In SAS Viya, which capability allows external Python scikit-learn models to be registered in SAS Model Manager for centralized governance?
Select an answer to reveal the explanation.
Short Explanation
Here's the deal — you need to package the scikit-learn model as a ZIP archive with the model binary, score.py, and metadata files, then import it through Model Manager. Raw .pkl uploads won't cut it, and there's no ASTORE bridge or PROC HPML PYTHON option for this.
Full Explanation
The correct capability is packaging the scikit-learn model as a ZIP archive with the model binary, score.py, and metadata files such as fileMetadata.json and modelProperties.json, then importing it via the Model Manager API. Model Manager needs the serialized artifact, Python scoring code, and schema metadata. Uploading only a .pkl file is wrong because a raw pickle lacks score code and metadata. Converting the model to an ASTORE using a SAS Python bridge is wrong because no such bridge exists. Using PROC HPML with a PYTHON= option is wrong because PROC HPML is not a real SAS procedure for this purpose.