This API call triggers the model being tested against the 20% of data allocated in the previous API call for testing.
If you set ‘contextual’ as false for ‘Build Data and Structures for Training and Testing’ API, then please also set ‘contextual’ as false here.
However, if you set ‘contextual’ as true for the ‘Build Data and Structures for Training and Testing’ API query, then you can set either true or false.
Using contextual during training should improve results for non-contextual testing. Please note that contextual testing results will always be significantly lower than non-contextual testing. The reason is that you are asking the algorithm to be more exact. Instead of predicting that a particular user watched a specific movie or bought a certain item, you ask how long the user watched or how many items the user purchased.
Once the model has been tested, an email will notify you and include the results.
For more information, please refer to the API documentation.