Training existing engines

To recognize content, cognitive engines need to be trained with libraries of assets, which involves showing the engine various examples of the objects you want the engine to detect. Training an engine requires creating an empty library, adding content (assets) to that library, publishing it, and polling the status of training tasks. 

Below is our documentation on both engine libraries and training them. 

Engine libraries

• Overview of engine libraries
• Engine configuration for libraries
• Running library-enabled engines

Training engines

• Training library-enabled engines manually
• Train an engine

• Developer's reference
• Engines
• Training existing engines