Data unit manager
From Piki
The data unit manager is a GUI element found in preprocessing that allows you to create new data units and manage existing ones.
Usage
| Data unit manager |
|---|
|
The central element in the data unit manager is the data unit list. It displays the currently loaded data units:
- Name: Name of the data unit.
- Size: Size of the data unit [tX : vY]xZ where X is the number of samples in the training set, Y the number of samples in the validation set and Z the number of features.
- Description: A description provided by the input format
Operations
- Adding a new data unit: Click on the "Add" button, choose an input format and follow the wizard.
- Changing format on exiting data unit: Double click on the data unit you wish to change or right-click on it and select "Modify Format..."
- Removing a data unit: Select the data unit in the list and click on "Remove" or press the DEL key or right-click on it and select "Remove". You will be asked to confirm the removal.
- Cloning a data unit: To create an identical copy of an exiting data unit, right-click and choose "Clone".
- Reloading a data unit: Right-click on the target data unit and select "Reload" from the menu.
- Saving data unit to CSV: To save the contents of a data unit right-click on the target data unit and select "Save to CSV" from the menu. The output buffer is saved (i.e. after all filters have been applied.
- Embedding data in solution: To embed the data from the data unit into the current solution select the target data unit and check the "Embed Data in Solution" checkbox. This is not recommended for larger quantities of data.
Indicators
- Data unit has new data: The name of the data unit will be written in green text.
- Errors in the data unit': The name of the data unit will be written in red text.