Tile Tool

The Tile tool groups datasets by assigning a value (tile) based on ranges in the data. The Tile tool groups datasets using one of the following methods: Equal sum, Equal records, Smart tile, Unique value, and Manual tile.

Two columns are appended to the data. Tile number is the assigned tile of the row. Tile sequence number is the row number of the row's position within the Tile.

Configure the tool

In Properties, select a Tile method, and then select which columns to gropu by (optional).

  • Equal sum: Creates tiles with equal total values. It does this by assigning tiles to cover a range of values where each tile has the same total of the Sum column based on the sort order of the in-coming rows.

    • The total of the Sum column is as close as it can be for each tile without breaking a row.

    1. Number of tiles: Specify how many groups to assign to the data.
    2. Sum column: Specify the column to tile by that results in ranges where the sum of this number is equally distributed across the tiles.
    3. Sort column (optional): If you select a column from this drop down list, the data will be sorted on the specified column before it is assigned to a tile. If no column is specified, the rows are assigned to tiles based on their in-coming sort order.
    4. Sort method: This option shows only if you select Sort Column (above). Set the sorting method that you want to use. Your options include Unicode Code Point, Dictionary Order (English, French, German, Italian, Portuguese, and Spanish languages), Gojūon (Japanese), and Pinyin (Simplified Chinese).

  • Equal records: Divides rows equally by dividing .input rows (records) into the specified amount of tiles so that each tile is assigned the same amount of rows.

    1. Number of tiles: Specify how many groups to assign to the data.
    2. Do not split tile on column (optional): If you identify a column , a tile will not be split across it. If no column is selected, the tile number takes preference over the specified column data, thus potentially splitting groups.
    3. Sort Column (optional): If you select a column from this drop down list, the data will be sorted on the specified column before it is assigned to a tile. If no column is specified, the rows are assigned to tiles based on their in-coming sort order. The number of rows in the input may not be equally divisible by the number of tiles you specify. Unless you specify the sort column, this is based solely on the row's position in the input file.
    4. Sort Method: This option shows only if you select Sort Column (above). Set the sorting method that you want to use. Your options include Unicode Code Point, Dictionary Order (English, French, German, Italian, Portuguese, and Spanish languages), Gojūon (Japanese), and Pinyin (Simplified Chinese).

  • Unique value: Creates tiles for unique values. For every unique value in a specified column or columns, a unique tile is assigned. If multiple columns are specified, a tile is assigned based on that combination of values.

    1. Unique column: Check the box of the column or columns to tile by.

    2. If the Leave unsorted check box is left unchecked, the table is automatically sorted by the column or columns you specified for the tile.

  • Smart tile: Creates tiles based on the Standard Deviation of the values in the specified column. The tiles assigned indicate whether the row's value falls within the average range (=0), above the average (1) or below the average (-1), etc.

    • The tile value represents the mean of the data range. The tile numbers explain the range directly. The range always start at -0.5 and end at +0.5:

    Tile Range
    0 -0.5 to 0.5
    -1 -1.5 to -0.5
    1 0.5 to 1.5
    1. Tile numeric column (must be numeric): Choose the column to tile by. The column must contain a numeric value.
    2. Select the naming convention to apply to the output column:

      3. Do not output name column: No additional output column name is generated.

      Output name column: An additional descriptive output column name is appended to the output. Descriptors include: Average, Above Average, High, Extremely High, Below Average, Low, Extremely Low.

      Output verbose name column: In addition to the descriptors mentioned above, the value range that the tile indicates is listed in parenthesis.

      High (12750 to 155000)

    Manual: You can set up rules manually. To do this, specify the cutoffs for the tiles by typing a value on a new line for each range.

    1. The numeric column (must be numeric): Select the column to tile by.
    2. Enter one or more tile cutoffs: Type each tile's upper limit on a new line in the box provided.

    Tiles are applied in Manual mode but the rows are not sorted. You must add a Sort Tool and specify the columns by which to sort by, if that is how you want the data displayed.

    Group by columns (optional): Tile rows based on a grouping column. By selecting a Group or Groups of columns, a series of tiles is created for every row within the specified group.