in Knime there are 2 predictors nodes “H2O MOJO Predictor (Autoencoder)” and “H2O MOJO Predictor (Word Embedding)” without a related learning node. How should I use these 2 nodes?
Thanks in advance
It seems you are really testing our H2O Integration. That’s great. Happy to receive your feedback :-). Regarding the MOJOs for Autoencoder and Word Embedding: In case you’ve created a MOJO with H2O itself (e.g. via Python or R) you can then use those MOJOs using those two nodes. Autoencoders and Word Embedding in KNIME can be done using the KNIME DL4J or KNIME Deep Learning - Keras Integrations.
Does that help?
Many thanks for your answer, much appreciated! Yes, I am testing H2O because it seems very powerful and fast! It would be good to also add “H2O Autoencoder Learning node” and “H2O Word Embedding Learning node” in Knime.
By chance, do you have any plan to do that?