You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
As we further broaden the scope of the Columns interfaces and provide support for various out-of-memory data backends it's becoming clear that the current interface isn't quite sufficient. When accessing a Columns datastructure it is sometimes necessary to do so by index, e.g. when displaying a Table and only a small subset of rows should be shown or when trying to randomly sampling from the data. Currently this requires each column to be loaded into memory in its entirety to index into them. Instead there should be an API on the Columns class to slice the row and column directly. I could imagine something like this table.cells[0:10, 0:2], where cells is a small object that allows slicing by index.
The text was updated successfully, but these errors were encountered:
As we further broaden the scope of the Columns interfaces and provide support for various out-of-memory data backends it's becoming clear that the current interface isn't quite sufficient. When accessing a Columns datastructure it is sometimes necessary to do so by index, e.g. when displaying a Table and only a small subset of rows should be shown or when trying to randomly sampling from the data. Currently this requires each column to be loaded into memory in its entirety to index into them. Instead there should be an API on the Columns class to slice the row and column directly. I could imagine something like this
table.cells[0:10, 0:2]
, where cells is a small object that allows slicing by index.The text was updated successfully, but these errors were encountered: