Trouble Loading "01-Basic-loading-the-kernel"

[0v3r-9000]

ServiceInitializationError: chat_deployment_name is required.

Hi,
I'm new to this and been getting a lot of errors along the way to trying to run sk locally. I've downloaded most updates successfully but keep getting the error thats up top after running this:

---------------- Code below

Remove all services so that this cell can be re-run without restarting the kernel

kernel.remove_all_services()

service_id = None
if selectedService == Service.OpenAI:
from semantic_kernel.connectors.ai.open_ai import OpenAIChatCompletion

service_id = "default"
kernel.add_service(
    OpenAIChatCompletion(
        service_id=service_id,
    ),
)

elif selectedService == Service.AzureOpenAI:
from semantic_kernel.connectors.ai.open_ai import AzureChatCompletion

service_id = "default"
kernel.add_service(
    AzureChatCompletion(
        service_id=service_id,
    ),
)

---

Any help is appreciated.

Ultimate goal is to follow the DeepLearning.ai activities locally versus in their notebook.

[moonbox3]

Hi @0v3r-9000, sorry to hear you're hitting issues. Let's help get you unblocked.

In the getting_started notebooks (and in that particular notebook you referenced) there is a section called Configuring the Kernel. There is also a file called CONFIGURING_THE_KERNEL.md. Have you taken a look at this? https://github.com/microsoft/semantic-kernel/blob/main/python/samples/getting_started/CONFIGURING_THE_KERNEL.md#configuring-the-kernel

Per the text:

Let's get started with the necessary configuration to run Semantic Kernel. For Notebooks, we require a .env file with the proper settings for the model you use. Create a new file named .env and place it in this directory. Copy the contents of the .env.example file from this directory and paste it into the .env file that you just created.

NOTE: Please make sure to include GLOBAL_LLM_SERVICE set to either OpenAI, AzureOpenAI, or HuggingFace in your .env file. If this setting is not included, the Service will default to AzureOpenAI.

You need to make sure you have a .env file in the same directory as the notebook(s). If you're using AzureOpenAI, you'll need to include, in the .env file:

GLOBAL_LLM_SERVICE="AzureOpenAI"
AZURE_OPENAI_CHAT_DEPLOYMENT_NAME="" # similar to "gpt-4o" or "gpt-4o-mini" whatever your deployment name is in the portal
AZURE_OPENAI_ENDPOINT="" # similar to: "https://***.openai.azure.com/"
AZURE_OPENAI_API_KEY="" # your api key, if using api auth

Do you have a .env file in the notebooks directory? If so, do you have AZURE_OPENAI_CHAT_DEPLOYMENT_NAME configured?

[0v3r-9000]

Thank you for the response! I will do this and keep you updated. I was trying to use the example.env file with an openai key. i suppose i can move that into said directory?

[moonbox3]

The .env.example is only showing what values should exist in the .env file. You’ll want to make sure you only use a .env file (remove the example at the end). And yes, please place that in the notebook directory. Let us know.

[0v3r-9000]

not sure if you have seen this before, but this is the last error i'm getting from this code:

kernel.config.add_text_completion_service(
"openai-gpt-3.5", OpenAITextCompletion("gpt-3.5-turbo", api_key)
)

gives error:

AttributeError Traceback (most recent call last)
Cell In[22], line 2
1 # Set up the text completion service
----> 2 kernel.config.add_text_completion_service(
3 "openai-gpt-3.5", OpenAITextCompletion("gpt-3.5-turbo", api_key)
4 )

File /Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/pydantic/main.py:892, in BaseModel.getattr(self, item)
889 return super().getattribute(item) # Raises AttributeError if appropriate
890 else:
891 # this is the current error
--> 892 raise AttributeError(f'{type(self).name!r} object has no attribute {item!r}')

AttributeError: 'Kernel' object has no attribute 'config'

[moonbox3]

Do you have a screenshot you can share? I think you have a very old reference to code that no longer exists -- meaning inside your notebook. If I search our codebase, that add_text_completion_service no longer exists (I remember we removed it well before May of 2024):