How to use URI_File type for running an AutoML training Job on Azure Machine Learning SDK V2

Question

How to use URI_File type for running an AutoML training Job on Azure Machine Learning SDK V2

Sagar 0

Hi Team,

I am trying to run an automl.classification training job. The training and validation files are of URI_File type. The job is failing with the following error -

Encountered user error while fetching data from Dataset. Error: UserErrorException:
	Message: Failed to download mltable yaml with error: 
Error Code: UserError
Error Message: 
Error Code: ScriptExecution.StreamAccess.NotFound
Native Error: error in streaming from input data sources
	StreamError(NotFound)
=> stream not found
	NotFound
Error Message: The requested stream was not found. Please make sure the request uri is correct.| session_id=18388af5-ad59-4506-a40e-4291d3c468c0; Not able to find MLTable file| session_id=18388af5-ad59-4506-a40e-4291d3c468c0
	InnerException None
	ErrorResponse 
{
    "error": {
        "code": "UserError",
        "message": "Failed to download mltable yaml with error: \nError Code: UserError\nError Message: \nError Code: ScriptExecution.StreamAccess.NotFound\nNative Error: error in streaming from input data sources\n\tStreamError(NotFound)\n=> stream not found\n\tNotFound\nError Message: The requested stream was not found. Please make sure the request uri is correct.| session_id=18388af5-ad59-4506-a40e-4291d3c468c0; Not able to find MLTable file| session_id=18388af5-ad59-4506-a40e-4291d3c468c0"
    }
}

Is there a way to setup the job to use the URI_file type rather than an ML Table into the training job?

Thanks & Regards,

Sagar

SriLakshmi C 6,250 Reputation points Microsoft External Staff Moderator

2025-05-15T17:01:33.36+00:00
Hello @Sagar,

I understand that you're running into a bit of trouble with your AutoML classification job using the URI_File type. The error message suggests that the job can't find the specified MLTable YAML file.

To set up your job to use the URI_File type effectively, you need to ensure a few things:

Define your training and validation datasets using type: uri_file. These should be accessible via a valid URI, such as from Azure Blob Storage, Azure Data Lake, or any publicly accessible HTTP location.

Ensure that the files you reference are individual files, not folders, as uri_file is meant to point to a single file (e.g., .csv, .json, .parquet).

Confirm that the URI path is correctly formatted and points to an actual file location:

inputs: training_data: type: uri_file uri: "https://your-storage-account.blob.core.windows.net/container/train_data.csv" validation_data: type: uri_file uri: "https://your-storage-account.blob.core.windows.net/container/val_data.csv"

In your Python code using the SDK, define input data like this:

from azure.ai.ml import Input from azure.ai.ml.constants import AssetTypes input_data = Input( type=AssetTypes.URI_FILE, path="https://your-storage-account.blob.core.windows.net/container/train_data.csv" )

Ensure that your AutoML job is configured to accept uri_file type inputs instead of using MLTable or uri_folder.

Set proper authentication. Make sure the job has permission to access the file location via managed identity or the workspace identity, depending on how access is configured.

You can also refer to the documentation on JobInputType.UriFile and UriFileJobInput Class to ensure your setup aligns with the expected input configuration.

I hope this helps, do let me know if you have further queries.

Thank you!
SriLakshmi C 6,250 Reputation points Microsoft External Staff Moderator

2025-05-16T12:42:56.51+00:00

Hi @Sagar,

Following up to see if the above suggestion was helpful. And, if you have any further query do let us know.

Thank you!
Sagar 0 Reputation points

2025-05-19T05:48:05.6666667+00:00

Is there a parameter in Automl class where we can set the below recommendation?
Ensure that your AutoML job is configured to accept uri_file type inputs instead of using MLTable or uri_folder.

1 answer

Your answer

SriLakshmi C 6,250 Reputation points Microsoft External Staff Moderator

2025-05-15T17:01:33.36+00:00

Hello @Sagar,

I understand that you're running into a bit of trouble with your AutoML classification job using the URI_File type. The error message suggests that the job can't find the specified MLTable YAML file.

To set up your job to use the URI_File type effectively, you need to ensure a few things:

Define your training and validation datasets using type: uri_file. These should be accessible via a valid URI, such as from Azure Blob Storage, Azure Data Lake, or any publicly accessible HTTP location.

Ensure that the files you reference are individual files, not folders, as uri_file is meant to point to a single file (e.g., .csv, .json, .parquet).

Confirm that the URI path is correctly formatted and points to an actual file location:

inputs: training_data: type: uri_file uri: "https://your-storage-account.blob.core.windows.net/container/train_data.csv" validation_data: type: uri_file uri: "https://your-storage-account.blob.core.windows.net/container/val_data.csv"

In your Python code using the SDK, define input data like this:

from azure.ai.ml import Input from azure.ai.ml.constants import AssetTypes input_data = Input( type=AssetTypes.URI_FILE, path="https://your-storage-account.blob.core.windows.net/container/train_data.csv" )

Ensure that your AutoML job is configured to accept uri_file type inputs instead of using MLTable or uri_folder.

Set proper authentication. Make sure the job has permission to access the file location via managed identity or the workspace identity, depending on how access is configured.

You can also refer to the documentation on JobInputType.UriFile and UriFileJobInput Class to ensure your setup aligns with the expected input configuration.

I hope this helps, do let me know if you have further queries.

Thank you!
SriLakshmi C 6,250 Reputation points Microsoft External Staff Moderator

2025-05-16T12:42:56.51+00:00

Hi @Sagar,

Following up to see if the above suggestion was helpful. And, if you have any further query do let us know.

Thank you!
Sagar 0 Reputation points

2025-05-19T05:48:05.6666667+00:00

Is there a parameter in Automl class where we can set the below recommendation?
Ensure that your AutoML job is configured to accept uri_file type inputs instead of using MLTable or uri_folder.

Answer 1

Hello @Sagar ,

Here are the ways to create input dataset with uri_file type and pass it to auto ml classification job.

As per this documentation you need to pass Input type object as training data for automl class, below is the how you create and pass it as training data to classification.


my_training_data_input = Input(
type="uri_file", 
path="./data/training-data-folder/your_file.csv" #Paths can be local paths, remote data uris, or a registered AzureML asset ID.

)


classification_job = automl.classification(
    compute=compute_name,
    experiment_name=exp_name,
    training_data=my_training_data_input,
    target_column_name="y",
    primary_metric="accuracy",
    n_cross_validations=5,
    enable_model_explainability=True,
    tags={"my_custom_tag": "My custom value"},
)

In portal you can create it, follow below steps.

User's image

Next, give a name and select uri_file

User's image

Select data source and click on next.

User's image

According to data source add your training data files and click on create.

User's image

After creating use below code for creating auto ml classification object.

data_asset = ml_client.data.get("<Name given while creating above dataset>", version="1")
input_data = Input(path=data_asset.id)

classification_job = automl.classification(
    compute=compute_name,
    experiment_name=exp_name,
    training_data=input_data ,
    target_column_name="y",
    primary_metric="accuracy",
    n_cross_validations=5,
    enable_model_explainability=True,
    tags={"my_custom_tag": "My custom value"},
)

Above are the ways to create dataset as uri file, please try it and let us know if you have any query in comments or in private chat.

Thank you

JAYA SHANKAR G S 3,960 Reputation points Microsoft External Staff Moderator

2025-05-21T05:54:14.6733333+00:00

Hello @Sagar ,

Please check above solution and let us know if you have any query in comments.

Thank you
JAYA SHANKAR G S 3,960 Reputation points Microsoft External Staff Moderator

2025-05-22T04:33:37.4833333+00:00

Hello @Sagar ,

I believe above given solution helped you, do let us know in comments if you have any query.

Thank you

Share via

How to use URI_File type for running an AutoML training Job on Azure Machine Learning SDK V2

1 answer

Your answer