Menu
I have saved a trained model (deep net, but it is more general I think) in H2O. Now I want to load it by another instance of H2O and use it for scoring, but the problem is, that the version of H2O used for training (3.10.0.3) was different than the one I started the production cluster with (3.10.0.6). The error message is quite self-explanatory
ERROR MESSAGE:
Found version 3.10.0.3, but running version 3.10.0.6
Is there a way to migrate the saved model between versions? Or am I stuck with using the same version of H2O for training and scoring?
650
H2O's AutoML can be used for automating the machine learning workflow, which includes automatic training and tuning of many models within a user-specified time-limit. H2O offers a number of model explainability methods that apply to AutoML objects (groups of models), as well as individual models (e.g. leader model).
a quick way to see the model's corresponding h2o version number (if you used h2o. saveModel() to save the model) is to open the model's file - you should be able to see the version number within the first line (in the form 3.10. 4.2).
What is a MOJO? ¶ A MOJO (Model Object, Optimized) is an alternative to H2O's POJO. As with POJOs, H2O allows you to convert models that you build to MOJOs, which can then be deployed for scoring in real time.
H2O from Python is a tool for rapidly turning over models, doing data munging, and building applications in a fast, scalable environment without any of the mental anguish about parallelism and distribution of work.
Yes, you are stuck using the same version for training and scoring. No migration route.
(You can export a model as a POJO, which can be bundled with the version of h2o-genmodel.jar that it needs. But that requires writing Java code to get the data in and results out, which is not ideal if you are using R code for data preparation.)
This has been discussed on the h2o-stream mailing list before, but I couldn't see a feature request ticket for it, so I just created one: https://0xdata.atlassian.net/browse/PUBDEV-3432
52
If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With
