question

Jef-1256 avatar image
0 Votes"
Jef-1256 asked GiftA-MSFT commented

Remove a published python script from designer in "Microsoft Auzre Machine Learning" framework ?

Hello,

I created a module that i have published using

 from azureml.pipeline.core.graph import InputPortDef, OutputPortDef
 from azureml.pipeline.core.module import Module
    
 datastore = ws.get_default_datastore()
    
 p_in = InputPortDef(
     name="p_in", 
     default_datastore_mode="mount", 
     default_data_reference_name=datastore.name, 
     label="Données de production"
     )
    
 module = Module.create(ws, name="Well Clustering", description="use well prod to create n class of producters")
 entry_version = module.publish_python_script("main.py", "initial", 
                                              inputs=[p_in], outputs=[], params = { "n_classes": 4},
                                              version="1", source_directory=".")


It works and i can see it in the Designer as a "Custom Module" with the right version, but how can i "unregister it"

azure-machine-learning
· 3
5 |1600 characters needed characters left characters exceeded

Up to 10 attachments (including images) can be used with a maximum of 3.0 MiB each and 30.0 MiB total.

Hi, have you tired using the deprecate method to deprecate the module?


0 Votes 0 ·

Hi, you can set the module to deprecate. Hope this helps.


0 Votes 0 ·

Whatever i tried, deprecated, disabled, module is still there.

0 Votes 0 ·

1 Answer

GiftA-MSFT avatar image
0 Votes"
GiftA-MSFT answered

Hi, please go to Azure ML Studio > Assets > Components to view all your custom components. Select target component and click Archive on the top action bar to archive that component from your list. You can also turn on View archived button to view all your archived component and restore them back to the list if needed. Hope this helps.

134439-image.png






image.png (64.9 KiB)
5 |1600 characters needed characters left characters exceeded

Up to 10 attachments (including images) can be used with a maximum of 3.0 MiB each and 30.0 MiB total.