Advanced Data Visualizations In Jupyter Notebooks
summary
Jupyter notebooks combine Python callbacks and NumPy implementations to build interactive apps using ipywidgets and bqplot that run in real time. Examples include k‑means with 300 points and five clusters that shows local‑minima retries, gradient descent that diverges with high learning rates, Gaussian process regression with posterior bands, a tile‑map of the S&P 500's 500 constituents, a Wealth of Nations bubble/choropleth animator, and a Twitter sentiment dashboard trained on 100k tweets that exposes labeling errors via a confusion matrix and triangle plot. SVG rendering limits prevent plotting millions of points.
AI-written summary · about
chapters (22)
- 0:00Introduction
- 0:25Overview
- 0:51Jupyter Notebooks
- 1:32Features
- 3:04Libraries
- 3:59Widgets
- 4:55Interactive Widgets
- 7:17Theoretical Machine Learning
- 11:25Gradient Descent
- 15:09Gaussian Processes
- 18:54Financial Use Cases
- 19:20Time Series Analysis
- 22:37Tile Map
- 26:51Wealth of Nations
- 27:16Wealth of Nations Data
- 28:13Bubble Chart
- 30:53Choropleth
- 35:08Twitter Sentiment Analysis
- 38:08Polarity Inverting Words
- 39:40Positive vs Neutral
- 40:25Conclusion
- 41:25Scalability
related talks
32mAdvanced Visualization in Jupyter Notebooks
29mUsing Jupyter notebooks to develop and share interactive data displays
41mInteractive Visualization in Jupyter with Bqplot and Interactive Widgets
28mAdvanced Graph Visualizations in Jupyter Notebooks
01h 38mEffective Data Visualization
31mbqplot Interactive Data Visualization in Jupyter
38mJumping into Jupyter Notebooks
02h 31mTech Tutorial: Data visualization in Python