org.llm4s.rag.benchmark.RAGPipeline
See theRAGPipeline companion class
object RAGPipeline
Attributes
- Companion
- class
- Graph
-
- Supertypes
-
class Objecttrait Matchableclass Any
- Self type
-
RAGPipeline.type
Members list
Value members
Concrete methods
def createEmbeddingClient(config: EmbeddingConfig, resolveEmbeddingProvider: String => Result[EmbeddingProviderConfig]): Result[EmbeddingClient]
Create an embedding client for a specific embedding config.
Create an embedding client for a specific embedding config.
Reads API keys from configuration (environment or application.conf):
- OpenAI: OPENAI_API_KEY
- Voyage: VOYAGE_API_KEY
- Ollama: No API key required
Value parameters
- config
-
Embedding configuration
Attributes
- Returns
-
Embedding client or error
def fromConfig(config: RAGExperimentConfig, llmClient: LLMClient, embeddingClient: EmbeddingClient, tracer: Option[Tracing]): Result[RAGPipeline]
Create a RAG pipeline from experiment configuration.
Create a RAG pipeline from experiment configuration.
Value parameters
- config
-
Experiment configuration
- embeddingClient
-
Embedding client for vectorization
- llmClient
-
LLM client for answer generation
- tracer
-
Optional tracer for cost tracking
Attributes
- Returns
-
Configured pipeline or error
def withStores(config: RAGExperimentConfig, llmClient: LLMClient, embeddingClient: EmbeddingClient, vectorStore: VectorStore, keywordIndex: KeywordIndex, tracer: Option[Tracing]): RAGPipeline
Create a RAG pipeline with custom stores.
Create a RAG pipeline with custom stores.
Value parameters
- config
-
Experiment configuration
- embeddingClient
-
Embedding client
- keywordIndex
-
Custom keyword index
- llmClient
-
LLM client
- tracer
-
Optional tracer for cost tracking
- vectorStore
-
Custom vector store
Attributes
- Returns
-
Configured pipeline
In this article