Data Pipelines
Lyra's left hand panel lists data pipelines. A data pipeline allows you to import a raw data set (B), apply a series of data transformations (C), and construct scale functions (F) that map data values to visual values. Pipelines are listed in the order they were created. To create a new pipeline, hit the New Pipeline button at the bottom of the panel (G). Clicking a pipeline's name expands or contracts its inspector. Hovering over the pipeline name reveals two icons (A): the pencil renames the pipeline, the minus deletes it.
 |
Use the In the popover, give your dataset a name. If you wish to input your data directly, select Lyra will load your raw data and prompt you to verify that the data types it inferred for each field are correct. Currently, Lyra supports |
The current output of the pipeline is always shown in the data table (D). We transpose rows and columns, and paginate data, to make it easier to scan. Green fields are those found in the raw data set while yellow fields are those added subsequently (for example, through data transformations). By default, Lyra adds an index field that counts off data records, beginning with 0.
Fields can be dragged off the table and dropped over various elements in Lyra to establish data mappings. Viable drop targets include the cyan drop zones that appear on the visualization canvas when you begin dragging, or fields in property inspectors that are marked with a pale dot.