Issue using Chroma as Vector DB

[toshnav]

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.vectorstores.chroma import Chroma
from langchain_openai import AzureOpenAIEmbeddings
from langchain_text_splitters import RecursiveCharacterTextSplitter

text_splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=200)

splits = text_splitter.split_documents(docs)

vectorstore = Chroma.from_documents(documents=splits, embedding=AzureOpenAIEmbeddings(
    deployment_name=os.environ["azure_deployment"],
    api_version=os.environ['openai_api_version'],
    api_key=os.environ['api_key'],
    azure_endpoint=os.getenv("azure_endpoint"),  
))

Description

"I am currently working on a Proof of Concept (POC) and require Chroma as my vector database. However, despite upgrading to the latest version, I am encountering an issue that I have not been able to resolve. I have reviewed the LangChain documentation, but it lacks the necessary details to address this problem. The error I am receiving is as follows:

"81 """Initialize with a Chroma client."""
82 try:
---> 83 import chromadb
84 import chromadb.config
85 except ImportError:

File ~/.local/lib/python3.11/site-packages/chromadb/init.py:86
84 sys.modules["sqlite3"] = sys.modules.pop("pysqlite3")
85 else:
---> 86 raise RuntimeError(
87 "\033[91mYour system has an unsupported version of sqlite3. Chroma
88 requires sqlite3 >= 3.35.0.\033[0m\n"
89 "\033[94mPlease visit
90 https://docs.trychroma.com/troubleshooting#sqlite to learn how
91 to upgrade.\033[0m"
92 )
95 def configure(**kwargs) -> None: # type: ignore
96 """Override Chroma's default settings, environment variables or .env files"""

RuntimeError: Your system has an unsupported version of sqlite3. Chroma requires sqlite3 >= 3.35.0.
Please visit https://docs.trychroma.com/troubleshooting#sqlite to learn how to upgrade."

System Info

langchain==0.2.15
langchain-chroma==0.1.3
langchain-community==0.2.14
langchain-core==0.2.36
langchain-openai==0.1.22
langchain-text-splitters==0.2.2

Windows/ Linux
python 3

[dosubot[bot]]

To resolve the "unsupported version of sqlite3" error when using Chroma as a vector database in LangChain, follow these steps:

1. Install the sqlite-vss Python package:

   pip install sqlite-vss

2. Ensure your SQLite version supports the VSS extension:

   • Verify your SQLite version is >= 3.35.0. You can check your SQLite version with:

     sqlite3 --version

   • If your SQLite version is outdated, you can upgrade it. For Windows, you can download the latest version from the SQLite website. For Linux, you can use your package manager or compile from source.

3. Ensure compatibility with Chroma:

   • For versions of Chroma prior to 0.4.0, the implementation uses duckdb+parquet for persistence. Ensure your code handles this correctly:

     import chromadb
     major, minor, _ = chromadb.__version__.split(".")
     if int(major) == 0 and int(minor) < 4:
         client_settings.chroma_db_impl = "duckdb+parquet"

4. Update your environment variables for Azure OpenAI embeddings:

   • Ensure you have set the necessary environment variables:

     import os
     
     os.environ["OPENAI_API_TYPE"] = "azure"
     os.environ["OPENAI_API_BASE"] = "https://<your-endpoint.openai.azure.com/"
     os.environ["OPENAI_API_KEY"] = "your AzureOpenAI key"
     os.environ["OPENAI_API_VERSION"] = "2024-02-01"
     os.environ["AZURE_OPENAI_ENDPOINT"] = "https://<your-endpoint.openai.azure.com/"
     os.environ["AZURE_OPENAI_API_KEY"] = "your AzureOpenAI key"
     os.environ["AZURE_OPENAI_AD_TOKEN"] = "your Azure AD token"
     os.environ["OPENAI_PROXY"] = "http://your-corporate-proxy:8080"

By following these steps, you should be able to resolve the "unsupported version of sqlite3" error and successfully use Chroma as your vector database in LangChain ^[1][2].

To continue talking to Dosu, mention @dosu.

[toshnav]

@dosu I am still encountering the same issue. I am working in a Jupyter notebook, and when I attempt to run sqlite3 --version, I receive an 'undefined' error. I am unsure how to resolve this issue.

[toshnav]

Can any one please help me. "---------------------------------------------------------------------------
RuntimeError Traceback (most recent call last)
Cell In[5], line 4
2 from langchain_text_splitters import RecursiveCharacterTextSplitter
3 from langchain_community.document_loaders import PyPDFLoader
----> 4 from langchain_chroma import Chroma

File ~/.local/lib/python3.11/site-packages/langchain_chroma/init.py:6
1 """This is the langchain_chroma package.
2
3 It contains the Chroma class for handling various tasks.
4 """
----> 6 from langchain_chroma.vectorstores import Chroma
8 all = [
9 "Chroma",
10 ]

File ~/.local/lib/python3.11/site-packages/langchain_chroma/vectorstores.py:24
10 import uuid
11 from typing import (
12 TYPE_CHECKING,
13 Any,
(...)
21 Union,
22 )
---> 24 import chromadb
25 import chromadb.config
26 import numpy as np

File ~/.local/lib/python3.11/site-packages/chromadb/init.py:86
84 sys.modules["sqlite3"] = sys.modules.pop("pysqlite3")
85 else:
---> 86 raise RuntimeError(
87 "\033[91mYour system has an unsupported version of sqlite3. Chroma
88 requires sqlite3 >= 3.35.0.\033[0m\n"
89 "\033[94mPlease visit
90 https://docs.trychroma.com/troubleshooting#sqlite to learn how
91 to upgrade.\033[0m"
92 )
95 def configure(**kwargs) -> None: # type: ignore
96 """Override Chroma's default settings, environment variables or .env files"""

RuntimeError: Your system has an unsupported version of sqlite3. Chroma requires sqlite3 >= 3.35.0."

[toshnav]