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. Any plans to for ML Studio to support Teradata?

    Any plans for ML Studio Import/Export module to support Teradata?

    1 vote
    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. Add description tag for azureml.train.hyperdrive

    Currently when registering a best run model from a hyperdrive run instance, there is no way to add model description.

    e.g:

    from azureml.train.hyperdrive import HyperDriveRunConfig

    hypertuneconfig = HyperDriveRunConfig()
    hyperdrive
    run = exp.submit(config=hypertuneconfig)
    hyperdrive
    run

    no way to add a description tag here

    model = bestrun.registermodel(modelname='model', modelpath='outputs/model.pkl')

    The current workaround is to re train a single model after finding the best run parameters and registering the model with the description that way. This is cumbersome

    1 vote
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  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

  3. Provide predict_proba for multiclass prediction models

    I have successfully trained a multi-class decision forest in ML Studio, but I would like to return all class probabilities as a 2d-array [nscores, nclasses], not just the best class prediction. This is similar to the predict_proba() function from scikit learn models. Can you provide? My customer is asking for this feature.

    1 vote
    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. Enable downloading engineered feature tables

    Model interpretability posts as an important part in ML adoption. If we can download the engineered feature tables (like what H2O.AI) offers, that will speed up our customer adoption.

    1 vote
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Flag idea as inappropriate…  ·  Admin →
  5. excel

    ML Excel extension support MAC version

    1 vote
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Flag idea as inappropriate…  ·  Admin →
  6. 1 vote
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Flag idea as inappropriate…  ·  Admin →
  7. Need reason code for model interpretability

    Can we have reason code (like the ones in DataRobot) to enable model interpretability?

    Also feature importance ranking graph?

    1 vote
    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. How do i find out what is taking up my storage in the Free Azure ML Workspace?

    When I go to my Datasets, the biggest one i see is a 15MB file twice, so how is it that my Workspace says it is over the limit. Please provide a way to identify and delete whatever is taking up space in my workspace.

    1 vote
    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. Allow connecting a container registry from a different AAD tenant

    It's already possible to link an "Azure Container Registry" from a different subscription to a "Machine Learning service workspace", but that subscription has to be in the same AAD tenant.

    I'd like to be able to connect an "Azure Container Registry" from a different AAD tenant, so that we can have one central container registry for all our projects/environments, which are using different subscriptions in different AAD tenants.

    1 vote
    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. Please provide Azure datalake as a source.

    , Either you can provide data source like you have provided in data factory where we are open to connect cross platform connection as source.

    1 vote
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  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

  11. restrict access to Storage Account when using ML

    Currently the access to the required storage account cannot be restricted when using ML. For security reasons it is wise to restrict access ("firewalls and virtual networks")

    1 vote
    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. Please add support for federated learning in AML

    More and more users are opting not to send their data back to cloud. In order to respect user privacy and scale the training of our ML models, it would be great if AML would support federated learning.

    1 vote
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Flag idea as inappropriate…  ·  Admin →
  13. keras in R

    Need support for Keras in R

    1 vote
    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. Please add function to enable distributed processing with Azure Machine Learning Studio itself.

    Please enable distributed processing with Azure Machine Learning Studio itself.
    When I do data analysis with module, an error occurs if the work time exceeds 24 hours.
    So, To reduce processing time, please add function to enable distributed processing with Azure Machine Learning Studio itself.

    1 vote
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    triaged  ·  1 comment  ·  Flag idea as inappropriate…  ·  Admin →
  15. My account has access to 5 tenants. It seems as if it is only checking the default tenant before telling me I do not have access to my sub.

    I am trying to run Jupyter notebooks in a ML workspace in my sandbox environment. My default tenant is our production environment. It appears as if the azureml python authentication is only checking the default tenant before telling me I do not have access to me subscription.

    1 vote
    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. Model data collection for ACI deployed models

    As far as I can tell from the documentation it is only possible to enable data collection on models deployed AKS. In scenarios where it is not desirable to create and manage Kubernetes in a production environment it would be useful to have this feature also for ACI based deploys.

    1 vote
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    1 comment  ·  Flag idea as inappropriate…  ·  Admin →
  17. Please integrate Azure Machine Learning Studio with Microsoft Visual Studio?

    Hi,

    It will be very helpful to create a Machine learning experiment in visual studio if you integrate AML Studio in Microsoft Visual Studio (Community Edition). Currently, it supports the only web-based, I think it will be very helpful if you provide support for desktop-based AML Studio.

    Thanks.

    1 vote
    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. error during "ws = Workspace.from_config()"

    When I run the command "ws = Workspace.fromconfig()" in the notebook I get an error. The error is "AuthenticationException: You don't have access to d9c30d04-f393-4fbb-87fa-4b75ef106734 subscription. All the subscriptions that you have access to = [SubscriptionInfo(subscriptionname='P-CSA 250$', subscription_id='13c3cc54-b676-4845-9974-a4e7f6c8b175')]. Please run 'az login' or set d9c30d04-f393-4fbb-87fa-4b75ef106734 as default subscription and retry". The subscription d9c30d04-f393-4fbb-87fa-4b75ef106734 exists. It is on this subscription that I have created the machine learning workspace.

    1 vote
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    1 comment  ·  Flag idea as inappropriate…  ·  Admin →
  19. Limit of experiments in a workspace

    Is there a limit of how many experiments can we create in one workspace? Is there a limit of how many runs can we create in one experiments?

    1 vote
    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,

    We do not have limits on pipelines runs or limits on pipelines per workspace at this point in time (not sure if we want to introduce limits as part of the SKU/billing effort).
    We have a limit on the size of a pipeline (30,000 nodes) and the numbers of schedules that will trigger pipelines per month (100,000)

    We will update our document to make it clear. Thank you.

    Regards,
    Azure CXP Community

  20. 1 vote
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Flag idea as inappropriate…  ·  Admin →
  • Don't see your idea?

Machine Learning

Categories

Feedback and Knowledge Base