Can ConversationalRetrievalChain.from_llm still be used in v0.2.9 langchain release? Does it get deprecated?

[hteeyeoh]

Checked other resources

• I added a very descriptive title to this question.
• I searched the LangChain documentation with the integrated search.
• I used the GitHub search to find a similar question and didn't find it.

Commit to Help

• I commit to help with one of those options 👆

Example Code

from langchain.chains import ConversationalRetrievalChain
qa_chain = ConversationalRetrievalChain.from_llm(
            llm=llm,
            retriever=retriever,
            return_source_documents=True,
            condense_question_prompt=CONDENSED_PROMPT,
            combine_docs_chain_kwargs={"prompt": PROMPT},
            condense_question_llm=condense_question_llm,
        )

Description

LLMChain will be deprecated and replaced with LangChain runnables. Does it apply the same with ConversationRetrievalChain?

System Info

langchain=0.2.9

[dosubot[bot]]

ConversationalRetrievalChain.from_llm can still be used in the v0.2.9 LangChain release, but the ConversationalRetrievalChain class itself is deprecated. The deprecation notice indicates that while the class is deprecated, it is still available for use until a future release ^[1][2].

Here's the relevant part of the deprecation notice:

## Code targeted for removal

Code that has better alternatives available and will eventually be removed, so there’s only a single way to do things. (e.g., `predict_messages` method in ChatModels has been deprecated in favor of `invoke`).

### langchain

#### ConversationalRetrievalChain

In module: `chains.retrieval_qa.base`
Deprecated: 0.1.17
Removal: 0.3.0

Alternative: [create_retrieval_chain](https://api.python.langchain.com/en/latest/chains/langchain.chains.retrieval.create_retrieval_chain.html#langchain-chains-retrieval-create-retrieval-chain)

You can continue using ConversationalRetrievalChain.from_llm in v0.2.9, but it is recommended to transition to create_retrieval_chain for future compatibility and improvements.

To continue talking to Dosu, mention @dosu.

[hteeyeoh]

@dosu can you give me example how to switch from using ConversationalRetrievalChain.from_llm to create_retrieval_chain?

[dosubot[bot]]

To switch from using ConversationalRetrievalChain.from_llm to create_retrieval_chain, you can follow this example:

from langchain.chains import (
    create_history_aware_retriever,
    create_retrieval_chain,
)
from langchain.chains.combine_documents import create_stuff_documents_chain
from langchain_core.prompts import ChatPromptTemplate, MessagesPlaceholder
from langchain_openai import ChatOpenAI

# Initialize your retriever
retriever = ...  # Your retriever

# Initialize the language model
llm = ChatOpenAI()

# Define the prompt to contextualize the question
contextualize_q_system_prompt = (
    "Given a chat history and the latest user question "
    "which might reference context in the chat history, "
    "formulate a standalone question which can be understood "
    "without the chat history. Do NOT answer the question, just "
    "reformulate it if needed and otherwise return it as is."
)
contextualize_q_prompt = ChatPromptTemplate.from_messages(
    [
        ("system", contextualize_q_system_prompt),
        MessagesPlaceholder("chat_history"),
        ("human", "{input}"),
    ]
)
history_aware_retriever = create_history_aware_retriever(
    llm, retriever, contextualize_q_prompt
)

# Define the prompt to answer the question
qa_system_prompt = (
    "You are an assistant for question-answering tasks. Use "
    "the following pieces of retrieved context to answer the "
    "question. If you don't know the answer, just say that you "
    "don't know. Use three sentences maximum and keep the answer "
    "concise."
    "\n\n"
    "{context}"
)
qa_prompt = ChatPromptTemplate.from_messages(
    [
        ("system", qa_system_prompt),
        MessagesPlaceholder("chat_history"),
        ("human", "{input}"),
    ]
)

# Create the question-answering chain
question_answer_chain = create_stuff_documents_chain(llm, qa_prompt)

# Create the retrieval chain
rag_chain = create_retrieval_chain(
    history_aware_retriever, question_answer_chain
)

# Usage example
chat_history = []  # Collect chat history here (a sequence of messages)
query = "Your query here"
response = rag_chain.invoke({"input": query, "chat_history": chat_history})
print(response)

This example demonstrates how to set up the create_retrieval_chain using the provided components and prompts. You can refer to the LangChain documentation for more advanced usage and customization ^[1][2][3].

To continue talking to Dosu, mention @dosu.

[Thegr8carlos]

Hey bro, did you have the chance to solve make that change?