Trains or retrains an active solution in a Custom dataset group
Source:R/personalize_operations.R
personalize_create_solution_version.RdTrains or retrains an active solution in a Custom dataset group. A solution is created using the create_solution operation and must be in the ACTIVE state before calling create_solution_version. A new version of the solution is created every time you call this operation.
See https://www.paws-r-sdk.com/docs/personalize_create_solution_version/ for full documentation.
Usage
personalize_create_solution_version(
name = NULL,
solutionArn,
trainingMode = NULL,
tags = NULL
)Arguments
- name
The name of the solution version.
- solutionArn
[required] The Amazon Resource Name (ARN) of the solution containing the training configuration information.
- trainingMode
The scope of training to be performed when creating the solution version. The default is
FULL. This creates a completely new model based on the entirety of the training data from the datasets in your dataset group.If you use User-Personalization, you can specify a training mode of
UPDATE. This updates the model to consider new items for recommendations. It is not a full retraining. You should still complete a full retraining weekly. If you specifyUPDATE, Amazon Personalize will stop automatic updates for the solution version. To resume updates, create a new solution with training mode set toFULLand deploy it in a campaign. For more information about automatic updates, see Automatic updates.The
UPDATEoption can only be used when you already have an active solution version created from the input solution using theFULLoption and the input solution was trained with the User-Personalization recipe or the legacy HRNN-Coldstart recipe.A list of tags to apply to the solution version.