After training a yolov5 model with automl and registering it to my workspace I get the following error when trying to load the model: "TypeError: __init__() got an unexpected keyword argument 'required'"
I have set up AutoML for yolov5 and let it run for 3 trials. Then I got the best run as shown in the documentation and registered it to my workspace with the following code: local_dir = "./artifact_downloads" if not os.path.exists(local_dir): …
Cannot update the endpoint traffic and logs are also not showing
I am using Azure Machine Learning for deploying models and when I try to update the traffic for the new deployment it shows the below error It's happening for all my endpoints for the past 5 hours, not sure what to do.
Why am I getting this error when deploying a Prompt Flow endpoint?
Getting a weird error when deploying a Prompt Flow endpoint. This is the error: The request is invalid. Status: 400 (The request is invalid.) ErrorCode: BadRequest Content: { "error": { "code": "BadRequest",…
Azure ML Prompt Flow Deployment - Uvicorn Error
I've not had this issue until today. Any prompt flow deployments are failing with a missing uvicorn package. This is not a manual deployment; it's deployed directly from the prompt flow tab, and I don't have access to the requirements.txt to add this…
No web service input and void data schema during subsequent endpoint generation.
I am trying to create real-time inference pipeline in azure ml , following all straight forward steps, but I am unable to generate the web service input . only web service output is generated. Further, if I manually add web service input or add data…
Validation error during signup
Hi everyone, I'm trying to sign up for Azure Machine Learning using my Azure for Students Starter subscription, but I'm encountering an error during the validation process. The specific error message I receive is: "Validation could not be completed…
Error while creating a managed online endpoint in azure machine learning studio
while creating a managed online endpoint in azure machine learning studio I get the following error in my workspace's notebook: --------------------------------------------------------------------------- OperationFailed Traceback (most recent call last)…
How to add hidden secrets as user input
I am currently using a Jira API to retrieve jira issues in a prompt flow python node, but the user should only be able to retrieve jira issues which the user has access to, to achieve this the user should provide the username and password, but if I set…
Opaque Errors Upon Creation of Feature Set, Despite Following Documentation Precisely
I'm attempting to create a Feature Set Specification, using MLClient.feature_sets.begin_create_or_update. I'm passing an azure.ai.ml.entities.FeatureSet instance, configured as follows: fs = FeatureSet( name=feature_set_name, …
I'm being charged for classic ML Studio workspaces that I'm unable to delete
I have 3 old ML Studio workspaces that I'm not using. I have been unable to delete them but I'm being charged for them every month (I'm guessing I tried to delete them incorrectly in the past and left fragments behind). I currently have no other…
Machine Learning Workspace Notebook and Environment issues
Question. I'm trying to learn to train a vision model and azure machine learning workspace notebooks. I am trying to create an environment where i can run both Azure AI SK2 and pytorch to train a vision model and have access to data assets in both the…
Does the AML Python SDK v2 support running Sweep Jobs on Local Compute ?
I'm trying to run a sweep job for hyperparameter tuning using Azure Machine Learning SDK v2 on my local compute. While I can successfully run command jobs locally and sweep jobs on serverless compute, I encounter AttributeError: 'SweepJob' object has no…
Unable to create managed online endpoint in using azureml python sdk v2
Hi All, My first post here. I am trying to follow the ms-learn tutorials to deploy a model using a managed online endpoint using Azure ml sdk v2. import uuid # Create a unique name for the endpoint online_endpoint_name = "credit-endpoint-" +…
How to fix Score Model (AFx Library) : Error 10: Argument 1 Array must not be empty
When I test preview, I have this error: Score Model (AFx Library) : Error 10: Argument 1 Array must not be empty. So What is empty in Score Model and how to fix it? Error Message: Score Model (AFx Library) : Error 10: Argument 1 Array must not be…
Failed to execute command group with error An unexpected error occurred while executing command due to: Failed to pull Docker image `mcr.microsoft.com/azureml/runtime/cap/hosttools-capability/installed:westeurope-stable` due to: None
Failed to execute command group with error An unexpected error occurred while executing command due to: Failed to pull Docker image mcr.microsoft.com/azureml/runtime/cap/hosttools-capability/installed:westeurope-stable due to: None I get this error in…
How to get more metrics from AzureML Wide&Deep Recommender
Hello! I was wondering if there is a way to get more output metrics for the wide and deep recommender in azure ml v1? It only gives ndcg and I don't see a way to get other metrics. Appreciate the help in advance!
How to update the SKU of the Azure container registry that is automatically created when creating an Azure Machine learning workspace
I am using Azure Mahcine Learning product and I have confirmed that Azure container registry resource group is automatically created when workspace is activated. However, I am not using ACR specifically at the moment, so I want to change SKU to Basic…
I have a new compute instance that has Python3.8 and Python3.9 and in Azure Notebook some libraries not found.
I created a new compute based on an earlier one from Dec 2023. Dec 23 compute has Python3.8 New compute has Python3.8 and Python3.9. Azure notebook uses Python 3.9.19 From command line "import sklearn" works for 3.8 and 3.9 Azure notebook…
How can I get MLClient object in the azureml job runtime
I try to get azureml MLClient object in the job created on azureml python sdk v2. I use azure.identity.DefaultAzureCredentials as creds. I am able to get "Workspace" using "Run" class from sdk v1. I try to get ml client like…
Private Azure Machine Learning Workspace getting forbidden to reach private Storage Account for SAS
My Azure Machine learning workspace is configured with private endpoint , my aml is configured to a Storage account which is also configured private endpoint and disabled the public access, My Storage Account network will looks like below The…