DocsRuntime & DeploymentPlugins: Post-PredictOverview

Post-Predict Plugins

Post-predict plugins are used to postprocess the output of the predictor before it is returned to the client. This allows you to customize the output of the predictor to meet your specific needs and requirements. Post-predict plugins can be used to add new postprocessing algorithms, formatters, and other components to the scoring pipeline.

The following default post-predict plugins are available in the ecosystem.Ai platform:

  • Post Score Basic plugin: This plugin is used to format the output of the predictor before it is returned to the client.
  • Post Score Dynamic Engagement plugin: This plugin is used with the dynamic configurations for dynamic recommenders, experimentation, real-time convergence models etc.
  • Post Score Recommender Offers plugin: This plugin is used as a default recommender system to rank and filter product offers based on the model’s predicted probabilities.
Last updated on