langgraph_adaptive_rag_cohere.ipynb: JSONDecodeError: Expecting value: line 1 column 1 (char 0)

[kolyan288]

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

### Build Index

from langchain.text_splitter import RecursiveCharacterTextSplitter
from langchain_cohere import CohereEmbeddings
from langchain_community.document_loaders import WebBaseLoader
from langchain_community.vectorstores import Chroma

# Set embeddings
embd = CohereEmbeddings()

# Docs to index
urls = [
    "https://lilianweng.github.io/posts/2023-06-23-agent/",
    "https://lilianweng.github.io/posts/2023-03-15-prompt-engineering/",
    "https://lilianweng.github.io/posts/2023-10-25-adv-attack-llm/",
]

# Load
docs = [WebBaseLoader(url).load() for url in urls]
docs_list = [item for sublist in docs for item in sublist]

# Split
text_splitter = RecursiveCharacterTextSplitter.from_tiktoken_encoder(
    chunk_size=512, chunk_overlap=0
)
doc_splits = text_splitter.split_documents(docs_list)

# Add to vectorstore
vectorstore = Chroma.from_documents(
    documents=doc_splits,
    embedding=embd,
)

retriever = vectorstore.as_retriever()

Description

I tried to run the code from the RAG LangGraph examples, but it throws an exception, although I did not change it in any way. How to fix the error?

System Info

langchain==0.2.1
langchain-chroma==0.1.1
langchain-cli==0.0.24
langchain-cohere==0.1.8
langchain-community==0.2.1
langchain-core==0.2.3
langchain-experimental==0.0.59
langchain-huggingface==0.0.1
langchain-openai==0.1.8
langchain-text-splitters==0.2.0

platform: linux mint 21.1

python version: 3.10.14

[dosubot[bot]]

Hey there, @kolyan288! I'm here to help you with your bug, answer questions, and guide you to become a contributor. Let's tackle this together while waiting for a human maintainer.

To fix the JSONDecodeError: Expecting value: line 1 column 1 (char 0) error in the langgraph_adaptive_rag_cohere.ipynb notebook, you should ensure that the JSON string being parsed is correctly formatted and not empty. This error typically occurs when the JSON string is empty or malformed.

You can use the parse_partial_json function from the libs/core/langchain_core/utils/json.py file to handle potentially malformed JSON strings. This function attempts to parse a JSON string that may be missing closing braces or have other issues.

Here is an example of how you can use the parse_partial_json function to handle the JSON parsing in your notebook:

from langchain_core.utils.json import parse_partial_json

# Assuming `json_string` is the string you are trying to parse
json_string = '...'  # Replace with your actual JSON string

try:
    parsed_json = parse_partial_json(json_string)
    if parsed_json is None:
        raise ValueError("Failed to parse JSON string.")
    # Proceed with using `parsed_json`
except json.JSONDecodeError as e:
    print(f"JSON decoding failed: {e}")
    # Handle the error, e.g., log it or raise an exception

Make sure to replace '...' with the actual JSON string you are working with. This approach will help you handle cases where the JSON string might be incomplete or malformed, thus avoiding the JSONDecodeError.

Additionally, ensure that the documents are correctly formatted and that the from_documents method is used properly. Here is an example of how to use the from_documents method from the Chroma class:

from langchain_community.vectorstores.chroma import Chroma
from langchain.text_splitter import RecursiveCharacterTextSplitter
from langchain_core.documents import Document

# Load and split documents
loader = WebBaseLoader("https://lilianweng.github.io/posts/2023-06-23-agent/")
data = loader.load()

text_splitter = RecursiveCharacterTextSplitter(chunk_size=500, chunk_overlap=0)
all_splits = text_splitter.split_documents(data)

# Create a Chroma vectorstore from the split documents
vectorstore = Chroma.from_documents(
    documents=all_splits,
    collection_name="rag-private",
    embedding=embd,
)
retriever = vectorstore.as_retriever()

Ensure that the all_splits list contains properly formatted Document objects and that all required parameters are provided to the from_documents method. This will help avoid errors when adding documents to the vector store ^[1][2][3].

To continue talking to Dosu, mention @dosu.

[kolyan288]

Unfortunately, your recommendation didn't help at all.

[rootint]

Hi, I am also from Russia and I had the same problem. The issue was that I had a Russian mail attached to the cohere account and my API key was not working no matter the environment. I solved this issue by creating a new account in cohere by signing in with google with my gmail attached to it. Now it works, but only with a VPN or on Google Colab.