Effective Data Visualization
summary
Hands-on tutorial that uses seaborn, matplotlib, plotly and sklearn to show what actually changes a figure: histogram bin choice vs kernel density estimation (KDE gives lower error with optimal bandwidth), PCA separating simple linear variance but failing on a swiss‑roll, and t‑SNE unfolding that manifold into distinct clusters. Builds interactive candlestick and time‑series plots, a choropleth of COVID cases by FIPS county, and warns against perceptually misleading jet colormap while demonstrating better sequential and diverging palettes.
AI-written summary · about
chapters (19)
- 0:00Introduction
- 19:10Density Plot
- 24:40Levels
- 31:20Stats Material
- 32:22Data Sets
- 33:20Scatter Plots
- 34:22Standard Deviation
- 35:33Fit
- 37:38Robust Fit
- 38:22Color Species
- 38:47Exercise
- 42:34Joint Plot
- 43:44Single Plot
- 45:01Categorical Plot
- 46:18Exercises
- 48:31Plots
- 50:08Box Plot
- 52:04Questions
- 56:00Solution
related talks
01h 38mEffective Data Visualization
03h 09mEffective Data Visualization
01h 28mPython Visualization for Exploration of Data
02h 31mTech Tutorial: Data visualization in Python
31mFundamentals of Data Visualization
48mData Visualization in Practice
51mArt of Effective Visualization of Multi-dimensional Data
43mAdvanced Data Visualizations In Jupyter Notebooks