Code from BasicOptimizer.scala:75 executed in 193.19 seconds (2.148 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]

Logging:

    Reset training subject: 911836372354600
    Reset training subject: 911839704629500
    Adding measurement 238142bd to history. Total: 0
    LBFGS Accumulation History: 1 points
    Constructing line search parameters: GD+Trust
    th(0)=-0.0909639298915863;dx=-3.769855758406388E-9
    Armijo: th(2.154434690031884)=-0.0909639298915863; dx=-3.7384166608437395E-9 evalInputDelta=0.0
    Armijo: th(1.077217345015942)=-0.0909639298915863; dx=-3.738416659641526E-9 evalInputDelta=0.0
    Armijo: th(0.3590724483386473)=-0.0909639298915863; dx=-3.738416659641526E-9 evalInputDelta=0.0
    Armijo: th(0.08976811208466183)=-0.0909639298915863; dx=-3.738416659641526E-9 evalInputDelta=0.0
    Armijo: th(0.017953622416932366)=-0.0909639298915863; dx=-3.738416659641526E-9 evalInputDelta=0.0
    Armijo: th(0.002992270402822061)=-0.0909639298915863; dx=-3.738416659641526E-9 evalInputDelta=0.0
    WOLFE (weak): th(4.2746720040315154E-4)=-0.0909639298915863; dx=-3.738416659641526E-9 evalInputDelta=0.0
    WOLFE (weak): th(0.0017098688016126062)=-0.0909639298915863; dx=-3.738416659641526E-9 evalInputDelta=0.0
    Armijo: th(0.0023510696022173336)=-0.0909639298915863; dx=-3.738416659641526E-9 evalInputDelta=0.0
    Armijo: th(0.00203046920191497)=-0.0909639298915863; dx=-3.738416659641526E-9 evalInputDelta=0.0
    Armijo: th(0.001870169001763788)=-0.0909639298915863; dx=-3.738416659641526E-9 evalInputDelta=0.0
    WOLFE (weak): th(0.0017900189016881972)=-0.0909639298915863; dx=-3.738416659641526E-9 evalInputDelta=0.0
    WOLFE (weak): th(0.0018300939517259926)=-0.0909639298915863; dx=-3.738416659641526E-9 evalInputDelta=0.0
    Armijo: th(0.0018501314767448902)=-0.0909639298915863; dx=-3.738416659641526E-9 evalInputDelta=0.0
    WOLFE (weak): th(0.0018401127142354413)=-0.0909639298915863; dx=-3.738416659641526E-9 evalInputDelta=0.0
    Armijo: th(0.0018451220954901657)=-0.0909639298915863; dx=-3.738416659641526E-9 evalInputDelta=0.0
    Armijo: th(0.0018426174048628034)=-0.0909639298915863; dx=-3.738416659641526E-9 evalInputDelta=0.0
    Armijo: th(0.0018413650595491223)=-0.0909639298915863; dx=-3.73841

...skipping 1752 bytes...

    0
    Armijo: th(0.009913773265330113)=-0.0909639298915863; dx=-3.738416659641526E-9 evalInputDelta=0.0
    WOLFE (weak): th(0.008922395938797102)=-0.0909639298915863; dx=-3.738416659641526E-9 evalInputDelta=0.0
    Armijo: th(0.009418084602063607)=-0.0909639298915863; dx=-3.738416659641526E-9 evalInputDelta=0.0
    WOLFE (weak): th(0.009170240270430355)=-0.0909639298915863; dx=-3.738416659641526E-9 evalInputDelta=0.0
    Armijo: th(0.009294162436246981)=-0.0909639298915863; dx=-3.738416659641526E-9 evalInputDelta=0.0
    Armijo: th(0.009232201353338668)=-0.0909639298915863; dx=-3.738416659641526E-9 evalInputDelta=0.0
    WOLFE (weak): th(0.00920122081188451)=-0.0909639298915863; dx=-3.738416659641526E-9 evalInputDelta=0.0
    Armijo: th(0.00921671108261159)=-0.0909639298915863; dx=-3.738416659641526E-9 evalInputDelta=0.0
    Armijo: th(0.00920896594724805)=-0.0909639298915863; dx=-3.738416659641526E-9 evalInputDelta=0.0
    Armijo: th(0.00920509337956628)=-0.0909639298915863; dx=-3.738416659641526E-9 evalInputDelta=0.0
    Armijo: th(0.009203157095725396)=-0.0909639298915863; dx=-3.738416659641526E-9 evalInputDelta=0.0
    WOLFE (weak): th(0.009202188953804953)=-0.0909639298915863; dx=-3.738416659641526E-9 evalInputDelta=0.0
    WOLFE (weak): th(0.009202673024765174)=-0.0909639298915863; dx=-3.738416659641526E-9 evalInputDelta=0.0
    WOLFE (weak): th(0.009202915060245285)=-0.0909639298915863; dx=-3.738416659641526E-9 evalInputDelta=0.0
    WOLFE (weak): th(0.009203036077985342)=-0.0909639298915863; dx=-3.738416659641526E-9 evalInputDelta=0.0
    WOLFE (weak): th(0.009203096586855369)=-0.0909639298915863; dx=-3.738416659641526E-9 evalInputDelta=0.0
    mu ~= nu (0.009203096586855369): th(0.0)=-0.0909639298915863
    Fitness changed from -0.0909639298915863 to -0.0909639298915863
    Static Iteration Total: 87.4403; Orientation: 0.0651; Line Search: 80.6141
    Iteration 2 failed. Error: -0.0909639298915863
    Previous Error: 0.0 -> -0.0909639298915863
    Optimization terminated 2
    Final threshold in iteration 2: -0.0909639298915863 (> -Infinity) after 193.190s (< 720.000s)
    

Returns:

    -0.0909639298915863