How to use AzureMLDataset

Matteo Pocchiari 11 Reputation points
2020-12-11T08:43:20.88+00:00

Hi,

I have used the data labeling system within Azure Machine Learning studio to label a dataset of images.
Then the output of the labeling system is a new dataset that can be found in the "Dataset" section of the designer in ML studio.

The problem is that this new dataset module, that can be dragged and dropped in the pipeline, has datatype "AzureMLDataset" (or datasoruce type "amldataset"), which then I cannot connect to any other module in the pipeline because there is no module which accepts as input something with datatype "AzureMLDataset" (or with datasource type "amldataset").

I have seen that it is possible to consume the dataset using python, but I would like to use Azure ML studio because it is more convenient to the system I am working in.

How can I use the AzureMLDataset output module inside ML studio?

Thank you in advance!

Azure Machine Learning
Azure Machine Learning
An Azure machine learning service for building and deploying models.
2,687 questions
Azure Data Science Virtual Machines
Azure Data Science Virtual Machines
Azure Virtual Machine images that are pre-installed, configured, and tested with several commonly used tools for data analytics, machine learning, and artificial intelligence training.
68 questions
0 comments

1 answer

Sort by: Most helpful
Most helpful Newest Oldest
  1. romungi-MSFT 43,656 Reputation points Microsoft Employee
    2020-12-11T10:35:45.227+00:00

    @Matteo Pocchiari Yes, the azure ml dataset which is registered from your labels after exporting it cannot be used directly in the designer because all the modules are using dataframe ports as input to their respective modules. You can however follow these steps to convert the labeled datasets to pandas dataframe and re-upload the downloaded file as a dataset which can be used in the studio with the correct format.

    @Lu Zhang (MSFT) Is there any other workaround for this scenario?

    1 person found this answer helpful.