haive.core.engine.retriever.providers.PineconeHybridSearchRetrieverConfig

from typing import Any. Pinecone Hybrid Search Retriever implementation for the Haive framework.

This module provides a configuration class for the Pinecone Hybrid Search retriever, which combines vector similarity search with keyword search using Pinecone’s hybrid search capabilities.

The PineconeHybridSearchRetriever works by: 1. Connecting to a Pinecone index 2. Performing both vector and keyword search 3. Combining results using Pinecone’s hybrid scoring

This retriever is particularly useful when: - Using Pinecone as the vector database - Need both semantic and keyword search - Want Pinecone’s optimized hybrid search performance - Building applications that benefit from combined search approaches

The implementation integrates with LangChain’s PineconeHybridSearchRetriever while providing a consistent Haive configuration interface with secure API key management.

Classes

PineconeHybridSearchRetrieverConfig

Configuration for Pinecone Hybrid Search retriever in the Haive framework.

Module Contents

class haive.core.engine.retriever.providers.PineconeHybridSearchRetrieverConfig.PineconeHybridSearchRetrieverConfig

Bases: haive.core.common.mixins.secure_config.SecureConfigMixin, haive.core.engine.retriever.retriever.BaseRetrieverConfig

Configuration for Pinecone Hybrid Search retriever in the Haive framework.

This retriever uses Pinecone’s hybrid search capabilities to combine vector similarity search with keyword search for better retrieval performance.

retriever_type

The type of retriever (always PINECONE).

Type:

RetrieverType

api_key

Pinecone API key (auto-resolved from PINECONE_API_KEY).

Type:

Optional[SecretStr]

index_name

Name of the Pinecone index to search.

Type:

str

environment

Pinecone environment (e.g., “us-east1-gcp”).

Type:

str

top_k

Number of documents to retrieve (default: 10).

Type:

int

alpha

Weight for vector vs sparse search (0.0 = sparse only, 1.0 = vector only).

Type:

float

Examples

>>> from haive.core.engine.retriever import PineconeHybridSearchRetrieverConfig
>>>
>>> # Create the pinecone hybrid search retriever config
>>> config = PineconeHybridSearchRetrieverConfig(
...     name="pinecone_hybrid_retriever",
...     index_name="my-hybrid-index",
...     environment="us-east1-gcp",
...     top_k=5,
...     alpha=0.5  # Equal weight to vector and sparse search
... )
>>>
>>> # Instantiate and use the retriever
>>> retriever = config.instantiate()
>>> docs = retriever.get_relevant_documents("machine learning algorithms")

get_input_fields()

Return input field definitions for Pinecone Hybrid Search retriever.

Return type:

dict[str, tuple[type, Any]]

get_output_fields()

Return output field definitions for Pinecone Hybrid Search retriever.

Return type:

dict[str, tuple[type, Any]]

instantiate()

Create a Pinecone Hybrid Search retriever from this configuration.

Returns:

Instantiated retriever ready for hybrid search.

Return type:

PineconeHybridSearchRetriever

Raises:

• ImportError – If required packages are not available.

• ValueError – If API key or configuration is invalid.