Machine Learning

Welcome! The Azure Machine Learning team invites you to share and vote for features to help you build, manage, and deploy custom machine learning models.

Have a technical question, or want to learn more? Please visit our documentation, MSDN forum or StackOverflow.


  1. Allow for dynamic paths in DataReferences

    Currently, when using the DataReference object to access directories on storage like ADLS Gen 1, we can't dynamically define the path that the DataReference refers to, because it is only evaluated once (at execution time).

    So for instance, if we wanted a DataReference to output data to {CURRENT DATE}/results, such as /DIR/2020/02/10/results, we're not currently able to do this as a datetime.now() would only be called once (when we first evaluate the graph).

    This means we have to do unnecessary extra steps (using the ADLS Gen 1 SDK, or additional ADF steps) to move outputs into the proper directory.

    4 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Flag idea as inappropriate…  ·  Admin →
  2. Azure DevOps ML extension - bug when triggering release pipelines

    I have multiple release pipelines triggered by new model versions. Each pipeline uses different models, but with the same name (they are in different workspaces of the same resource group).

    The issue is that all pipelines seem to be triggered by the same model (the one in the oldest workspace), even though they're using different references, and different service connections. I expect them to be triggered by updates to their respective models, of course.

    I can provide more details if needed.

    4 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Flag idea as inappropriate…  ·  Admin →
  3. Please add function to monitor "Used studio experimentation hour","Used Production Web API's Compute hours about previous month", "Used Prod

    I want to monitor below 3 things.
    Please add function that monitor below 3 things.

    1.Used studio experimentation hour
    2.Used Production Web API's Compute hours about previous month
    3.Used Production Web API's transactions about previous month

    4 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Flag idea as inappropriate…  ·  Admin →
  4. Need AML SDK for .NET

    AML SDK is only available in Python. I am developing a .Net application (in C#) and need the same features as Python SDK. Is there any plan for porting python SDK to .NET SDK?

    4 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    1 comment  ·  Flag idea as inappropriate…  ·  Admin →
  5. Need hyper parameter tuning for ML models

    Should allow to tune hyper parameters in order to improve algorithm performance. Implement Grid search, Randomized Search and Hyperopt for parameter tuning.

    4 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    1 comment  ·  Flag idea as inappropriate…  ·  Admin →
  6. ML studio Experiments page to allow folders

    In ML Studio, at experiments, we need to allow a user to create folders to organize various projects. Usually one creates different experiments to tackle the problem in different ways. In the end, maybe only one or two experiments are put to use. However, the user would like to retain all prior experiments for future reference on what was tried. It also helps to allow user to add a Readme.txt to go along with each project to document. In our case, a workspace is shared, and the person who created the experiments may leave. In addition, after two years I…

    4 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    1 comment  ·  Flag idea as inappropriate…  ·  Admin →
  7. Script parameters in Azure ML portal

    The RunDetails widget displays the script parameters of the Estimator that were used for that Run. It would be great if they were also displayed on the Azure portal. Furthermore, i would like to be able to search and filter Runs based on the script parameters used.

    4 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Flag idea as inappropriate…  ·  Admin →
  8. Monetization of pre-trained ML Models via AI-store

    As a user it would be great to be able to use an existing pre-trained model for some of my AI use-cases. This model can be supplied by a third party supplier. As a user I pay a fee per minute of usage.
    The same approach is currently available witin the AWS Sagemaker marketplace.

    3 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Flag idea as inappropriate…  ·  Admin →
  9. Default compute target

    An awesome feature has been added to the new Visual interface: when I want to run my experiment, I have to specify what kind of machine I want to run it on.

    When the experiment has to run more than once, I found only one – tiny bit – annoying feature: every time you run the experiment, you have to choose a compute target. What if I want to use the same target for the next few days, or even longer?

    I wish there was a possibility to set a default size, and use that, unless I want to scale…

    3 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Flag idea as inappropriate…  ·  Admin →
  10. Expose Cluster Purpose On Arm Template Deployments

    In the AzureML SDK you are able to specify when attaching an AKS Compute Instance to the workspace whether the purpose is DEVTEST or FASTPROD.

    For development environments you can specify DEV_TEST so that the minimum 3 Nodes are not imposed on the cluster (keep cost down on test environmets).

    I noticed that this functionality does not seemed to be exposed when using ARM Templates (https://docs.microsoft.com/en-us/azure/templates/microsoft.machinelearningservices/2020-01-01/workspaces/computes#AKS).

    This means that if you are trying to easily spin up test environments using ARM templates it will set the purpose to FAST_PROD which means that you will need a…

    3 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Flag idea as inappropriate…  ·  Admin →
  11. Need a module to identify Low / High Correlated feature with Target Variable as well as High Correlated input features.

    A new module which outputs High/low Correlated input features with Target Variable as well as to identify High correlated input features.

    This module would help to create better Trained Data Set and that would turn to create better ML model.

    3 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Flag idea as inappropriate…  ·  Admin →
  12. Resync Keys

    It is very said we need manually "Resync Keys" in ML studio. We can not even find a API to call resync key in power-shell.

    3 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Flag idea as inappropriate…  ·  Admin →

    Hi Azure Customer,

    Thank you for your feedback here. Since Azure Machine Learning Studio will be deprecated. We will not be supporting in Studio, investigtaing how Resync Keys works with visual interface

    We will get back to you soon once we have any update.

    Thank you for your understanding.

    Regards,
    Azure CXP Community

  13. Add the function to delete any version of the dataset

    If you update a dataset with incorrect data, there is no way to delete that version. It would be helpful if there was a function that could delete only the specific version you want to delete.

    3 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Flag idea as inappropriate…  ·  Admin →
  14. model_data_collection should log failed API request too

    We plan to leverage modeldatacollection message type in traces to log our API request payload (instead of printing the payload to STDOUT) only to realize that failed requests were not logged. It's actually more useful to log payload for failed requests so we can reproduce the issue rapidly.

    3 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Flag idea as inappropriate…  ·  Admin →
  15. Enable Machine Learning Studio for Government Cloud

    Please enable azure machine learning services for government cloud customers

    Currently there is no timeline for releasing ML services to Gov Cloud users, but we are interested in the service.

    https://azure.microsoft.com/en-us/global-infrastructure/services/?products=machine-learning-studio

    3 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Flag idea as inappropriate…  ·  Admin →
  16. Add a chart for evaluating a time-series forecast model

    Evaluate automated machine learning experiment results (https://docs.microsoft.com/en-us/azure/machine-learning/how-to-understand-automated-ml)

    This site has charts and metrics for Classification and Regression, but no for time-series forecast models. Customers have asked us to add a chart like the one attached, so please consider it.

    "chart.png" file: a comparison chart of real data and prediction results.

    3 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Flag idea as inappropriate…  ·  Admin →
  17. Support ParallelRunStep and Many Models in Visual Designer

    In order to implement a Many Models solution code is required[1]. This solution with the ParallelRunStep should be accomplishable via the Visual Designer.

    I notice that an icon is available for this step AFTER it's created via code, but not an option in the pipeline editor.

    [1] <https://github.com/microsoft/solution-accelerator-many-models>

    3 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Flag idea as inappropriate…  ·  Admin →
  18. Add the AzureSMR R package with support to HTTPS

    Could be very useful interact directly into a R Script with azure resources (example: storage account, blob, ecc...)

    3 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    1 comment  ·  Flag idea as inappropriate…  ·  Admin →

    Dear Azure Customer,

    Thanks for the feedback. Your feedback is very important to us. Our engineering team will investigate it and consider your feedback seriously. We will get back to you soon.

    Thank you for understanding.
    Azure CXP Community

  19. Read HHTP response in Azure Machine Learning

    I created a logic app that invoke a Machine Learning pipeline when the blob storage is changed. When I create the method HHTP request pass two parameters in the JSON: file store and the file path (the same example as: https://docs.microsoft.com/en-us/azure/machine-learning/how-to-trigger-published-pipeline), however the Machine Learning receive the response but the file of pipelines is not change (the value is static). How can I read the HTTP response in Azure Machine Learning and pass this information in the pipline endpoint so that the result is always dynamic depending on blob storage?

    3 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Flag idea as inappropriate…  ·  Admin →
  20. Issue with NaN during Export csv filein Export data module

    Hi,
    I am using Export Data module in Machine learning designer to export csv file. Issue is with numeric columns which contains NaN. Result should be empty field/value but exported csv contains NA, which cause troubles in additional file processing.

    3 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    1 comment  ·  Flag idea as inappropriate…  ·  Admin →
  • Don't see your idea?

Machine Learning

Categories

Feedback and Knowledge Base