I have deployed an ML skill using OOB DU model (invoices)
Then also used it in studio as an ML extractor and everything is working
Then problem occured when i wanted to add a new field to this extractor
I went to AI Center - data labelling to add a new field to the extractor (and also removed / hid some of the fields i dont need).
then i exported this result and ran a pipeline to retrain the ML skill using this exported result.
Finally i updated the ML Skill version.
In studio when i clicked “Get capabilities” to update the field list, its showing this error
This is what the Studio log is showing An error occurred while getting the capabilities: System.Exception: The selected ML Skill 'InvoiceSkill2' does not have an access uri
is your ML skill set to “auto update” in AI center? If so, you need to do that, so the ML extractor gets updates every time you modify the ML model.
You can delete the add the data extraction scope again. Get capabilities using the same endpoint and API key that you use and it should update the new fields, if you already have done option no 1.
have tried it but same error, log = An error occurred while getting the capabilities: System.Exception: The selected ML Skill 'InvoiceSkill2' does not have an access uri
that’s awesome to hear! Sometimes there are little hiccups with the endpoints but always test the endpoint if the skill doesn’t work. In your case, a link might have been broken hence once you created a new skill it works.