Issue with SemanticSimilarityExampleSelector.from_examples

[NikhilKosare]

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

example_selector = SemanticSimilarityExampleSelector.from_examples(
    examples,
    embeddings,
    vector_store,
    k=5,
    input_keys=["input"],
)

Description

ServiceRequestError: Invalid URL "/indexes('langchain-index')?api-version=2023-10-01-Preview": No scheme supplied. Perhaps you meant https:///indexes('langchain-index')?api-version=2023-10-01-Preview?

System Info

langchain 0.1.8

platform - windows

python version - 3.11.5

[dosubot[bot]]

🤖

Hey @NikhilKosare, great to see you diving into another challenge! Hope you're doing well.

Based on the error message you provided, it seems like the URL for the vector store is not correctly formatted. The error message ServiceRequestError: Invalid URL "/indexes('langchain-index')?api-version=2023-10-01-Preview": No scheme supplied. Perhaps you meant https:///indexes('langchain-index')?api-version=2023-10-01-Preview? indicates that the URL scheme (http or https) is missing.

The vectorstore_cls.from_texts method, which is called within SemanticSimilarityExampleSelector.from_examples, expects a properly formatted URL for the vector store. This URL should be passed as a keyword argument when calling SemanticSimilarityExampleSelector.from_examples.

Here is an example of how to call the method with a URL:

example_selector = SemanticSimilarityExampleSelector.from_examples(
    examples,
    embeddings,
    vector_store,
    k=5,
    input_keys=["input"],
    vectorstore_cls_kwargs={"url": "https://your-vector-store-url"}
)

In this example, replace "https://your-vector-store-url" with the actual URL of your vector store.

Please ensure that the URL for your vector store is correctly formatted and includes the scheme (http or https). If you continue to experience issues, please provide more details about your vector store and how you are initializing it.

Sources

• libs/core/langchain_core/example_selectors/semantic_similarity.py
• libs/langchain/langchain/vectorstores/astradb.py

About Dosu

This response is meant to be useful and save you time. It is not meant to be a precise solution, but rather a starting point for your own research.

Help me be more useful! Please leave a 👍 if this is helpful and 👎 if it is irrelevant.

If you want to continue the conversation, start your reply with @dosu-bot.

[NikhilKosare]

@dosu-bot I am creating a vector store like this
vector_store = AzureSearch(
azure_search_endpoint=vector_store_address,
azure_search_key=vector_store_password,
index_name=index_name,
embedding_function=embeddings.embed_query,
)

example_selector = SemanticSimilarityExampleSelector.from_examples(
examples,
embeddings,
vector_store,
k=5,
input_keys=["input"],
vectorstore_cls_kwargs={"url": "https://ss-abc.search.windows.net"}
)

I have am using azure_search_endpoint as vectore store url and i am still getting the same error.

ServiceRequestError: Invalid URL "/indexes('langchain-index')?api-version=2023-10-01-Preview": No scheme supplied. Perhaps you meant https:///indexes('langchain-index')?api-version=2023-10-01-Preview?

[calls9-amirbraham]

@NikhilKosare Have you found a solution to this issue please ?

[dosubot]

It looks like the core issue you're encountering with the ServiceRequestError is related to how the Azure Search service URL is being handled within your setup. The error message suggests that the URL is not correctly formatted or passed to the Azure Search service, specifically indicating a missing scheme (http or https).

Since you've confirmed that you're using the azure_search_endpoint with the correct https:// prefix, the problem might not be with how you're providing the URL but rather with how it's being utilized internally by the AzureSearch vector store or the SemanticSimilarityExampleSelector.from_examples method.

Here are a few steps to troubleshoot and potentially resolve this issue:

1. Verify the URL Format: Double-check that the vector_store_address variable indeed contains the full URL, including https://. It should match the format https://ss-abc.search.windows.net.

2. Initialization of AzureSearch: Since you're initializing the AzureSearch vector store directly with its endpoint, ensure that the initialization parameters are correctly set. The azure_search_endpoint should be the full URL of your Azure Search service. You don't need to pass the URL again through vectorstore_cls_kwargs in the SemanticSimilarityExampleSelector.from_examples call, as this might be redundant or cause confusion.

3. Review AzureSearch Implementation: If the URL is correctly formatted and the issue persists, it might be helpful to review the implementation of the AzureSearch class, focusing on how it constructs and executes requests to the Azure Search service. There could be an issue in how the URL is handled or passed in the class methods.

4. Check for Updates or Documentation: Ensure that you're using the latest version of the libraries involved and check any available documentation for updates or changes in how the Azure Search service should be integrated. Sometimes, issues like these can be resolved by updating to a newer version of the library that contains bug fixes or improvements.

If after following these steps, the issue remains unresolved, it might be beneficial to reach out to the maintainers of the library or service for more direct support, as they might have more insight into the specific configurations and requirements of the Azure Search service integration.

Details

About Dosu
This response is meant to be useful and save you time. It is not meant to be a precise solution, but rather a starting point for your own research.

Help me be more useful! Please leave a 👍 if this is helpful and 👎 if it is irrelevant.

To continue the conversation, mention @dosu.