LangChain_embedding-distance.md

LangChain: EmbeddingDistance

• Dimension: Generated Answer <-> GroundTruth Answer
• Reference: Evaluating RAG Architectures on Benchmark Tasks
• Type: Semantic Similarity

Simply calculate the embedding distance between the generated answer and the ground truth answer.

Calculation

INPUT

prediction: |-
  Toolkits in LangChain are collections of tools that allow agents to interact with various services or data. Examples include:

  - **SQLDatabaseToolkit:** Helps agents query and interact with SQL databases [0].
  - **GitHubToolkit:** Enables agents to manage issues, pull requests, and comments on GitHub [1][2].
  - **GmailToolkit:** Allows agents to read, send, update, and delete emails in Gmail [3][2].

reference: Toolkits are collections of tools that are designed to be used together for specific tasks and have convenience loading methods.

OUTPUT

score: 0.14001019253292135