Skip to main content

LangChain helpers for working with Logseq documents

Project description

Langchain Logseq

Collection of Langchain utilities for working with Logseq files.

Components

This section provides an overview of the components provided, listed by type

Retrievers

Retrievers inject context into a conversation. Works in tandem with a Contextualizer and Document Loader.

  • Input:
    • natural-language user-input, usually query-like
    • (optional) chat history
  • Output:
    • list of Documents to provide context for an LLM to answer the user-input

Implementations

  • LogseqJournalDateRangeRetriever
    • retrieve Logseq journal Documents, intended for queries that require context from a date range
    • required to set up:
      • RetrieverContextualizer
      • LogseqJournalLoader
    • examples:
      • "What did I do over Christmas break 2024?"
      • "How did I spend the last Independence Day?"

Contextualizers

Contextualizers serve as the bridge between natural-language input and a downstream component that handles fetching of relevant Documents.

  • Input:
    • natural-language user-input, usually query-like
    • (optional) chat history
  • Output:
    • structured downstream query, based on

In this library, an instance of RetrieverContextualizer is provided directly to Retrievers during the latter's instantiation. To set up the RetrieverContextualizer, provide RetrieverContextualizerProps, which includes:

  • llm - this is the backbone of the contextualizer
  • prompt - instructions provided to the LLM
  • output_schema - (optional) structured schema used to fetch relevant Documents
    • if no schema is provided, a string shall be returned instead
  • other flags and settings

Loaders

Loaders are one type of component that can fetch relevant Documents. Loaders are typically specific to a corresponding Retriever component.

  • Input:
    • each loader specifies its own schema
      • the Contextualizer is usually responsible for creating an instance of the query obj to act upon
  • Output:
    • list[Document]

Implementations

  • LogseqJournalFilesystemLoader
    • loads from the filesystem, where journal files are expected to be present at specified path

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

langchain_logseq-0.1.0.tar.gz (9.5 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

langchain_logseq-0.1.0-py3-none-any.whl (11.2 kB view details)

Uploaded Python 3

File details

Details for the file langchain_logseq-0.1.0.tar.gz.

File metadata

  • Download URL: langchain_logseq-0.1.0.tar.gz
  • Upload date:
  • Size: 9.5 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.1.0 CPython/3.8.10

File hashes

Hashes for langchain_logseq-0.1.0.tar.gz
Algorithm Hash digest
SHA256 17df4934779ac40305b1f4554ebeea3c8afdb0634b6bd48748734f4c16472cfd
MD5 7e13f88fcc8afb945988d50e8e160367
BLAKE2b-256 6f744425be08e6c986f8647345903ba1e146f2fd70e0781e6f7134131491e741

See more details on using hashes here.

File details

Details for the file langchain_logseq-0.1.0-py3-none-any.whl.

File metadata

File hashes

Hashes for langchain_logseq-0.1.0-py3-none-any.whl
Algorithm Hash digest
SHA256 72498c483ca803cf51e851b19483dd322da9b44dffd4293eceed533ef21130fd
MD5 aba3178e506f183422d6651104a8de27
BLAKE2b-256 398b04b495ff33e6b60070e233e0f4f769fdfb8ddc73ade984edcafe83dbf229

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page