You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Costing a work load is hard in cognitive search because customers don't know what's inside PDFs or blobs. Since image analysis is so expensive, it would be great to run an initial skillset that returns information about the data itself: image count, image type,
For text, I don't know if this is even possible, but something to indicate what level of granularity would work best for splitting, or suggestions of skills that might add value.
The text was updated successfully, but these errors were encountered:
Costing a work load is hard in cognitive search because customers don't know what's inside PDFs or blobs. Since image analysis is so expensive, it would be great to run an initial skillset that returns information about the data itself: image count, image type,
For text, I don't know if this is even possible, but something to indicate what level of granularity would work best for splitting, or suggestions of skills that might add value.
The text was updated successfully, but these errors were encountered: