Getting error while using AzureOpenAIEmbeddings

[deepak-hl]

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

embedding = AzureOpenAIEmbeddings(
                    openai_api_version="2024-04-01-preview",
                    openai_api_type='azure',
                    api_key = os.getenv("AZURE_OPENAI_API_KEY"), 
                    azure_endpoint = os.getenv("AZURE_OPENAI_ENDPOINT"),
                    azure_deployment=DEPLOYMENT_NAME,
                    validate_base_url = True)

below is my .env file
DEPLOYMENT_NAME="exxxxxxxxxxxxxxxxx"
AZURE_OPENAI_API_KEY="axxxxxxxxxxxxxxxx49"
AZURE_OPENAI_ENDPOINT="https://xxxxxxxxxx.openai.azure.com/ "
AZURE_OPENAI_API_VERSION="2024-05-01-preview"

Description

Below is the error i am getting :
Traceback (most recent call last):
File "/home/deepak/Summarizer-bot/langchain_qa.py", line 99, in
print(generate_embeddings("Code of Conduct Policy.pdf"))
File "/home/deepak/Summarizer-bot/langchain_qa.py", line 65, in generate_embeddings
embedding = AzureOpenAIEmbeddings(
File "/home/deepak/Summarizer-bot/envs/lib/python3.10/site-packages/pydantic/main.py", line 212, in init
validated_self = self.pydantic_validator.validate_python(data, self_instance=self)
pydantic_core._pydantic_core.ValidationError: 1 validation error for AzureOpenAIEmbeddings
Value error, As of openai>=1.0.0, Azure endpoints should be specified via the azure_endpoint param not openai_api_base (or alias base_url). [type=value_error, input_value={'openai_api_version': '2...rue, 'model_kwargs': {}}, input_type=dict]
For further information visit https://errors.pydantic.dev/2.9/v/value_error

System Info

langchain==0.3.0
langchain-community==0.3.0
langchain-core==0.3.5
langchain-openai==0.2.0
langchain-text-splitters==0.3.0
langchainhub==0.1.21
ubuntu
openai==1.47.0

[dosubot[bot]]

I found a similar solved discussion that might help you resolve the issue ^[1]. The ValidationError you're encountering is likely due to the validation logic in the validate_environment method of the AzureOpenAIEmbeddings class. Here are some key points to ensure correct initialization:

1. Ensure azure_endpoint includes /openai:

   azure_endpoint = "https://<your-endpoint>.openai.azure.com/openai"

2. Do not specify both deployment_name and openai_api_base:

   • Use azure_endpoint and azure_deployment together.

Here's an example of how to correctly instantiate the AzureOpenAIEmbeddings class:

from langchain_openai import AzureOpenAIEmbeddings

embedding = AzureOpenAIEmbeddings(
    openai_api_version="2024-04-01-preview",
    openai_api_type='azure',
    api_key=os.getenv("AZURE_OPENAI_API_KEY"), 
    azure_endpoint=os.getenv("AZURE_OPENAI_ENDPOINT").rstrip("/") + "/openai",
    azure_deployment=os.getenv("DEPLOYMENT_NAME"),
    validate_base_url=True
)

Ensure that your .env file is correctly set up and that the azure_endpoint includes the /openai path to avoid validation errors. This should resolve the issue you're facing ^[1][2][3].

To continue talking to Dosu, mention @dosu.

[dosubot[bot]]

The error persists because the AzureOpenAIEmbeddings class enforces that Azure endpoints should be specified via the azure_endpoint parameter and not openai_api_base. Ensure that you are not using openai_api_base in your code or environment variables.

Here is the corrected code snippet:

from langchain_openai import AzureOpenAIEmbeddings
import os

embedding = AzureOpenAIEmbeddings(
    openai_api_version="2024-04-01-preview",
    openai_api_type='azure',
    openai_api_key=os.getenv("AZURE_OPENAI_API_KEY"), 
    azure_endpoint=os.getenv("AZURE_OPENAI_ENDPOINT").rstrip("/") + "/openai",
    deployment=os.getenv("DEPLOYMENT_NAME"),
    validate_base_url=True
)

Ensure your .env file is correctly set up:

DEPLOYMENT_NAME="exxxxxxxxxxxxxxxxx"
AZURE_OPENAI_API_KEY="axxxxxxxxxxxxxxxx49"
AZURE_OPENAI_ENDPOINT="https://xxxxxxxxxx.openai.azure.com"
AZURE_OPENAI_API_VERSION="2024-05-01-preview"

Make sure the azure_endpoint includes the /openai path and that you are not using openai_api_base anywhere in your configuration ^[1][2].

To continue talking to Dosu, mention @dosu.

[deepak-hl]

@dosu
not using openai_api_base anywhere but still getting same error as above

[dosubot[bot]]

The validation error for AzureOpenAIEmbeddings is likely occurring because the openai_api_type is set to "azure", "azure_ad", or "azuread", which triggers the validation check in the validate_environment method. Ensure that you are using the correct parameters and environment variables.

Here is the corrected code snippet:

from langchain_openai import AzureOpenAIEmbeddings
import os

embedding = AzureOpenAIEmbeddings(
    openai_api_version="2024-04-01-preview",
    openai_api_type='azure',
    api_key=os.getenv("AZURE_OPENAI_API_KEY"), 
    azure_endpoint=os.getenv("AZURE_OPENAI_ENDPOINT").rstrip("/") + "/openai",
    azure_deployment=os.getenv("DEPLOYMENT_NAME"),
    validate_base_url=True
)

Ensure your .env file is correctly set up:

DEPLOYMENT_NAME="exxxxxxxxxxxxxxxxx"
AZURE_OPENAI_API_KEY="axxxxxxxxxxxxxxxx49"
AZURE_OPENAI_ENDPOINT="https://xxxxxxxxxx.openai.azure.com"
AZURE_OPENAI_API_VERSION="2024-05-01-preview"

Make sure the azure_endpoint includes the /openai path and that you are not using openai_api_base anywhere in your configuration ^[1][2][3][4].

To continue talking to Dosu, mention @dosu.

[deepak-hl]

@dosu now getting below error:
Error code: 404 - {'error': {'code': 'DeploymentNotFound', 'message': 'The API deployment for this resource does not exist. If you created the deployment within the last 5 minutes, please wait a moment and try again.'}}
True

[dosubot[bot]]

To resolve the "DeploymentNotFound" error, ensure that the deployment parameter matches the deployment name configured in your Azure OpenAI resource. Here is an example of how to set the deployment parameter correctly:

from langchain_openai import AzureOpenAIEmbeddings
import os

# Initialize the AzureOpenAIEmbeddings with the correct deployment name
embedding = AzureOpenAIEmbeddings(
    openai_api_version="2024-04-01-preview",
    openai_api_type='azure',
    api_key=os.getenv("AZURE_OPENAI_API_KEY"), 
    azure_endpoint=os.getenv("AZURE_OPENAI_ENDPOINT").rstrip("/") + "/openai",
    deployment=os.getenv("DEPLOYMENT_NAME"),  # Ensure this matches your actual deployment name
    validate_base_url=True
)

Make sure that the DEPLOYMENT_NAME in your .env file matches exactly with the deployment name configured in your Azure OpenAI resource. Additionally, ensure that the azure_endpoint and api_key are correctly set.

If you prefer, you can also set these values using environment variables:

export AZURE_OPENAI_API_KEY="your-api-key"
export AZURE_OPENAI_ENDPOINT="https://example-resource.azure.openai.com/"
export AZURE_OPENAI_API_VERSION="2024-04-01-preview"
export DEPLOYMENT_NAME="your-deployment-name"

Then, you can initialize the AzureOpenAIEmbeddings without explicitly passing these parameters:

from langchain_openai import AzureOpenAIEmbeddings

# Initialize the AzureOpenAIEmbeddings using environment variables
embedding = AzureOpenAIEmbeddings(
    deployment=os.getenv("DEPLOYMENT_NAME")  # Ensure this matches your actual deployment name
)

Ensure that the environment variables are correctly set and that the deployment name matches the one configured in your Azure OpenAI resource ^[1][2][3].

To continue talking to Dosu, mention @dosu.