Pack: Online update experiment pack

Description:

This is a data iteration workflow used to iterate throug query update sets.

Description:

This is a data iteration workflow used to iterate throug query update sets.

Description:

This is a file containing query datasets for online updates of item recommendation operators.

Description:

This workflow trains and saves a model for a selected item recommendation operator.

Description:

This workflow trains and saves model for a selected rating prediction operator.

Description:

This workflow measures performance of three models.

Model learned on train data and upgraded using online model updates. Model learned on train data + all query update sets. Model learned on train data only.

Description:

This workflow measures performance of three models. Model learned on train data and upgraded using online model updates. Model learned on train data + all query update sets. Model learned on train data only.

Description:

This is a Model update workflow called from data iteration workflow on every given query set. In the Loop operator model and current training set are retrieved from the repository. Model update is performed on a given query set creating new model. Model and updated train set are saved in the repository.

Description:

This is a Model update workflow called from data iteration workflow on every given query set. In the Loop operator model and current training set are retrieved from the repository. Model update is performed on a given query set creating new model. Model and updated train set are saved in the repository.

Description:

This is a file containing query sets for online updates of rating prediction operators.

Description:

This is a main online update experimentation workflow. It consists of three Execute Process operators.

First operator executes model training workflow. Second operator executes online updates workflow for multiple query update sets. The last operator executes performance testing and comparison workflow.

Final performance results are saved in an Excel file.

Description:

This is a main online update experimentation workflow. It consists of three Execute Process operators. First operator executes model training workflow. Second operator executes online updates workflow for multiple query update sets. The last operator executes performance testing and comparison workflow. Final performance results are saved in an Excel file.