The Return of Clojure Data Science
summary
Clojure data science revives a practical workflow: abandon Incanter baggage, use Gorilla REPL notebooks for interactive reports, keep data as vectors-of-maps so specs and transducers work naturally, and use transducer libraries (inc. kixi.stats) for fast summaries. Demonstrations: a Kaggle universal basic income dataset with a homemade Kramers V transducer for categorical correlation, and MNIST digit recognition using cortex whose training UI updates live — both examples run and produce useful outputs.
AI-written summary · about
chapters (26)
- 0:00Intro
- 0:35The bastard child
- 1:10Be careful
- 1:31The return of Clojure
- 3:10Structure
- 3:46Basic Tuning
- 4:12Sampling
- 4:38Making sense of data
- 5:19Notebooks
- 7:22Data Structures
- 7:50Matrix Data Set
- 10:02Closure Data Structures
- 10:53KickSee Stats
- 11:18Correlation Matrix
- 11:32Huard
- 12:14Kaggle
- 13:04Guerrilla Notebook
- 13:31Visual Check
- 15:01Standard Closure
- 15:35Filter
- 15:49Correlation
- 17:29Coretex
- 18:08Neural Networks
- 19:01Cortex
- 19:44Confusion Matrix
- 21:16Roadmap
related talks
43mDoing Data Science with Clojure
35m"Clojure for Data Science in the Real World"
27mData and Clojure: Using Clojure for more than software development
38mScientific Clojure, a bird's eye view
18mNavigating Clojure: A Fresh Look From a Data Analysis Background
26mThrough the looking glass w Data Rabbit: "A System built for Seeing"
34mClojure by neophyte
01h 03mUnleash Your Data with Clojure: Using Transducers and Sequences