) button to open the Change Sift Rules dialog.Using Sift Rules, CM4D Reporter takes the highly structured data loaded out of the Query and converts it into an organized form that can be easily used for statistical analysis in a report. Sift Rules are sets of predefined instructions that can define layouts, data conditions, or sorting within a report. Sift Rules can also be used to place annotation in order according to statistics.
Feature types are derived from the characteristics of the data. Using Sift Rules, you can turn specific feature types and characteristics on or off to determine what kind of data will appear in your report.
See also, Multiple Sift Rules.
1. Click the Change Sift Rules () button to open the Change Sift Rules dialog.
2. Select the Single Sift Rule label and click OK.
3. Click the Sift Rules (
) button to open the Sift Rules dialog.
Click the Sift Rules () button to open the Sift Rules dialog. Sift Rules allow you to select which Feature Type, Characteristic, Direction, Hand, and Feature Filter types you want reported. You can also apply Conditions and Hide or Show options to further filter the content of your data.
There may already be checked boxes in the Sift Rules dialog, as well as locked values. These defaults are due to the Sift Rule settings in the 4Reporter.ini file. If you do not have this file (located in the CM4D Reporter\User Config folder), than the Sift Rules will be read from the standard Cm4d.ini file (located in the CM4D Reporter installation directory).
SMI TypeTo select a Feature Type (Surface, Hole, etc.), select the check box to the left of the Type name. Select or clear the check boxes next to the SMI Characteristic Type (X, Y, Z, etc.) you wish to include. The SMI Characteristic Value (act or dev) can be selected from the list next to the SMI Characteristic Type. If 'dev' is used, the deviation type must be set in the Document Defaults Deviation Types dialog.
When you set a SMI Characteristic Value for a SMI Characteristic Type, the SMI Characteristic Type of the other Feature Types will change to match your selection.
If you changed the SMI Characteristic Value of the X to "act" in the Surface Feature Type section of the Sift Rules, the X in all of the other Feature Types (Holes, Slots, Reference, etc.) will change to "act".
When Significant or Toleranced is selected in combination with other rules, CM4D will only load those features which have all of the selections. For example, if both Significant and Primary are selected, only those features which are both Significant and Primary will be loaded.
CM4D looks at the characteristics of a feature, and if it meets certain criteria, CM4D will assume that the feature is one of the feature types available in CM4D.
Feature FiltersSelect the check boxes next to the pre-defined filter labels you wish to use. Under Must Match you can choose whether the data must match "at least one filter specified", or "all filters specified".
The filter value <unassigned> can be used in order to select all of the feature values that have not been assigned for Key, Restricted, or Unrestricted feature filters.
Feature Filters are defined during the creation of the database or assigned to your data using DataUtility. Open DataUtility to see if there are any Feature Filters available to you.
ConditionsSelect a check box next to a condition and then enter the condition in the field. The Must Match section allows you to choose whether the data must match at least one of the conditions selected, or all of the filters selected. A condition could be a variable or an expression, or a combination thereof.
If you entered ~nom, x~ > 1500 as a condition, you would get all of the points in a specific area of the car, those greater than 1500.
Conditions can use the following arguments:
|
Symbol |
Definition |
|
< |
Less Than |
|
> |
Greater Than |
|
[ |
Less Than or Equal To |
|
] |
Greater Than or Equal To |
|
& |
And |
|
| |
Or |
|
^ |
Direction |
SortingThe sorting available in the Sift Rules allows you to sort the contents of a DataSet. Sorting will only work on numerical data, and will be applied consecutively (i.e. sorting rule 1, sorting rule 2, etc). Variables and expressions can be used to sort a DataSet. The sorted data will be rounded to the precision (number of decimal places) set in your Document Defaults.
To create a sort rule, click on the first field in the Sorting section of the Sift Rules dialog and enter the expression {(-1)*(~mean)}. Enable the sort rule by selecting the check box next to number 1 to use the sorting parameter entered in the field. Click OK to apply the Sift Rule sorting in the DataSet grid. This same process can be used for Multi Sift Rules.
The main advantage of using Sift Rule sorting versus sorting on the DataSet level is that Sift Rules accommodate three sorting options at a time (or more when combining Single Sift Rules within a Multi Sift Rule). Using Multiple Sift Rules affords more control over DataSet sorting, as each Sift Rule within the Multi Sift has more options than the DataSet level sort. Also, Sift Rules have the Keep only Top Values options (see section below).
When setting up Sift Rule sorting, multiplying a variable by (-1) will sort the data in descending order. Using a variable without multiplying by (-1) will keep the results in the default ascending order.
After the other sift rules are applied for the sift, you can then sort the list and filter out (remove) from the list those rows whose value of the sorting criteria is not in the top so many or top percentage of the list of rows. For instance, after entering a variable such as ~mean~ or ~stddev~ into the first sort field, you can choose to show the Top number (e.g. 10 values) or the Top percent (e.g. 5% of all the values).
Say the Sift Rule conditions identify 50 rows in the DataSet. The variable ~mean~ is entered in the first Sort field, the Top number radio button is selected, and the Top number value is set to 10. In such a case, the 10 rows out of 50 that have the highest ~mean~ value will be listed in descending order in the grid, while the remaining 40 rows will be sifted out.
Hide/ShowThe Hide/Show section of the Sift Rules allow you to remove rows from a DataSet according to the following rules:
When the check box is selected, rows in the DataSet grid that are empty, i.e. rows that do not have values for any of the samples, will be removed. When the check box is cleared, empty rows will not be removed from the DataSet grid.
When the check box is selected, rows in the DataSet grid that are non-triggered for Event 1, i.e. rows where none of the samples have triggered Event 1, will be removed from the DataSet. When the check box is cleared, non-triggered rows will remain in the DataSet.
When the check box is selected, rows in the DataSet grid that are non-triggered by Event 2, i.e. rows where none of the samples have triggered Event 2, will be removed. When the check box is cleared, non-triggered rows will not be removed from the DataSet grid.
To show only rows-triggered by "Either Event 1 or Event 2", select both the "Show ONLY Rows Triggered by Event 1" and "Show ONLY Rows Triggered by Event 2".
When the check box is selected, rows in the DataSet grid that are non-triggered for both Event 1 and Event 2, i.e. rows where none of the samples have triggered both Event 1 and Event 2, will be removed. When the check box is cleared, non-triggered rows will not be removed from the DataSet grid.
Ignore the three Alignment options, as these do not apply in Reporter.
When this check box is selected, all Reference Features are added to the DataSet, regardless of which feature types are selected in the DataSet Sift Rules for SMI Type.
When the Reference Features used to Augment the DataSet are added, they are added according to the characteristics selected in the Sift Rules. However, even if the main check box of an SMI Type is off, any characteristics selected within the SMI Type will be used to determine which characteristics of the Reference Features will augment the DataSet.
When this check box is selected, all of the Reference Features used to create the Constructed Feature are added to the DataSet, but not the Constructed Feature itself.