Hi, I'm trying to finetune a model lately.
While the finetune module works perfectly fine on the sampledata/DIC-Erythrocytes provided by your website, it throws an exception on my own data once it starts:
Exception in thread "Thread-5" java.lang.IllegalArgumentException: n out of range: 2
at ij.ImagePlus.convertIndexToPosition(ImagePlus.java:1824)
at de.unifreiburg.unet.TrainingSample.getROIPosition(TrainingSample.java:1656)
at de.unifreiburg.unet.TrainingSample.createLabelsAndWeightBlobsFromRois(TrainingSample.java:1256)
at de.unifreiburg.unet.TrainingSample.createLabelsAndWeightsBlobs(TrainingSample.java:195)
at de.unifreiburg.unet.FinetuneJob.run(FinetuneJob.java:1983)
I figured it out myself.
When I was labeling my data pictures I made a stack of two channels, in order to check the accuracy of my labeling.
Reverting to labeling on single channel data resolved the problem.
Hi, I'm trying to finetune a model lately. While the finetune module works perfectly fine on the sampledata/DIC-Erythrocytes provided by your website, it throws an exception on my own data once it starts:
And the log window halts at:
Info of my image:
Any idea what might went wrong?