it feels like 5.1 has become quite instable. Sorting a table (in preview), which contains about an estiamte of 800 to 1000 MB of images, their binary objects and some text values causes this error.
Scrolling to the right where the binary object column resides causes the preview to freeze (moderate CPU but no disc usage). the preview window cannot be close requiring me to force quit Knime.
There is another post of mine about the binary object column causing issuses but predominantly with Selenium nodes. This time it’s a core node of Knime (albeit in preview).
the default row based backend. In case I have some sepcial setup I usually point that out like in the upcoming post, spoiler alert, about a performance regression of the new columnar backend.
circling back to this issue as it appeared again but in a different context using a Phyton Script node while trying to extract the meta data form an image. Not even saving is possible:
I had to reset the failing node to save the workflow. Upon trying to replicate the save issue as well as the error thrown in the Python node, they were not triggered again but instead this one:
❌ KeyError: 'Sample Images'
Traceback (most recent call last):
File "<string>", line 26, in <module>
File "C:\Program Files\knime_5.2.0\bundling\envs\org_knime_pythonscripting\Lib\site-packages\pandas\core\frame.py", line 3761, in __getitem__
indexer = self.columns.get_loc(key)
^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Program Files\knime_5.2.0\bundling\envs\org_knime_pythonscripting\Lib\site-packages\pandas\core\indexes\base.py", line 3655, in get_loc
raise KeyError(key) from err
Thanks for providing a workflow, I can reproduce the original error when fixing your script (it referenced column “Sample Images” but only has “Image” available). We’ll take a look (internal ticket reference AP-22425)!