Is it possible to use open source embedding methods rather than OpenAIEmbeddings?

[ykemiche]

System Info

LangChain version : 0.0.216
Python version : 3.11.4
System: Windows

Who can help?

No response

Information

• The official example notebooks/scripts
• My own modified scripts

Related Components

• LLMs/Chat Models
• Embedding Models
• Prompts / Prompt Templates / Prompt Selectors
• Output Parsers
• Document Loaders
• Vector Stores / Retrievers
• Memory
• Agents / Agent Executors
• Tools / Toolkits
• Chains
• Callbacks/Tracing
• Async

Reproduction

I want to create a chatbot to retrieve information from my own pdf in response to a query using google PaLM model, I followed these steps :
-load the pdf
-split it using RecursiveCharacterTextSplitter
-store its embeddings in a Chroma vectorestore
and then create a chain ...

from langchain.document_loaders import PyPDFLoader
from langchain.document_loaders import PyPDFLoader
from langchain.text_splitter import RecursiveCharacterTextSplitter
from langchain.vectorstores import Chroma
from langchain.embeddings.openai import OpenAIEmbeddings

import langchain

loader=PyPDFLoader("path/to/pdf.pdf")
chroma_dir="./chroma
pages=loader.load()
splitter=RecursiveCharacterTextSplitter(
    chunk_size=1000,
    chunk_overlap=150,
    separators=['\n\n','\n'," ",""]
)

splits=splitter.split_documents(pages)

#I want to change this with another embedding method which doesn't require API authentification
embeddings=OpenAIEmbeddings()

vector_db=Chroma.from_documents(
documents=splits,
embedding=embeddings,
persist_directory=chroma_dir
)

But the only embedding method that is available in the LangChain documentation is OpenAIEmbeddings,how can we do without it?

Expected behavior

all the splits embeddings stored in Chroma vectorestore without using OpenAIEmbeddings()

[mswoff]

If you want Google PALM

from langchain.embeddings.google_palm import GooglePalmEmbeddings
...
embeddings = GooglePalmEmbeddings()

You can alternatively use VertexAIEmbeddings as an alternative to OpenAIEmbeddings. This method does not require API authentication. Here is how you can modify your code:

from langchain.embeddings.vertexai import VertexAIEmbeddings

# Replace OpenAIEmbeddings with VertexAIEmbeddings
embeddings = VertexAIEmbeddings()

vector_db = Chroma.from_documents(
    documents=splits,
    embedding=embeddings,
    persist_directory=chroma_dir
)

[Fofna]

It requieres GOOGLE_API_KEY