Last week at the meet the developers Q&A you showed a PLS node written in python. Will this be made available soon? Many thanks.
we were basically waiting for someone to ask We’ll release the PLS node as an advanced tutorial for writing Python Node Extensions soon, but we are not planning to release the node as an installable KNIME extension on its own.
We’d prefer to have a full collection of scikit-learn algorithms wrapped in Python nodes at some point, but we currently don’t have that on the roadmap. We hope that this is something that could be a community effort.
In the meantime you could build the same functionality using the Python Scripting (Labs) node, assuming you are familiar with Python?
That’s great the tutorial will be perfect and an interesting example to see how to write a node in Python. I have a full set of working PLS, OPLS and PLS-DA nodes I’ve written in R-scripting nodes. I should share these at some point,
Let me know if you want to work together to get those KNIME Interactive R Statistics Integration nodes inside components.
We should be able to also couple them with the Conda Environment Propagation node so that R and the library you are using is installed automatically when someone drags and drop your component from KNIME Hub!
More on that if you scroll here:
Yes I was thinking of turning them into components, The PCA one is done but not the PLS ones yet. Also need to get permission to share from my employers but that should be a formality.
Great! Once you get the green light, please share a first draft of the component on your KNIME Hub space. You can find details on how to do that here:
Then share the link of the component you publicly shared in here and my team can provide feedback to further improve it.
Thanks a lot for the contribution in advance!