http://imagej.273.s1.nabble.com/Error-cant-get-around-Expected-tp3684050p3684061.html
> loads the classes from the macros. Still working on it.
Thank you for your attention to this. I updated Fiji and ran the new
Weka. However, now I am getting and error in updating the feature stack.
> On Wed, Jun 29, 2011 at 5:36 PM, Ignacio Arganda-Carreras
> <
[hidden email] <mailto:
[hidden email]>> wrote:
>
> Very interesting, Aryeh. I guess it's a problem with my
> multi-threading. Let me have a look and i will be back to you.
> Thanks for reporting!
>
> ignacio
>
>
> On Wed, Jun 29, 2011 at 3:28 PM, Aryeh Weiss <
[hidden email]
> <mailto:
[hidden email]>> wrote:
>
> I am starting to use the Advanced Weka Segmentation plugins,
> which are proving to be very interesting.
>
> Two issues have come up when using them in macros.
>
> 1. there is some timing issue. Consider the following code:
>
> run("Advanced Weka Segmentation");
>
> wait(3000); //Without this wait statement, the following
> statements will fail and return segmentation faults.
>
> call("trainableSegmentation.__Weka_Segmentation.setFeature",
> "Variance=true");
> call("trainableSegmentation.__Weka_Segmentation.setFeature",
> "Mean=true");
> call("trainableSegmentation.__Weka_Segmentation.setFeature",
> "Maximum=true");
> call("trainableSegmentation.__Weka_Segmentation.setFeature",
> "Gabor=true");
>
> 2. If I do the following:
>
> call("trainableSegmentation.__Weka_Segmentation.__trainClassifier");
>
> then the segmentation will occur. However, a number buttons will
> be grayed out. I will lose "Get Probability", "Apply
> Classifier", and "Save Classifier".
>
> On the other hand, if I do the trainClassifier by manually
> clickingon "Train Classifier", then it does the segmentation and
> leaves all the other things functional. Yet the macro recorder
> records that very command
> call("trainableSegmentation.__Weka_Segmentation.__trainClassifier");
> when I do it by hand.
>
> Since I am trying to automate my procedures as much as possible,
> any help with this second issue would be greatly appreciated.
>
> --aryeh
> --
>