on 06-23-2023 11:17 AM
I'm doing a relatively new SAP Mission on AI Launchpad and AI Core.
I currently want to run a ipynp file with the following code:
response = ai_core_client.configuration.create(<br> name = "azure-proxy-serve",<br> scenario_id = "azure-openai-proxy",<br> executable_id = "azure-openai-proxy",<br> input_artifact_bindings = [],<br> parameter_bindings = [<br> ParameterBinding(key = "OPENAI_API_BASE", value = OPENAI_API_BASE),<br> ParameterBinding(key = "OPENAI_API_KEY", value = OPENAI_API_KEY), <br> ParameterBinding(key = "DOCKER_NAMESPACE", value = DOCKER_NAMESPACE)<br> ],<br> resource_group = resource_group_id<br>)<br><br>serve_config_resp = response<br>print(response.__dict__)
The part of the file the code is reading from:
metadata:<br> name: azure-openai-proxy # executable ID, must be unique across your SAP AI Core instance, for example use `server-pipeline-yourname-1234`azure-openai-proxy_kuhnsa1234<br> annotations: <br> scenarios.ai.sap.com/description: "Azure OpenAI Proxy"<br> scenarios.ai.sap.com/name: "Azure OpenAI Proxy" # Scenario name should be the use case<br> executables.ai.sap.com/description: "Azure OpenAI Completions Proxy as Inference"<br> executables.ai.sap.com/name: "azure-openai-proxy" # Executable name should describe the workflow in the use case<br> labels:<br> scenarios.ai.sap.com/id: "azure-openai-proxy"<br> ai.sap.com/version: "1.0"<br> ext.ai.sap.com/islm_released_version: "true"<br> ext.ai.sap.com/islm_executable_type: "inference"<br> ext.ai.sap.com/islm_inference_type: "online"
But when I run it I get an AIAPI Exception with the following message :
AIAPINotFoundException: Failed to post configurations: Not Found, Could not create configuration because executable azure-openai-proxy for scenario azure-openai-proxy wasn't found.. Status Code: 404, Request ID:9686e691-5c09-4359-9e8b-0239d5638fe4
do I have to create an executable manually first or why do I get this error?
the tutor is the following https://github.com/SAP-samples/smart-co2nverter-ai
Thanks for the help 🙂
Hi Sarah,
Please use the snippet available in the step below to check if the scenario is synced from your file.
https://developers.sap.com/tutorials/ai-core-helloworld.html#0045ff2a-9cf1-4893-95f0-f73b561a4dad
It would be great if you could also share, how much the link helped you debug, so that we suggest improvements in mentioned notebook file.
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
screenshot2.pngI tried to code snippets you provided and as far as I see the gitHub is onboarded and connected wit the AI Launchpad, or at least it pops up there. I also created an application and entered the path to the GitHub.
My Problem is, that there is still no scenario and I cannot configure to serve my AI Core model
Here are some screenshoots:
I will add some in an additional comment as there is a limit on how many png files to attach
You can also use the below snippet to get the application' health/ sync message
response = ai_core_client.applications.get_status("azure-openai-aicore-cap-api") # repo name
print(response.message)
# If response is `rpc error: code = Unknown desc = authorization failed`
# then you Git PAT(used for onboarding Git is incorrect) # to resolve offboard GitHub and onboard again.User | Count |
---|---|
74 | |
8 | |
8 | |
7 | |
7 | |
6 | |
5 | |
5 | |
5 | |
4 |
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.