Building a maintainable plotting library-PyData NYC
summary
Concrete how-to for making reusable plotting code: use matplotlib rcParams style files for lab-wide defaults, wrap plotting into functions that accept an ax= None and propagate kwargs, and when needed implement a custom artist — here a LineCollection with 25 segments driven by widths computed from an image using numpy blurring and masks to sketch a dog — packaged and testable, and it visibly works (adjusted widths ×10 and smoothed to improve results).
AI-written summary · about
chapters (6)
- 0:00Domain-specific plotting
- 4:01Custom style sheets
- 12:12Wrapping plotting functions
- 17:51Custom plots for domains
- 29:34Matplotlib's design philosophy
- 33:48Q&A
related talks
30mFrom plotting to chart porn - Creating slick charts with Matplotlib
28mThe bumps in the road: A retrospective on my data visualisation mistakes
59mPlotting with Matplotlib; Telling Static, Animated, & Interactive Stories
01h 28mPython Visualization for Exploration of Data
39mRubyPlot - Creating a Plotting Library for Ruby
42mRefactoring PyChart
01h 38mEffective Data Visualization
01h 26mTurning DataFrames into Pretty Pictures with Plotnine