation: 181.1498; Line Search: 2.6449
Final threshold in iteration 2: 1.2434780975500326 (> 0.0) after 200.174s (< 30.000s)
Returns
1.2434780975500326
This training apply resulted in the following configuration:
TrainingTester.java:610 executed in 0.00 seconds (0.000 gc):
RefList<double[]> state = network.state();
assert state != null;
String description = state.stream().map(RefArrays::toString).reduce((a, b) -> a + "\n" + b)
.orElse("");
state.freeRef();
return description;
Returns
TrainingTester.java:622 executed in 0.00 seconds (0.000 gc):
return RefArrays.stream(RefUtil.addRef(data)).flatMap(x -> {
return RefArrays.stream(x);
}).limit(1).map(x -> {
String temp_18_0015 = x.prettyPrint();
x.freeRef();
return temp_18_0015;
}).reduce((a, b) -> a + "\n" + b).orElse("");
Returns
[
[ [ -1.4467957401238019 ], [ 0.004150489159482018 ], [ 0.7278054206665104 ], [ -1.4566205088642983 ], [ 0.8831261033865616 ], [ -1.488936193330393 ], [ -0.8728940198055933 ], [ 0.5695218680682244 ], ... ],
[ [ -0.02555667372896927 ], [ -1.4260889803190107 ], [ 0.7444386829983118 ], [ 0.5338595844045493 ], [ -0.9346727409052733 ], [ -1.2309168340788137 ], [ 0.3483156653173686 ], [ 0.7356878218419194 ], ... ],
[ [ 0.31376788428138463 ], [ 1.0728773659966235 ], [ -0.9088077315971157 ], [ -1.146379116346186 ], [ 1.6102414113236823 ], [ -0.8339017531421673 ], [ 0.38270019323508586 ], [ -0.2584120082843072 ], ... ],
[ [ -0.8358824219273653 ], [ -0.01074082588981351 ], [ -1.0544653672866793 ], [ 0.9823349985826056 ], [ 0.3378889597739819 ], [ 0.940034834126467 ], [ 0.3546011687614449 ], [ -0.23463247743117221 ], ... ],
[ [ -1.296 ], [ -1.8919852890656879 ], [ -1.2680642286446258 ], [ -1.0735272604910293 ], [ -1.5161284572892517 ], [ -0.1371620637881143 ], [ -1.783992657864921 ], [ -0.5257532150590827 ], ... ],
[ [ -0.38794665057199096 ], [ -1.2967416035056218 ], [ -0.3937640382021449 ], [ 1.1163470020238733 ], [ 0.8659036761817872 ], [ -0.43326847030170784 ], [ 0.5692978645765683 ], [ 0.46807351433217403 ], ... ],
[ [ -0.13606616057229698 ], [ -1.3560274803665902 ], [ 0.5144982776872148 ], [ -0.9791996125883985 ], [ 0.1909775653034768 ], [ 1.0069408148034282 ], [ -0.41977058493681285 ], [ -0.1128003874116017 ], ... ],
[ [ 0.4149922434138301 ], [ -0.8573381280506691 ], [ -0.6778228272384623 ], [ -0.10252959721388943 ], [ -0.4015547511182941 ], [ -0.3662634288323683 ], [ 1.349650282787576 ], [ 1.264150470010756 ], ... ],
...
]
To produce the following output:
TrainingTester.java:633 executed in 0.15 seconds (0.000 gc):
Result[] array = ConstantResult.batchResultArray(pop(RefUtil.addRef(data)));
@Nullable
Result eval = layer.eval(array);
assert eval != null;
TensorList tensorList = Result.getData(eval);
String temp_18_0016 = tensorList.stream().limit(1).map(x -> {
String temp_18_0017 = x.prettyPrint();
x.freeRef();
return temp_18_0017;
}).reduce((a, b) -> a + "\n" + b).orElse("");
tensorList.freeRef();
return temp_18_0016;
Returns
[
[ [ -1.4467957019805908, 1.4618943929672241 ], [ 0.004150489345192909, 0.3413914740085602 ], [ 0.7278054356575012, -0.8245102167129517 ], [ -1.456620454788208, -0.68016517162323 ], [ 0.8831260800361633, -1.0097365379333496 ], [ -1.488936185836792, -0.2808866500854492 ], [ -0.8728940486907959, -0.6287214756011963 ], [ 0.5695218443870544, -0.5884405970573425 ], ... ],
[ [ -0.025556674227118492, 0.45399534702301025 ], [ -1.4260889291763306, -0.011709882877767086 ], [ 0.7444387078285217, 1.3943991661071777 ], [ 0.5338596105575562, -0.2955814301967621 ], [ -0.9346727132797241, -0.7568498849868774 ], [ -1.2309168577194214, -0.390817791223526 ], [ 0.34831565618515015, -0.729729175567627 ], [ 0.735687792301178, 0.23402667045593262 ], ... ],
[ [ 0.3137678802013397, 0.032938115298748016 ], [ 1.0728774070739746, 0.6253198981285095 ], [ -0.9088077545166016, -0.9129949808120728 ], [ -1.1463791131973267, 0.8736576437950134 ], [ 1.610241413116455, 0.49632495641708374 ], [ -0.8339017629623413, 0.48252415657043457 ], [ 0.38270020484924316, -0.26150715351104736 ], [ -0.2584120035171509, 0.9427682757377625 ], ... ],
[ [ -0.8358824253082275, -1.671517252922058 ], [ -0.010740825906395912, -0.9010205268859863 ], [ -1.054465413093567, -0.37479767203330994 ], [ 0.9823349714279175, 0.4859183728694916 ], [ 0.3378889560699463, 0.31964021921157837 ], [ 0.940034806728363, 0.41210290789604187 ], [ 0.3546011745929718, -0.5855767726898193 ], [ -0.2346324771642685, 0.7839505076408386 ], ... ],
[ [ -1.2960000038146973, -1.4779016971588135 ], [ -1.891985297203064, -0.3244386911392212 ], [ -1.268064260482788, 0.22982476651668549 ], [ -1.073527216911316, 0.5213527679443359 ], [ -1.516128420829773, -1.4823698997497559 ], [ -0.13716205954551697, -0.22270213067531586 ], [ -1.7839926481246948, -0.05910792574286461 ], [ -0.5257532000541687, 0.6840843558311462 ], ... ],
[ [ -0.3879466652870178, 0.21413849294185638 ], [ -1.2967416048049927, 1.379098653793335 ], [ -0.39376404881477356, -0.6513427495956421 ], [ 1.1163469552993774, -0.12218261510133743 ], [ 0.8659036755561829, 0.16742554306983948 ], [ -0.4332684576511383, 0.9376081228256226 ], [ 0.5692978501319885, -0.6040421724319458 ], [ 0.4680735170841217, -1.5313005447387695 ], ... ],
[ [ -0.13606615364551544, -1.1484882831573486 ], [ -1.3560274839401245, 0.30837640166282654 ], [ 0.5144982933998108, -0.9122073650360107 ], [ -0.9791995882987976, 1.0193500518798828 ], [ 0.1909775584936142, -0.7096228003501892 ], [ 1.0069408416748047, 0.40934738516807556 ], [ -0.41977059841156006, -1.388763666152954 ], [ -0.11280038952827454, 0.8112197518348694 ], ... ],
[ [ 0.41499224305152893, -0.33844876289367676 ], [ -0.8573381304740906, -1.495344638824463 ], [ -0.6778228282928467, 0.7855895757675171 ], [ -0.1025296002626419, 1.147091269493103 ], [ -0.40155476331710815, -1.175337314605713 ], [ -0.36626341938972473, 0.48687276244163513 ], [ 1.349650263786316, -1.1322166919708252 ], [ 1.2641505002975464, 1.6064139604568481 ], ... ],
...
]
TrainingTester.java:432 executed in 0.14 seconds (0.000 gc):
return TestUtil.compare(title + " vs Iteration", runs);
Plotting range=[1.0, -47.46189405167957], [6.0, 0.3261081200071431]; valueStats=DoubleSummaryStatistics{count=11, sum=7.032972, min=0.000000, average=0.639361, max=2.118889}
Plotting 6 points for GD
Plotting 3 points for CjGD
Plotting 2 points for LBFGS
Returns
TrainingTester.java:435 executed in 0.02 seconds (0.000 gc):
return TestUtil.compareTime(title + " vs Time", runs);
Plotting range=[0.0, -47.46189405167957], [184.398, 0.3261081200071431]; valueStats=DoubleSummaryStatistics{count=11, sum=7.032972, min=0.000000, average=0.639361, max=2.118889}
Plotting 6 points for GD
Plotting 3 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": "NonConverged", "value": 1.2434780975500326 }, "CjGD": { "type": "Converged", "value": 3.4522794915609565E-48 }, "GD": { "type": "NonConverged", "value": 4.386733097203415E-6 } }, "model":null, "complete":null}
LayerTests.java:425 executed in 0.00 seconds (0.000 gc):
throwException(exceptions.addRef());
details | result |
---|---|
{"input":{ "LBFGS": { "type": "NonConverged", "value": 1.2434780975500326 }, "CjGD": { "type": "Converged", "value": 3.4522794915609565E-48 }, "GD": { "type": "NonConverged", "value": 4.386733097203415E-6 } }, "model":null, "complete":null} | OK |
{
"result": "OK",
"performance": {
"execution_time": "269.185",
"gc_time": "7.190"
},
"created_on": 1586739517201,
"file_name": "trainingTest",
"report": {
"simpleName": "Float",
"canonicalName": "com.simiacryptus.mindseye.layers.cudnn.ImgConcatLayerTest.Float",
"link": "https://github.com/SimiaCryptus/mindseye-cudnn/tree/59d5b3318556370acb2d83ee6ec123ce0fc6974f/src/test/java/com/simiacryptus/mindseye/layers/cudnn/ImgConcatLayerTest.java",
"javaDoc": ""
},
"training_analysis": {
"input": {
"LBFGS": {
"type": "NonConverged",
"value": 1.2434780975500326
},
"CjGD": {
"type": "Converged",
"value": 3.4522794915609565E-48
},
"GD": {
"type": "NonConverged",
"value": 4.386733097203415E-6
}
}
},
"archive": "s3://code.simiacrypt.us/tests/com/simiacryptus/mindseye/layers/cudnn/ImgConcatLayer/Float/trainingTest/202004135837",
"id": "516dc22c-e593-4d6c-a45b-6e481f46e924",
"report_type": "Components",
"display_name": "Comparative Training",
"target": {
"simpleName": "ImgConcatLayer",
"canonicalName": "com.simiacryptus.mindseye.layers.cudnn.ImgConcatLayer",
"link": "https://github.com/SimiaCryptus/mindseye-cudnn/tree/59d5b3318556370acb2d83ee6ec123ce0fc6974f/src/main/java/com/simiacryptus/mindseye/layers/cudnn/ImgConcatLayer.java",
"javaDoc": ""
}
}