A Dash of Bokeh - Speaker Deck

Slide 1

Slide 1 text

Bryan Van de Ven @bigreddot Oct 2018 A Dash of Bokeh

Slide 2

Slide 2 text

No JavaScript Bokeh ? https://bokeh.pydata.org/en/latest Interactive visualization, widgets, and tools For the browser, with or without a server Versatile and high-level graphics Streaming, dynamic, large data What

Slide 3

Slide 3 text

No JavaScript No JavaScript Bokeh ? https://bokeh.pydata.org/en/latest Interactive visualization, widgets, and tools For the browser, with or without a server Versatile and high-level graphics Streaming, dynamic, large data What

Slide 4

Slide 4 text

Simple (to write) Apps for Data Science What

Slide 5

Slide 5 text

Concentrate on your work Why

Slide 6

Slide 6 text

Concentrate on your work Mostly or completely written in Python, little HTML or CSS coding Why

Slide 7

Slide 7 text

Concentrate on your work Mostly or completely written in Python, little HTML or CSS coding Simple python scripts, no special classes of frameworks Why

Slide 8

Slide 8 text

Concentrate on your work Mostly or completely written in Python, little HTML or CSS coding Simple python scripts, no special classes of frameworks Useful for exploratory analysis or sharing and publishing Why

Slide 9

Slide 9 text

Slide 10

Slide 10 text

Concentrate on your work Mostly or completely written in Python, little HTML or CSS coding Connect the full PyData stack to interactive web apps Simple python scripts, no special classes of frameworks Useful for exploratory analysis or sharing and publishing Automatically mirrors and synchronizes Python and browser state Why

Slide 11

Slide 11 text

Users Gonna Use Who

Slide 12

Slide 12 text

KeplerGO/lightkurve Who

Slide 13

Slide 13 text

KeplerGO/lightkurve Who

Slide 14

Slide 14 text

Bokeh Today

Slide 15

Slide 15 text

Circa June 2014 Who

Slide 16

Slide 16 text

Circa Yesterday 2018 Who

Slide 17

Slide 17 text

Makes the Dream Work Damian Luke Bryan Philipp Mateusz Havoc Jean-Luc Peter Maggie Sarah Brendan Julia Who

Slide 18

Slide 18 text

Makes the Dream Work New faces too: !!! Damian Luke Bryan Philipp Mateusz Havoc Jean-Luc Peter Maggie Sarah Brendan Julia Who

Slide 19

Slide 19 text

Makes the Dream Work Special Recent Thanks: Karel Van de Plassche, Eugene Pakhomov, Anthony Doucuoliagos. New faces too: !!! Damian Luke Bryan Philipp Mateusz Havoc Jean-Luc Peter Maggie Sarah Brendan Julia Who

Slide 20

Slide 20 text

Slide 21

Slide 21 text

one point oh !!!

Slide 22

Slide 22 text

one point oh Support for MultiPolygons with holes !!!

Slide 23

Slide 23 text

one point oh Support for MultiPolygons with holes Several ﬁxes and improvements to DataTable !!!

Slide 24

Slide 24 text

one point oh Support for MultiPolygons with holes Several ﬁxes and improvements to DataTable !!! CustomAction for user-deﬁned Toolbar buttons

Slide 25

Slide 25 text

one point oh Support for MultiPolygons with holes Several ﬁxes and improvements to DataTable Plain JSON export/embed functions !!! CustomAction for user-deﬁned Toolbar buttons

Slide 26

Slide 26 text

one point oh Support for MultiPolygons with holes Several ﬁxes and improvements to DataTable Plain JSON export/embed functions !!! CustomAction for user-deﬁned Toolbar buttons Callback to allow AjaxDataSource to adapt JSON responses

Slide 27

Slide 27 text

one point oh Support for MultiPolygons with holes Several ﬁxes and improvements to DataTable Plain JSON export/embed functions Scatter glyph for parameterizable marker type !!! CustomAction for user-deﬁned Toolbar buttons Callback to allow AjaxDataSource to adapt JSON responses

Slide 28

Slide 28 text

one point oh Support for MultiPolygons with holes Reuse webdrivers for faster PNG/SVG export by default Several ﬁxes and improvements to DataTable Plain JSON export/embed functions Scatter glyph for parameterizable marker type !!! CustomAction for user-deﬁned Toolbar buttons Callback to allow AjaxDataSource to adapt JSON responses

Slide 29

Slide 29 text

Slide 30

Slide 30 text

Basic Principles

Slide 31

Slide 31 text

Building Blocks How

Slide 32

Slide 32 text

Building Blocks How

Slide 33

Slide 33 text

Building Blocks How

Slide 34

Slide 34 text

Not just for Python http://hafen.github.io/rbokeh Plays well with R ecosystem: HTMLwidget, RMarkdown…

Slide 35

Slide 35 text

Map Visual Properties To Data Columns How Pick what graphical primitives to use, provide the data, and specify how to map visual properties to data ﬁelds. Bokeh will take care of the rest.

Slide 36

Slide 36 text

Streaming Data Two data source methods: .stream to append data incrementally to column ends .patch for random access updates anywhere How

Slide 37

Slide 37 text

Styling How Simple YAML or JSON format Useful for a consistent look across plots Any Bokeh property can be targeted

Slide 38

Slide 38 text

Some User Contributed Styles How Caliber Minimal Dark Monokai Dark

Slide 39

Slide 39 text

Built-in Layouts How Bokeh offers "rows and columns" layouts

Slide 40

Slide 40 text

Built-in Layouts How

Slide 41

Slide 41 text

Built-in Layouts How

Slide 42

Slide 42 text

Built-in Layouts How Bokeh objects can be responsive

Slide 43

Slide 43 text

Built-in Layouts How Bokeh objects can be responsive Some problems and known issues

Slide 44

Slide 44 text

Built-in Layouts How Bokeh objects can be responsive Improved, simpler layout system for 1.1 Some problems and known issues

Slide 45

Slide 45 text

Custom Templates How Import embed function Load BokehJS resources Embed Bokeh objects Boilerplate script

Slide 46

Slide 46 text

Custom Templates How

Slide 47

Slide 47 text

Dask UI Uses CSS Grid How http://examples.dask.org

Slide 48

Slide 48 text

Custom Extensions http://bokeh.pydata.org/en/latest/docs/user_guide/extensions.html Python class interface JavaScript implementation Normal usage How

Slide 49

Slide 49 text

Custom Extensions Add capability for… How

Slide 50

Slide 50 text

Custom Extensions Add capability for… Wrapping different widgets How

Slide 51

Slide 51 text

Custom Extensions Add capability for… Wrapping different widgets Adapting 3D JavaScript libraries How

Slide 52

Slide 52 text

Custom Extensions Add capability for… Wrapping different widgets Adapting 3D JavaScript libraries Adding LaTeX labels How

Slide 53

Slide 53 text

Custom Extensions Add capability for… Wrapping different widgets Adapting 3D JavaScript libraries Adding LaTeX labels Community doesn’t have to wait! How

Slide 54

Slide 54 text

Bokeh Server Applications How Bokeh Server automatically keeps Python and JavaScript in sync

Slide 55

Slide 55 text

Bokeh Server Applications How Python callbacks can execute on property changes (and JS will react too)

Slide 56

Slide 56 text

Bokeh Server Applications How Document level callbacks can perform server-driven updates

Slide 57

Slide 57 text

An Example

Slide 58

Slide 58 text

Goal Example of streaming API Connects to PyData tools Custom layout for nice look Demonstrates:

Slide 59

Slide 59 text

Goal Example of streaming API Connects to PyData tools Custom layout for nice look Demonstrates:

Slide 60

Slide 60 text

The Python Bits

Slide 61

Slide 61 text

The Template Bits

Slide 62

Slide 62 text

Live Demo

Slide 63

Slide 63 text

Final Thoughts

Slide 64

Slide 64 text

Resources • GitHub: https://github.com/Bokeh/bokeh • Documentation: http://bokeh.pydata.org/en/latest • Example Apps: https://demo.bokehplots.com • Tutorials: https://mybinder.org/v2/gh/bokeh/bokeh-notebooks/master?filepath=tutorial%2F00%20-%20Introduction%20and%20Setup.ipynb • Mailing List: https://groups.google.com/a/continuum.io/forum/#!forum/bokeh • Gitter Chat: https://gitter.im/bokeh/bokeh Do

Slide 65

Slide 65 text

Call to Action Do

Slide 66

Slide 66 text

Call to Action Install Bokeh and provide feedback • BSD license for everything (JS, Python, server) • https://github.com/Bokeh/bokeh Do

Slide 67

Slide 67 text

Call to Action Install Bokeh and provide feedback • BSD license for everything (JS, Python, server) • https://github.com/Bokeh/bokeh Donate to Bokeh via NumFOCUS • https://www.flipcause.com/secure/cause_pdetails/MzE5NjE= Do

Slide 68

Slide 68 text

Call to Action Install Bokeh and provide feedback • BSD license for everything (JS, Python, server) • https://github.com/Bokeh/bokeh Engage companies for custom work • Code integration, analytical and visualization expertise • Contact us to be put in touch bokehplots@gmail.com Donate to Bokeh via NumFOCUS • https://www.flipcause.com/secure/cause_pdetails/MzE5NjE= Do

Slide 69

Slide 69 text

No content