Regex evaluation metric

The regex metric measures whether the rows in the prediction contain the specified regex expression.

Metric details

Regex is a content validation metric that uses string-based functions to analyze and validate generated LLM output text. The metric is available only when you use the Python SDK to calculate evaluation metrics.

Scope

The regex metric evaluates generative AI assets only.

  • Types of AI assets: Prompt templates
  • Generative AI tasks:
    • Text summarization
    • Content generation
    • Question answering
    • Entity extraction
    • Retrieval augmented generation (RAG)
  • Supported languages: English

Scores and values

The regex metric score indicates whether the prediction rows contain the specified regex expression.

  • Range of values: 0.0-1.0
  • Ratios:
    • At 0: The prediction rows do not contain the specified regex expression
    • At 1: The prediction rows contain the specified regex expression.

Settings

  • Thresholds:
    • Lower limit: 0
    • Upper limit: 1

Parent topic: Evaluation metrics