I want to change the color of my labels after applying a running forest classification. I use the "Color Manager" node to change colors depending on the classification score. However, when I feed everything back into the "Apply Color Settings to Labels" node, it says 'Couldn't find color for label' and sets everything to default color. Maybe I am missing something here, but I do not see why the label names would have changed in between.
I attach an example workflow that throws the error.
I managed to color the labels as you wanted, the issue you faced is caused by a limitation of the apply colors node: "the values in the colored column must be the labels in the labeling."
This means you can not color labels with e.g. names like "label1" using a color
definition created on a double value column. What I did in the attached workflow
is to create a new labeling with the prediction confidence as the labels using the GroupBy node.
You can find some tutorials about color management with labels on the example server:
I hope this helps you,
Thanks a lot! This works perfectly now.
Just a quick question: It looks like the Interactive Segmentation View node overrides the color settings back to random. Is there an option to prevent this?
Currently there is no option to do this, but this is something we should implement, I created an issue on Github for this: https://github.com/knime-ip/knip/issues/432. Thank you for reporting this.
As a workaround you can use the Image Viewer node's ability to render several images together. To do this select both desired image and labeling and select the Combine Images checkbox. This should offer you most of the options of the Interactive Segmentation View. I attached a screenshot that shows these settings.