'Iterate folders in Datastore in Azure Machine Learning service
We have training pictures coming in to Azure blob storage. They are places in folders like ProductA/yyyy-mm-dd, where the date is when the material was put in blob storage.
We have registered the blob container as a Datastore. When we run the training, we want to use the python sdk to create a Dataset, or a new version of the dataset, called ProductA and that should reference all date folders existing at that moment.
This is described here: https://docs.microsoft.com/en-us/azure/machine-learning/how-to-version-track-datasets#versioning-best-practice
In that example data is in folders called "Week xx". What is not described is how to know which all folders to reference when creating the dataset.
How can we access the blob storage to find the relevant folders to include? Can we use the Datastore to access files and folders directly?
Sources
This article follows the attribution requirements of Stack Overflow and is licensed under CC BY-SA 3.0.
Source: Stack Overflow
| Solution | Source |
|---|