Code from BasicOptimizer.scala:75 executed in 279.92 seconds (2.844 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: 923828410001600
Reset training subject: 923832937838000
Adding measurement 215f76f4 to history. Total: 0
LBFGS Accumulation History: 1 points
Constructing line search parameters: GD+Trust
th(0)=1.6109168827533722;dx=-6.91516442073461E-7
New Minimum: 1.6109168827533722 > 1.6109162494540215
WOLFE (weak): th(2.154434690031884)=1.6109162494540215; dx=-6.915170055508436E-7 evalInputDelta=6.332993507385254E-7
New Minimum: 1.6109162494540215 > 1.6109150797128677
WOLFE (weak): th(4.308869380063768)=1.6109150797128677; dx=-6.915244797967002E-7 evalInputDelta=1.8030405044555664E-6
New Minimum: 1.6109150797128677 > 1.6109100356698036
WOLFE (weak): th(12.926608140191302)=1.6109100356698036; dx=-6.915212358180307E-7 evalInputDelta=6.847083568572998E-6
New Minimum: 1.6109100356698036 > 1.610889084637165
WOLFE (weak): th(51.70643256076521)=1.610889084637165; dx=-6.916170559020059E-7 evalInputDelta=2.7798116207122803E-5
New Minimum: 1.610889084637165 > 1.6107778549194336
WOLFE (weak): th(258.53216280382605)=1.6107778549194336; dx=-6.914404770245374E-7 evalInputDelta=1.3902783393859863E-4
New Minimum: 1.6107778549194336 > 1.610082745552063
WOLFE (weak): th(1551.1929768229563)=1.610082745552063; dx=-6.91488259498342E-7 evalInputDelta=8.341372013092041E-4
New Minimum: 1.610082745552063 > 1.6050939932465553
WOLFE (weak): th(10858.350837760694)=1.6050939932465553; dx=-6.86220285160728E-7 evalInputDelta=0.005822889506816864
New Minimum: 1.6050939932465553 > 1.565880797803402
WOLFE (weak): th(86866.80670208555)=1.565880797803402; dx=-6.422097730482063E-7 evalInputDelta=0.045036084949970245
New Minimum: 1.565880797803402 > 1.343377560377121
END: th(781801.26031877)=1.343377560377121; dx=-2.6850213640761446E-7 evalInputDelta=0.2675393223762512
Fitness changed from 1.6109168827533722 to 1.343377560377121
Iteration 1 complete. Error: 1.343377560377121 Total: 49.3837; Orientation: 0.0627; Line Search: 37.6423
<a id="p-3"></a>Iteration 1
<a id="p-2"></a>=0.20588916540145874;dx=-4.1304387512094816E-8
New Minimum: 0.20588916540145874 > 0.18196988105773926
END: th(1624556.091764473)=0.18196988105773926; dx=-2.2994071815183374E-9 evalInputDelta=0.023919284343719482
Fitness changed from 0.20588916540145874 to 0.18196988105773926
Iteration 14 complete. Error: 0.18196988105773926 Total: 13.3092; Orientation: 1.9178; Line Search: 7.6845
Adding measurement 336dd178 to history. Total: 3
Rejected: LBFGS Orientation magnitude: 6.517e+02, gradient 1.618e-04, dot -0.680; [f2dada80-4fb5-4eba-87e8-63c6a2ca45b1 = 1.000/1.000e+00]
Orientation rejected. Popping history element from 0.2558424770832062, 0.22729569673538208, 0.20588916540145874, 0.18196988105773926
LBFGS Accumulation History: 3 points
Removed measurement 5e1ab04e to history. Total: 4
th(0)=0.18196988105773926;dx=-2.618975250247354E-8
New Minimum: 0.18196988105773926 > 0.17047765851020813
WOLF (strong): th(3500000.000000001)=0.17047765851020813; dx=5.43498728953671E-9 evalInputDelta=0.011492222547531128
New Minimum: 0.17047765851020813 > 0.1627921462059021
END: th(1750000.0000000005)=0.1627921462059021; dx=-4.073344107536047E-9 evalInputDelta=0.019177734851837158
Fitness changed from 0.18196988105773926 to 0.1627921462059021
Iteration 15 complete. Error: 0.1627921462059021 Total: 17.2149; Orientation: 1.9656; Line Search: 11.6244
<a id="p-15"></a>Iteration 15
<a id="p-14"></a>
Final threshold in iteration 16: 0.1627921462059021 (> -Infinity) after 279.923s (< 1800.000s)
0.1627921462059021