This folder contains several notebooks designed as an introduction to the various features of perspective-python
:
table_tutorial.ipynb
shows how to load, update, query, and serialize data usingTable
andView
, and how to connect multipleTable
instances together usingon_update
.widget_tutorial.ipynb
demonstrates how to usePerspectiveWidget
as a powerful interactive visualization component within a Jupyter notebook.pandas_pivots.ipynb
displays Perspective's ability to read pivots from a pivoted DataFrame and automatically apply it as part of aPerspectiveWidget
.
Each notebook is fully self-contained and should offer a good place to start for those interested in using perspective-python
whether within a Jupyter environment or in a pure-Python context.
For examples pertaining to perspective-python
Tornado servers, check out:
- tornado-python: a simple Tornado server that delivers a static dataset to the user using
perspective-python
and<perspective-viewer>
. - tornado-streaming-python: a streaming Tornado server that demonstrates
perspective-python
's high throughput and performance in streaming scenarios. - workspace-editing-python: a full-featured example using
<perspective-workspace>
that illustrates a deep and powerful integration between<perspective-workspace>
andperspective-python
.