Code from BasicOptimizer.scala:75 executed in 2484.68 seconds (24.464 gc):
val lineSearchInstance: LineSearchStrategy = lineSearchFactory
IterativeTrainer.wrap(trainable)
.setOrientation(orientation())
.setMonitor(new TrainingMonitor() {
override def clear(): Unit = trainingMonitor.clear()
override def log(msg: String): Unit = trainingMonitor.log(msg)
override def onStepFail(currentPoint: Step): Boolean = {
BasicOptimizer.this.onStepFail(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, currentPoint)
trainingMonitor.onStepComplete(currentPoint)
super.onStepComplete(currentPoint)
}
})
.setTimeout(trainingMinutes, TimeUnit.MINUTES)
.setMaxIterations(trainingIterations)
.setLineSearchFactory((_: CharSequence) => lineSearchInstance)
.setTerminateThreshold(java.lang.Double.NEGATIVE_INFINITY)
.runAndFree
.asInstanceOf[lang.Double]
Reset training subject: 1080801645733899
Reset training subject: 1080821140999200
Adding measurement 1fba15d7 to history. Total: 0
LBFGS Accumulation History: 1 points
Constructing line search parameters: GD+Trust
th(0)=1.8756767897340563;dx=-3.9332071706361496E-7
New Minimum: 1.8756767897340563 > 1.8756761936876085
WOLFE (weak): th(2.154434690031884)=1.8756761936876085; dx=-3.907858024930739E-7 evalInputDelta=5.960464477539062E-7
New Minimum: 1.8756761936876085 > 1.8756750090554015
WOLFE (weak): th(4.308869380063768)=1.8756750090554015; dx=-3.9078556587729047E-7 evalInputDelta=1.7806786547502895E-6
New Minimum: 1.8756750090554015 > 1.8756714030047164
WOLFE (weak): th(12.926608140191302)=1.8756714030047164; dx=-3.9078783663155227E-7 evalInputDelta=5.386729339917906E-6
New Minimum: 1.8756714030047164 > 1.875655578082715
WOLFE (weak): th(51.70643256076521)=1.875655578082715; dx=-3.9078293203737343E-7 evalInputDelta=2.1211651341168647E-5
New Minimum: 1.875655578082715 > 1.875570306824581
WOLFE (weak): th(258.53216280382605)=1.875570306824581; dx=-3.907605683504312E-7 evalInputDelta=1.064829094752362E-4
New Minimum: 1.875570306824581 > 1.8750373707668873
WOLFE (weak): th(1551.1929768229563)=1.8750373707668873; dx=-3.906566527534067E-7 evalInputDelta=6.394189671690143E-4
New Minimum: 1.8750373707668873 > 1.87120681733738
WOLFE (weak): th(10858.350837760694)=1.87120681733738; dx=-3.8993397128044885E-7 evalInputDelta=0.004469972396676347
New Minimum: 1.87120681733738 > 1.840191935428055
WOLFE (weak): th(86866.80670208555)=1.840191935428055; dx=-3.8376470014851404E-7 evalInputDelta=0.035484854306001345
New Minimum: 1.840191935428055 > 1.5808163191276852
END: th(781801.26031877)=1.5808163191276852; dx=-3.240635165329574E-7 evalInputDelta=0.2948604706063711
Fitness changed from 1.8756767897340563 to 1.5808163191276852
Iteration 1 complete. Error: 1.5808163191276852 Total: 257.1535; Orientation: 0.0359; Line Search: 197.6802
<a id="p-3"></a>Iteration 1
<a id="p-2"></a>=-0.4487953229116337;dx=-4.93747079462735E-9
New Minimum: -0.4487953229116337 > -0.45448373708333767
END: th(5875853.91986572)=-0.45448373708333767; dx=-1.962306190227703E-9 evalInputDelta=0.005688414171703959
Fitness changed from -0.4487953229116337 to -0.45448373708333767
Iteration 29 complete. Error: -0.45448373708333767 Total: 60.8205; Orientation: 1.0365; Line Search: 40.2932
Adding measurement 3f5fd883 to history. Total: 3
Rejected: LBFGS Orientation magnitude: 1.434e+03, gradient 7.262e-05, dot -0.710; [e43dc92b-88c7-4bef-8d64-7bf49af310fb = 1.000/1.000e+00]
Orientation rejected. Popping history element from -0.4339739408240664, -0.43832636981737827, -0.4487953229116337, -0.45448373708333767
LBFGS Accumulation History: 3 points
Removed measurement 4b56971a to history. Total: 4
th(0)=-0.45448373708333767;dx=-5.273860700768678E-9
Armijo: th(1.2659143518518532E7)=-0.44614095120027875; dx=-3.2325730106766817E-10 evalInputDelta=-0.008342785883058923
New Minimum: -0.45448373708333767 > -0.45851746402966076
END: th(6329571.759259266)=-0.45851746402966076; dx=-1.3199939118495762E-9 evalInputDelta=0.004033726946323091
Fitness changed from -0.45448373708333767 to -0.45851746402966076
Iteration 30 complete. Error: -0.45851746402966076 Total: 78.4993; Orientation: 1.0764; Line Search: 57.8412
<a id="p-21"></a>Iteration 30
<a id="p-20"></a>
Final threshold in iteration 31: -0.45851746402966076 (> -Infinity) after 2484.669s (< 3600.000s)
-0.45851746402966076