The Data Science Journey
summary
Dyalog shows concrete work: a booth at JSM with ~5,000 attendees and over 100 direct pitches, plus demos of native APL toolchains. The TamStat GUI runs regression wizards that push R² from ~68% to ~80% when adding features. APLearn implements scikit‑learn–style models in APL and fits an SVC on the breast‑cancer set with ~96% accuracy after extensive training (hundreds of thousands of updates). New primitives key and stencil enable fast grouping and sliding‑window ops but GPU/support for heavy numeric kernels remains missing.
AI-written summary · about
chapters (15)
- 0:00The importance of data science
- 1:08Dyalog at the Joint Statistical Meetings
- 3:14What is data science?
- 3:33Collection of data
- 4:38Structuring of data
- 7:27Cleansing of data
- 9:15Visualisation of data
- 11:12Summarisation of data
- 11:42APL libraries for data science
- 15:49TamStat demonstration
- 17:40Issues with ML libraries
- 18:27APLearn demo
- 24:24Key (⌸) and Stencil (⌺)
- 26:35Conclusion
- 27:20Questions
related talks
01h 26mData Driven Action: a primer on Data Science
39mSoftware Analytics with Data Science on Software Data
54mDesign Patterns vs Anti pattern in APL
46mJump-Start Your Data Science Learning Java & Jupyter Notebooks
04h 02mHands-on Session on Data Science for Beginners
42mDevelopment of Data Science Ecosystem for Ruby
53mCan Haskell Become a Great Language for Data Science?
43mAdvanced Data Visualizations In Jupyter Notebooks