Train the model:

User Rating (TrainUsersRatingModel):

• RatingList: List of your Users ratings of the items to train the model 
• TestPortion: Portion of the RatingList to be used to evaluate the model (if empty all data will be used to train the model and no evaluation will be done on the model)
• NumberOfIteractions: Number of iterations that will be done on the training process (default: 100)

• RMSE: Root Mean Square Error, indicates the error of the algorithm (only valid if provided a value for TestPortion)
• Result: Indicates whether or not the training of the model was successful

Bought Together (TrainBoughtTogetherModel):

• ItemList: List of items and the item that is connected (for example, was bought together) to train the model 
• TestPortion: Portion of the ItemList to be used to evaluate the model (if empty all data will be used to train the model and no evaluation will be done on the model)
• Alpha and Lambda: Parameters of the algorithm that can be adjusted depending of the RMSE. Check the documentation of ML.NET for more details. 

• RMSE: Root Mean Square Error, indicates the error of the algorithm (only valid if provided a value for TestPortion)
• Result: Indicates whether or not the training of the model was successful

Make Predictions:

To make predictions just use the Server Actions available for that. Pass the a single Object (User Rating or Item) or list, depending of the Action that is being used, and one should get a prediction. It's necessary that you have trained respective model to make predictions.