Skip to main content
Version: 0.0.36

Suggestions

Datasource Suggestions within Lakehousecat enhance user interaction with Custom Models by providing contextually relevant prompts derived from the semantic processing of datasources. This guide explains how to utilize and manage these suggestions for optimal model interaction.

Suggestion Generation Process​

Suggestions are automatically generated as semantic layers are built around your datasource. These suggestions aid users by offering insights and prompts tailored to the specific context of your data and model configurations.

Accessibility and Management​

View Options​

  • List View: Access the suggestions in a detailed list format for in-depth evaluation.
  • Card View: Utilize the card view for a more graphical representation of the suggestions, making quick assessments easier.

Suggestion Actions​

You have several options for managing suggestions:

  • Evaluate: Review each suggestion to decide its relevance and applicability.
  • Remove: Delete any suggestions that are deemed unnecessary or redundant.
  • Lock: Lock suggestions to prevent them from being altered or removed, ensuring consistent use.
  • Add: Create and insert custom suggestions to better suit your specific needs and enhance model interaction.

Integration with Custom Models​

Suggestions are designed to work in tandem with Custom Models. Ensure they are activated within the model configuration settings if not automatically integrated. This activation allows users to leverage suggestions for more efficient and meaningful interactions with the Custom Models.

Suggestion Structure​

Each suggestion is structured to provide clarity and functionality:

  • Dual Titles: Use two titles to categorize the suggestion effectively and facilitate easier organization.
  • Custom Prompts: Define the prompts within each suggestion to guide user interaction with the models.

Purpose and Utility​

Suggestions serve as interactive aids, enhancing user engagement with Custom Models. By strategically utilizing these suggestions, users can streamline model interactions and extract valuable insights, thereby optimizing data-driven decision-making.

Best Practices​

  • Contextual Relevance: Tailor suggestions to align with the specific data context and model objectives for enhanced user benefit.
  • Prompt Development: Craft prompts with clear intent and actionable insights to maximize interaction effectiveness.

By implementing these guidelines, you can effectively manage and use Datasource Suggestions within Lakehousecat to promote dynamic and intelligent model interactions.