Downsampling#
When an interactive table is displayed by itables, the table data is embedded in the notebook output. As we don’t want your notebook to become super heavy just because you displayed a large table, we have a downsampling mechanism in place.
When the data in a table is larger than maxBytes, which is equal to 64KB by default, itables will display only a subset of the table - one that fits into maxBytes, and display a warning that points to the itables documentation.
If you wish, you can increase the value of maxBytes or even deactivate the limit (with maxBytes=0). Similarly, you can set a limit on the number of rows (maxRows, defaults to 0) or columns (maxColumns, defaults to 200).
import itables
itables.init_notebook_mode()
itables.options.lengthMenu = [2, 5, 10, 20, 50, 100, 200, 500]
itables.options.maxBytes = "8KB"
df = itables.sample_dfs.get_indicators()
itables.downsample.as_nbytes(itables.options.maxBytes), itables.downsample.nbytes(df)
(8192, 28000)
df
| id | name | unit | source | sourceNote | sourceOrganization | topics |
|---|---|---|---|---|---|---|
Loading ITables v2.4.0-dev from the init_notebook_mode cell...
(need help?) |
To show the table in full, we can modify the value of maxBytes either locally:
itables.show(df, maxBytes=0)
| id | name | unit | source | sourceNote | sourceOrganization | topics |
|---|---|---|---|---|---|---|
Loading ITables v2.4.0-dev from the init_notebook_mode cell...
(need help?) |
or globally:
itables.options.maxBytes = "1MB"
df
| id | name | unit | source | sourceNote | sourceOrganization | topics |
|---|---|---|---|---|---|---|
Loading ITables v2.4.0-dev from the init_notebook_mode cell...
(need help?) |