Model Performance Signal (Preview) | Model Monitor - Data Joiner | error:" join column 'correlationid is not present in left_input data.
Greetings, I am working on my MLOps project and have been trying to use the Azure ML "model performance" monitoring function to detect the model's degradation. Left_input_data is the deployed model's inference output data, which were…
(new)AML Continuous Monitoring | Mapping Ground Truth Data for Model Performance Monitoring in Azure Machine Learning
This is the new post as my first post couldn't be viewed( shown 404 error after I clicked it) Greetings, I am reading the document regarding MLOps and would like to continuously monitor the ML model deployed inside Azure Machine Learning. The goal of…
What rank is used in LoRA fine-tuning?
I previously fine-tuned a few instances of gpt-4 using the standard Azure ML training flow. I am seeing online that Azure ML uses Low rank approximation in order to optimize fine tuning. I am wondering what rank is used in the LoRA fine-tuning process…
Where are auto-generated environments and scoring scripts for endpoints located?
When deploying a real-time or batch endpoint, you have the option to have both the environment and scoring script auto-generated. Where are these scoring scripts and environments stored for reference and troubleshooting, afterwards? They are not located…
Inquiry about Machine Learning Model Deployment Error in Azure for Students Account
I am contacting you because an error occurred at the "Deploy and test the model" stage while performing machine learning practice at the link below using an Azure for Students account. Link:…
What is the best input for video indexer
I am new to video indexer api, I want to leverage this api for one of my app development. My question is which format is working the best for this api to make the best result? Best result with the smaller size is the best
Move data from one component to the next in Azure Machine Learning
I have got 2 components in Azure Machine Learning. I have got 2 dataframes in the first component (called prep) which I want to pass into the next component (called middle) for further processing. In the prep code, I have tried to save the dataframe into…
How do I configure search space ranges in an Azure Machine Learning sweep job from input parameters?
I am using Azure Machine Learning pipeline SDK v2 and I'd like to do a sweep job while specifying the search space range from the inputs parameters. The example currently available currently does not cover this, and I've been having problems getting it…
I have available cores but I can't access it for my endpoint deployment
I have available cores for either Standard_NC6s_v3 or Standard_NC24ads_A100_v4 while creating computer instance. But when I try to deploy an endpoint with the choice of Standard_NC6s_v3, it does not appear in my list of choices. Please help me. Thank…
Error during batchpoint invoke: Access to this resource is denied. Please check your ACL rules on the resource
Hello, I am getting a "Access to this resource is denied. Please check your ACL rules on the resource." when attempting to invoke a batch endpoint. I have tried both with Azure CLI and Azure Data Factory to invoke a batch endpoint, with both…
Issue with the sandbox being unavailable on Microsoft Azure Fundamentals
Hey everyone, I’ve been trying to use the MS Learn Sandbox for a module, but I keep running into this annoying error message: “The sandbox for this module is currently unavailable. We're working to resolve this as quickly as possible. In the meantime,…
What is the latency and cost of A100 gpu
Hi team, I want to deploy fine-tuned model on Azure. If I use llama3.1-8b model, the input is 4k token and output is 512 token. What is the latency if I deploy the model to A100 gpu ? What is the economy cost for each A100 gpu ? Do we support vllm…
How to do A/B test on Azure
Hi Team, I want to deploy fine-tuned Qwen2.5B model on Azure, how do I do A/B test or online experiment for model version A and model version B ?
Feature request for Azure ML online endpoints
Hi, I use AML online endpoints a lot. The "test" panel is very helpful. However, the data in the test request gets zeroed out when I navigate off this page. I currently duplicate the browser tab and navigate to the "logs" panel so I…
How to deploy an LLM in a batch endpoint in Azure Machine Learning? Any alternatives?
Hi, I'm trying to deploy an LLM in Azure Machine Learning in a batch endpoint. I was successful in creating an online endpoint, but facing issues with batch endpoint. The goal of the project is to deploy an LLM that will run every morning and produce…
AZURE ML STUDIO (FACING PROBLEM IN DEPLOYMENT)
Not enough quota available for Standard_D2as_v4. Current usage/limit: 0/4. Additional requested: 8 (20% more quota than expected may be needed for redundancy purposes on some SKUs) I am facing a problem while deploying model in selecting Virtual machine…
Automl Model Deploy Fails On Gunicorn Version Conflict
Hello, I am attempting to deploy an automl model to a real time endpoint, when I do I am getting a pip error about dependency conflicts between gunicorn and mlflow. It seems these install are dynamically added by azure? What can I do to get past this…
Azure Machine Learning Studio creates v1 deployments that can't be tested in Studio
If you use AutoML in Azure Machine Learning Studio to train a model and then deploy that model to an Azure Container Instance, you can't test the deployment using Studio. When you click on the Test tab, it comes back with "This deployment is based…
Not able to subscribe LLama-3.3-70B-Instruct with Microsoft Azure Sponsorship Account
Hello Team, Greetings of the day!! I would like to inform you that one of my customers is not able to subscribe LLama-3.3-70B-Instruct with Microsoft Azure Sponsorship Account. He attempted to subscribe to the LLama-3.3-70B-Instruct via both the…
what is difference between AKS and machine learning compute
Azure offers managed compute resources, including Azure Kubernetes Service (AKS) or Azure Machine Learning Compute. I want to upload fine-tuned LLM like Qwen2.5 or llama3.1 to Azure and online serving with these fine-tuned models, what is difference…