Function reference

Vectorisers

Utilities to project text into numerical representation in a semantic vector space

vectorisers	This module provides classes for creating and utilizing embedding models.
vectorisers.base	This module provides classes for creating and utilizing embedding models from different services.
vectorisers.base.VectoriserBase	Abstract base class for all vectorisers.
vectorisers.base.VectoriserBase.transform	Transforms input text(s) into embeddings.
vectorisers.huggingface.HuggingFaceVectoriser	A general wrapper class for Huggingface Transformers models to generate text embeddings.
vectorisers.ollama.OllamaVectoriser	A wrapper class allowing a locally-running ollama server to generate text embeddings.
vectorisers.gcp.GcpVectoriser	A class for embedding text using Google Cloud Platform’s GenAI API.

Indexers

Creation of Vector Stores for efficient similarity search and retrieval

indexers	This module provides functionality for creating a VectorStore from a CSV (text) file.
indexers.main	This module provides functionality for creating a VectorStore from a CSV (text) file.
indexers.VectorStore	A class to model and create VectorStore objects for building and searching vector databases from CSV text files.
indexers.dataclasses	This module defines dataclasses for structuring and validating input and output data for
indexers.dataclasses.VectorStoreSearchInput	DataFrame-like object for forming and validating search query input data.
indexers.dataclasses.VectorStoreSearchOutput	DataFrame-like object for storing and validating search results with rankings
indexers.dataclasses.VectorStoreReverseSearchInput	DataFrame-like object for forming and validating reverse search query input data.
indexers.dataclasses.VectorStoreReverseSearchOutput	DataFrame-like object for storing reverse search results.
indexers.dataclasses.VectorStoreEmbedInput	DataFrame-like object for forming and validating text data to be embedded.
indexers.dataclasses.VectorStoreEmbedOutput	DataFrame-like object for storing and validating embedded vectors and associated metadata.
indexers.VectorStore.embed	Converts text (provided via a VectorStoreEmbedInput object) into vector embeddings using the Vectoriser.
indexers.VectorStore.search	Searches the VectorStore using queries from a VectorStoreSearchInput object.
indexers.VectorStore.reverse_search	Reverse searches the VectorStore using a VectorStoreReverseSearchInput object.
indexers.VectorStore.from_filespace	Creates a VectorStore instance from stored metadata and Parquet files.

Hooks

Creation of Pre and post Processing hooks for improved search and RAG

indexers.hooks	This module contains a factory for, and example pre-built hooks. Allowing users to manipulate the content of ClassifAI dataclass objects as they enter or leave the VectorStore.
indexers.hooks.hook_factory
indexers.hooks.default_hooks	Submodule containing the prebuilt hooks for the service.
indexers.hooks.default_hooks.preprocessing
indexers.hooks.default_hooks.postprocessing
indexers.hooks.hook_factory.HookBase	Abstract base class for all post-processing hooks requiring customisation.
indexers.hooks.default_hooks.preprocessing.CapitalisationStandardisingHook	A pre-processing hook to handle upper-/lower-/sentence-/title-casing.
indexers.hooks.default_hooks.postprocessing.DeduplicationHook	A post-processing hook to remove duplicate knowledgebase entries, i.e. entries with the same label.
indexers.hooks.default_hooks.postprocessing.RagHook	A post-processing hook to perform Retrieval Augmented Generation.

Servers

Expose ClassifAI functionality via Fast-API endpoints

servers	This module provides functionality for creating or extending a REST-API service.
servers.main	This module provides functionality for creating a start a restAPI service.
servers.get_router	Create and return a FastAPI.APIRouter with search endpoints.
servers.get_server	Create and return a FastAPI server with search endpoints.
servers.run_server	Create and run a FastAPI server with search endpoints.
servers.make_endpoints	Create and register the different endpoints to your app.