233 questions
0
votes
0
answers
18
views
Avoid scheduled pipeline to be triggered immediately after creation in azureml (Python SDK v2)
When I submit a pipeline with an associated schedule, it runs immediately after submission, instead of waiting until the scheduled time of the day, day of the month, etc. Is there a way to avoid this ...
- 507
0
votes
1
answer
60
views
How to retrieve AzureML schedules that have at least one run successfully completed using Azure Python SDK v2?
I would like to retrieve those schedules in AzureML workspace that have been triggered and have at least one complete run (e.g., a pipeline that has been triggered under that schedule) successfully ...
- 507
0
votes
1
answer
51
views
How to programmatically get experiments with metrics above threshold azureml sdk v2
I have a number of experiments completed in AzureML, each one consisting of one pipeline with multiple jobs, and some of the jobs (2 in my case, say job X and job Y) having associated logged metrics. ...
- 507
0
votes
0
answers
156
views
Unable to use serverless connection in a PromptFlow pipeline
I am running my prompt flow (version 1.18.0) code as a pipeline. However, it seems like that the connection that I created in prompt flow which is referred in flow.dag.yaml is not working through a ...
- 394
0
votes
0
answers
95
views
mltable will not load data asset from AzureML
I am struggling to read a data asset from AzureML into my local python session. My code works up until I use mltable.load() (disguising my inputs with <...>):
from azure.identity import ...
0
votes
1
answer
173
views
Azure ML - can't use the model path
The ultimate goal - to publish my model to production. After a lot of iterations and not explicit errors I decided to do line by line check. I'm using Azure ML Notebooks.
Everything is working until I ...
- 518
1
vote
1
answer
167
views
Getting ModuleNotFoundError: No module named 'mltable' error while executing command job in Azure ml SKD v2
I am trying to run a command job that'd read in a Data Asset and perform some pre processing for further ML tasks.
To read in the data in the .py file I used the code in the consume section as below:
...
- 11.6k
2
votes
1
answer
275
views
How to trigger an Azure Machine Learning Studio notebook from a Databricks notebook?
I have a notebook in Azure Machine Learning Studio Workspace, and I have also created a compute cluster in the workspace. Now, I want to trigger this notebook from a Databricks notebook.
I have ...
0
votes
1
answer
310
views
Passing environment to the yml configuration file for Azure Machine Learning component
I want to create Azure Machine Learning component with custom environment. I've created the Environment and uploaded it to my workspace.
from azure.ai.ml.entities import Environment
from azure.ai.ml....
- 41
1
vote
1
answer
226
views
Cancel current pipeline job "from within" in Azure ML sdk v2
I am porting a sdk v1 machine learning pipeline to SDK v2. We have a step which, under certain conditions, cancels the whole pipeline job (ie, the other steps won't run). Its code is like this:
from ...
- 659
0
votes
1
answer
199
views
Retrieving Generated Data From Azure ML
I am having a very difficult time trying to download a JSON file from a successful Azure ML pipeline job. In the pic below you can see the contents of the saved file. However using the Python SDK, ...
- 47
0
votes
1
answer
155
views
How to use predict_proba with AzureML batch endpoint within invoke method, whilst using URI folder URL as the data?
I have a AutoML generated binary classification model deployed to a batch endpoint. I can successfully invoke the model using the below code to output a file that contains a binary prediction (1|0).
# ...
- 29
0
votes
1
answer
893
views
Load Registered Component in Azure ML for Pipeline using Python sdk v2
I'm working in Azure Machine Learning Studio to create components that I will run together in a pipeline. In this basic example, I have a single python script and a single yml file that make up my ...
0
votes
0
answers
185
views
AzureML SDKV2 - ml_client job stream error - Unauthorized
I have a aml pipeline that uses azureml sdkv2. I submit the pipeline as:
def get_chained_credentials(client_id: str):
try:
managed_identity_creds = ManagedIdentityCredential(client_id=...
- 854
0
votes
1
answer
289
views
AzureML command job: 'Your file exceeds 100 MB.'
I am trying to submit a job to AzureML using Python SDK V2.
I am developing locally on VSCode.
As my repository grew over time, I now get the message when submitting the job:
Your file exceeds 100 MB....
- 25
0
votes
1
answer
473
views
Azure deployment error "TypeError: metaclass conflict" when importing tensorflow (1.13.1) in project code
Any help fixing the problem would be greatly appreciated.
I am trying to deploy an old CNN model (MRCNN), last deployed in December 2020, on Azure ML studio using the Python SDK V2. After a successful ...
- 2,086
0
votes
1
answer
256
views
Azure Machine Learning dataset creation hangs forever
I'm trying to create a Dataset from a datastore using Azure ML, however, the execution hangs forever and never finishes.
This is the code I'm running which I've adapted from the Msft documentation:
...
- 15.3k
0
votes
2
answers
915
views
Unable to read current version of delta table in azure ml studio using Data asset
I am trying to create data assest with ADLS gen 2, and read a delta table on adls gen folder something like this:
/
└── my-data
├── _delta_log
├── part-0000-xxx.parquet
└── part-0001-xxx....
- 1
0
votes
1
answer
689
views
Azure ML Pipeline: Specify input-path on compute
I got a python script which shall be executed as an Azure ML pipeline step. This script expects to find several file-sets in a certain tree structure e.g.
data/
├─ project_A/
│ ├─ data.csv
│ ├─ ...
- 50
0
votes
1
answer
443
views
How do I write a pandas data frame from an Azure ML Jupyter notebook to an Azure SQL database?
I'm working with some data in Azure ML, using Jupyter notebooks.
I've gathered the data from an Azure SQL database by setting the tables up as assets and then used Python to run some machine learning ...
- 11
1
vote
0
answers
212
views
getting mlflow exception in azure ml compute cluster if I am trying to get the latest version from production
I was trying to get a latest model from production using mlflow command,
version = client.get_latest_versions(model_name, stages=["Production"])[0].version
However this is working fine in ...
0
votes
0
answers
323
views
Azure ML: how to use remote location as code path?
I am using Azure ML python SDK v2 to submit the job. The code parameter at "command" is pointed to the local "./source" path.
train_func = command(
environment="AzureML-...
- 1
4
votes
1
answer
499
views
Azure ML: DeploymentIdentityError: Failed to create Kubernetes deployment identity, Reason:RefreshExtensionIdentityNotSet
I'm using Azure Machine Learning v2 SDK to create a model deployment on a kubernetes compute attached to an AML workspace. I'm able to deploy it locally as part of testing before deploying online. ...
0
votes
0
answers
648
views
Azure Machine Learning SDK V2 - Sequencing pipeline steps with sweep job
We are trying to migrate our azure ml pipelines from SDKV1 to SDKV2. The problem we are facing is that we cannot sequence the pipeline steps like we used to to in SDKV1.
In SDKV1, we could simply do ...
- 854
0
votes
1
answer
3k
views
Register dataset in AzureML Python SDK v2 from a dataframe
I am looking to upgrade to the new Azure AML Python SDK v2.
However, I can't replicate registering a dataset (now called data assets) directly from a dataframe. it seems that now the only option is to ...
- 37