Machine Learning Recipes
Modern application delivery teams can integrate Machine Learning (ML) recipes, which serve as building blocks for insights such as recommendation engines, user segmentation, NLP/NLU pipelines and anomaly/novelty detection.
These recipes not only deliver results faster, but also make the process repeatable. Several common pre-processing tasks (data profiling, data quality inspection, imputation, feature engineering, etc.) are handled automatically and at scale.
Data scientists can leverage several machine learning recipes such as prediction, classification, clustering and ranking to solve real-world problems with strategic business apps.
Build process flows that can orchestrate the execution of various recipes in a chain to solve business problems using the low-code wizard interface. Process flows help operationalize model building steps and execute scoring on incremental new data.
Data scientists automate delivery of models and solutions from development to production without worrying about data size or infrastructure scaling issues.
Collaboration with Developers
Data science solutions are deployed at scale and the insights are consumed from prebuilt connectors for software developers that deliver seamless integration with modern business apps deployed on the platform.
Productivity is achieved with automatic feature engineering, algorithm selection, model selection, model lifecycle and ensemble management, and recipe and process flow management