Code from BasicOptimizer.scala:88 executed in 278.08 seconds (5.245 gc):
val lineSearchInstance: LineSearchStrategy = lineSearchFactory
val trainer = new IterativeTrainer(trainable)
trainer.setOrientation(orientation())
trainer.setMonitor(new TrainingMonitor() {
override def clear(): Unit = trainingMonitor.clear()
override def log(msg: String): Unit = {
trainingMonitor.log(msg)
BasicOptimizer.this.log(msg)
}
override def onStepFail(currentPoint: Step): Boolean = {
BasicOptimizer.this.onStepFail(trainable.addRef().asInstanceOf[Trainable], currentPoint)
}
override def onStepComplete(currentPoint: Step): Unit = {
if (0 < logEvery && (0 == currentPoint.iteration % logEvery || currentPoint.iteration < logEvery)) {
val image = currentImage()
timelineAnimation += image
val caption = "Iteration " + currentPoint.iteration
out.p(caption + "\n" + out.jpg(image, caption))
}
BasicOptimizer.this.onStepComplete(trainable.addRef().asInstanceOf[Trainable], currentPoint)
trainingMonitor.onStepComplete(currentPoint)
super.onStepComplete(currentPoint)
}
})
trainer.setTimeout(trainingMinutes, TimeUnit.MINUTES)
trainer.setMaxIterations(trainingIterations)
trainer.setLineSearchFactory((_: CharSequence) => lineSearchInstance)
trainer.setTerminateThreshold(java.lang.Double.NEGATIVE_INFINITY)
val result = trainer.run.asInstanceOf[lang.Double]
trainer.freeRef()
result
Reset training subject: 2169900202912
Reset training subject: 2173515648617
Adding measurement 5765adec to history. Total: 0
LBFGS Accumulation History: 1 points
Constructing line search parameters: GD+Trust
th(0)=23.13871928676963;dx=-1.5536625059813873E-5
New Minimum: 23.13871928676963 > 23.13865140452981
WOLFE (weak): th(2.154434690031884)=23.13865140452981; dx=-1.520811854989834E-5 evalInputDelta=6.7882239818573E-5
New Minimum: 23.13865140452981 > 23.138582352548838
WOLFE (weak): th(4.308869380063768)=23.138582352548838; dx=-1.4431186267808326E-5 evalInputDelta=1.3693422079086304E-4
Armijo: th(12.926608140191302)=23.151404973119497; dx=-1.5368024665574507E-5 evalInputDelta=-0.012685686349868774
New Minimum: 23.138582352548838 > 23.138467371463776
WOLF (strong): th(8.617738760127535)=23.138467371463776; dx=1.0967495714807414E-4 evalInputDelta=2.5191530585289E-4
WOLFE (weak): th(6.463304070095651)=23.13851356226951; dx=-1.5159354160235933E-5 evalInputDelta=2.0572450011968613E-4
WOLFE (weak): th(7.540521415111593)=23.1384795634076; dx=-1.5104663301669124E-5 evalInputDelta=2.3972336202859879E-4
New Minimum: 23.138467371463776 > 23.138461731374264
WOLFE (weak): th(8.079130087619564)=23.138461731374264; dx=-1.5052939440543968E-5 evalInputDelta=2.5755539536476135E-4
New Minimum: 23.138461731374264 > 23.138378901407123
END: th(8.348434423873549)=23.138378901407123; dx=-1.0977479325332343E-5 evalInputDelta=3.403853625059128E-4
Fitness changed from 23.13871928676963 to 23.138378901407123
Iteration 1 complete. Error: 23.138378901407123 Total: 39.6600; Orientation: 0.0336; Line Search: 29.5407
<a id="p-3"></a>Iteration 1
<a id="p-2"></a>
Adding measurement 5087f707 to history. Total: 1
LBFGS Accumulation History: 2 points
th(0)=23.138378901407123;dx=-3.518703040825683E-5
Armijo: th(17.98615673024952)=23.15095162205398; dx=-1.4406483156614046E-5 evalInputDelta=-0.012572720646858215
Armijo: th(8.99307836512476)=23.15125075634569; dx=-1.5539671455146727E-5 evalInp
...skipping 7311 bytes...
: th(6.116420036434031E-5)=23.138370836153626; dx=-1.5041424255071665E-5 evalInputDelta=7.450580596923828E-9
Fitness changed from 23.138370843604207 to 23.138370836153626
Iteration 9 complete. Error: 23.138370836153626 Total: 16.4007; Orientation: 0.0032; Line Search: 13.2750
Adding measurement 1df06221 to history. Total: 2
LBFGS Accumulation History: 3 points
th(0)=23.138370836153626;dx=-1.5594559352423243E-5
Armijo: th(1.3177427505299555E-4)=23.13837178517133; dx=-1.0028602237026998E-5 evalInputDelta=-9.490177035331726E-7
WOLFE (weak): th(6.588713752649777E-5)=23.138370836153626; dx=-1.5042356923756903E-5 evalInputDelta=0.0
Armijo: th(9.883070628974666E-5)=23.13837178517133; dx=-8.126901271497656E-6 evalInputDelta=-9.490177035331726E-7
WOLFE (weak): th(8.235892190812222E-5)=23.138370836153626; dx=-1.5059712225509507E-5 evalInputDelta=0.0
Armijo: th(9.059481409893444E-5)=23.13837178517133; dx=-1.6013070597721778E-5 evalInputDelta=-9.490177035331726E-7
Armijo: th(8.647686800352834E-5)=23.13837178517133; dx=-1.0702334290829139E-5 evalInputDelta=-9.490177035331726E-7
Armijo: th(8.441789495582528E-5)=23.13837178517133; dx=-2.8954976875222647E-6 evalInputDelta=-9.490177035331726E-7
WOLFE (weak): th(8.338840843197376E-5)=23.138370836153626; dx=-1.708566930593131E-5 evalInputDelta=0.0
WOLFE (weak): th(8.390315169389952E-5)=23.138370836153626; dx=-1.644166768116639E-5 evalInputDelta=0.0
WOLFE (weak): th(8.41605233248624E-5)=23.138370836153626; dx=-1.6655743779358174E-5 evalInputDelta=0.0
WOLFE (weak): th(8.428920914034384E-5)=23.138370836153626; dx=-1.4961435856203249E-5 evalInputDelta=0.0
END: th(8.435355204808456E-5)=23.138370836153626; dx=-1.1994918178594378E-5 evalInputDelta=0.0
Fitness changed from 23.138370836153626 to 23.138370836153626
Static Iteration Total: 44.9978; Orientation: 0.0023; Line Search: 41.9316
Iteration 10 failed. Error: 23.138370836153626
Previous Error: 0.0 -> 23.138370836153626
Retrying iteration 10
Final threshold in iteration 10: 23.138370836153626 (> -Infinity) after 278.064s (< 1800.000s)
23.138370836153626