Validation error for FastEmbedEmbeddings - extra fields not permitted

[Fernando7181]

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_community.vectorstores import Chroma
from langchain_community.document_loaders import PyPDFLoader
from langchain_community.chat_models import ChatOllama
from langchain_community.embeddings.fastembed import FastEmbedEmbeddings 
from langchain.text_splitter import RecursiveCharacterTextSplitter
from langchain.prompts import PromptTemplate
from langchain.chains.combine_documents import create_stuff_documents_chain
from langchain.chains import create_retrieval_chain
from huggingface_hub import login
from pydantic import BaseModel
from langchain_ollama import ChatOllama
import os
import sys
import chromadb

def ingest():
    loader = PyPDFLoader("assets/2409.12122v1.pdf")
    pages = loader.load_and_split()
    text_splitter = RecursiveCharacterTextSplitter(
        chunk_size=1024,
        chunk_overlap=100,
        length_function=len,
        add_start_index=True,
    )
    chunks = text_splitter.split_documents(pages)
    print(f"Split {len(pages)} pages into {len(chunks)} chunks.")

    embedding = FastEmbedEmbeddings()

    Chroma.from_documents(documents=chunks, embeddings=embedding, persist_directory="./sql_chroma_db")
ingest()

Description

for some reason I can ingest() my code for my vector dabase and im not sure what to do

System Info

[feijoes]

@Fernando7181, the issue seems to be with the parameter name embeddings in your code. The correct parameter name is embedding (singular), not embeddings.

Here's the corrected version of your code:

from langchain.vectorstores import Chroma

# Correct the parameter name from 'embeddings' to 'embedding'
vector_store = Chroma.from_documents(
    documents=chunks, 
    embedding=embedding, 
    persist_directory="./sql_chroma_db"
)

docs

[Fernando7181]

@Fernando7181, the issue seems to be with the parameter name embeddings in your code. The correct parameter name is embedding (singular), not embeddings.

Here's the corrected version of your code:

from langchain.vectorstores import Chroma

# Correct the parameter name from 'embeddings' to 'embedding'
vector_store = Chroma.from_documents(
    documents=chunks, 
    embedding=embedding, 
    persist_directory="./sql_chroma_db"
)

docs

that did helped me a lot and and fixed my issue, thank you but by any other means you know why this other one would be happening? I looked for it but quite didnt see anyone talking about it. Vlw msm man ajudou bastante kkk

here's the reference code for it
`ccess_token_read = "your hugginface token"
access_token_write = "your hugginface token"
login(token = access_token_read)

def rag_chain():
model = ChatOllama("llama3", 1024)

prompt = PromptTemplate.from_template(
    """
    <s> [Instructions] You are a friendly assistant. Answer the question based only on the following context. 
    If you don't know the answer, then reply, No Context availabel for this question {input}. [/Instructions] </s> 
    [Instructions] Question: {input} 
    Context: {context} 
    Answer: [/Instructions]
    """
)

embedding = FastEmbedEmbeddings()
vector_store = Chroma(
    persist_directory="./sql_chroma_db", 
    embedding_function=embedding)

retriever = vector_store.as_retriever(
    search_type="similarity_score_threshold",
    search_kwargs={
        "k": 3,
        "score_threshold": 0.5,
    },
)

document_chain = create_stuff_documents_chain(model, prompt)
chain = create_retrieval_chain(retriever, document_chain)

return chain

def ask(query: str):

chain = rag_chain()

result = chain.invoke({"input": query})
# print results
print(result["answer"])
for doc in result["context"]:
    print("Source: ", doc.metadata["source"])`

[feijoes]

@Fernando7181 In the line ChatOllama("llama3", 1024) idk what the 1024 is for but i recommend always pass the arguments ass keyword arguments

ChatOllama(
    model="llama3.1"
    ...

[Fernando7181]

@Fernando7181 In the line ChatOllama("llama3", 1024) idk what the 1024 is for but i recommend always pass the arguments ass keyword arguments

ChatOllama(
    model="llama3.1"
    ...

Hey Feijoes, just tryied and the 1024 is the chunk size for the llm, i have removed that in the final code. Will be adding to a repository with the final adjustments so you can look at it and see if it's working but that really helped <3