Table of contents

Model type support

Depending on the type of model different languages, real time scoring, and batch scoring are supported.

Table 1. Summary of deployment for all asset types in Watson Studio Local
Asset type Real time Batch Evaluation Application
Open source model Yes Yes Yes  
Script Yes Yes N/A  
PMML Yes Yes No  
SPSS flow No Yes No  
Custom model Yes Yes No  
Decision Optimization Model     N/A  
Notebook N/A Yes N/A Yes
Shiny application N/A N/A N/A Yes
Table 2. Supported model types in Watson Studio Local
Model Type API Algorithm Types
SparkML   Classification, Regression *
scikit-learn scikit-learn 0.19.1 (Python 2.7 and Python 3.5) - 0.19.1 (GPU-Python 3.5) with pickle or joblib format Classification, Regression *
XGBoost XGBoost 0.7.post3 (Python 2.7 and 3.5) - 0.71 (GPU-Python 3.5) Classification, Regression *
Keras Keras 2.1.3 (Python 2.7 and Python 3.5) - 2.1.5 (GPU-Python 3.5) Classification, Regression *
TensoFlow TensorFlow 1.5.0 (Python 2.7 and Python 3.5) - 1.4.1 (GPU-Python 3.5) Classification, Regression *
R Caret  

* Other algorithm types can be used in notebooks. Notebooks can be deployed for batch scoring.