LangChain

from langchain.llms import …

• EdenAI(edenai_api_key: str, provider: str, model: str)

• OpenAI(api_key: str, model: str)

from langchain.embeddings import …

• edenai.EdenAiEmbeddings(edenai_api_key: str, provider: str)

• OpenAIEmbeddings()

from langchain.prompts import …

• PromptTemplate.from_template(template: str)

  • .format(**kwargs: Any)

from langchain.output_parser import …

• ResponseSchema(name: str, description: str, type: str)

• StructuredOutputParser.from_response_schemas(response_schemas: list[ResponseSchema])

  • .get_format_instructions()

  • .parse(text: str)

from langchain.memory import …

• BaseChatMemory()

  • .clear()

  • save_context(inputs: dict[str, Any], outputs: dct[str, str])

• ConversationBufferMemory()

• ConversationBufferWindowMemory(k: int)

• ConversationTokenBufferMemory(max_token_limit: int)

• ConversationSummaryBufferMemory(llm: BaseLanguageModel, max_token_limit: int)

from langchain.chains import …

• ConversationChain(llm: BaseLanguageModel, memory: BaseMemory)

  • .predict(input: str)

• LLMChain(llm: BaseLanguageModel, prompt:: str)

  • run(str)

• SimpleSequentialChain(chains: listChain])

  • run(str)

• SequentialChain(chains: list[Chain], input_variables: list[str], output_variables: list[str])

  • run(str)

• RetrievalQA

from langchain.document_loaders import …

• CSVLoader(file_path: str, csv_args: dict[str, Any])

  • csv_args: "delimiter", "quotechar", "fieldnames", …

  • .load()

• PyPDFLoader(file_path: str)

  • .load()

from langchain.vectorstores import …

• DocArrayInMemorySearch

  • from_documents(documents: list[Document], embeddings: Embeddings)

  • similarity_search(query: str)

  • .as_retriever()

from langchain.indexes import …

• VectorstoreIndexCreator(vectorstore_cls: Chroma

  )

  • .from_loaders(loaders: list[BaseLoader])

  • .query(question: str)