Datastore.upload method is working, but impossible to delete files after

[StintouchDev]

Hi, I have wrongly used the upload method of datastore class (from a Bing conversation/chatgpt response) and so have uploaded files I do not want on my Azure ML datastore. I do not know how to remove them and when I came to the documentation here to find a delete method, I do not even find the upload method...

[...]

# Connect to the workspace
ws = Workspace(subscription_id, resource_group, workspace_name)

# Get the default datastore for the workspace
datastore = Datastore.get_default(ws)

# Upload a directory to the datastore
datastore.upload('directory_path')

The files are at the root of my workspaceblobstore in datastore tab of Data section of my Azure ML workspace, and cannot find any solution to delete them.

Please can you help me ?

---

Détails du document

⚠ Ne pas modifier cette section. Elle est requise pour learn.microsoft.com ➟ Liaison des problèmes GitHub.

• ID: b35c96a9-b082-7b3f-3c26-15b6ec753b75
• Version Independent ID: 771cf70f-497b-23b4-4a87-a9d12f129b24
• Content: azureml.core.Datastore class - Azure Machine Learning Python
• Content Source: AzureML-Docset/stable/docs-ref-autogen/azureml-core/azureml.core.Datastore(class).yml
• Service: machine-learning
• Sub-service: core
• GitHub Login: @DebFro
• Microsoft Alias: debfro

[StintouchDev]

Finally, I have succeeded to delete unwanted files by using the Azure Storage explorer (on Azure Platform, not Azure ML Platform). I have followed this article: https://learn.microsoft.com/en-us/azure/machine-learning/how-to-export-delete-data

But still do not know why the upload method is working on Datastore object.

[StintouchDev]

In my terminal, there is this log : "Datastore.upload" is deprecated after version 1.0.69. Please use "Dataset.File.upload_directory" to upload your files from a local directory and create FileDataset in single method call. See Dataset API change notice at https://aka.ms/dataset-deprecation.

This explains all. Bing conversation (chatgpt online) gave me an old method and I stupidly trusted him.