2

I am trying to run an image classifier using Keras on Azure ML Studio. However, I have huge amount of images sorted into class-specific subfolders on my local machine that I'd like to upload to Azure Blob Storage. When I use "Create Dataset" there is an option to upload the entire folder but it removes all subfolders, uploads only images in each subfolder and dumps them into the main folder. The reason I want to keep the subfolders is that each subfolder is a class and I want to use Keras flow_from_directory(). So, the folder structure I want to upload images should be as follows:

- Main_folder
  --- subfolder1
  --- subfolder2
  --- subfolder3
  --- subfolder4
  --- subfolder5
        .
        .
        .
  --- subfolderN

Does anyone know how to do this? Any other alternatives would be appreciated as well

Shora Dehkordi
  • 51
  • 3

1 Answers1

0

Rather than uploading from the "Create Dataset UI" you should upload to blob before hand, then browse to your already created folder (and subfolders) during the "Create Dataset" flow

Anders Swanson
  • 3,637
  • 1
  • 18
  • 43