Parens for Python
summary
Demonstrates using libpython-clj to call Python libs directly from the Clojure REPL (requires pip/conda), showing live numpy/matplotlib plots and Seaborn charts, and loading GPT-2 via Hugging Face into PyTorch tensors to tokenize and generate text. Shows REPL completion, memory/structure sharing and garbage collection across runtimes, notes GPT-2’s repetition and that adding a temperature parameter reduces it, and reports the examples run interactively and produce plots and generated text.
AI-written summary · about
chapters (7)
- 0:00Introduction
- 1:06The case for interop
- 5:41Why libpython-clj matters
- 12:28Plotting with Python
- 17:14Text generation demos
- 24:36Advanced data science tools
- 30:38Community and conclusion
related talks
31mAdding Python support to Scala tooling
24mEngage the Hyper-Python
34mIntegrating Python into the CLR with Python for .NET
25mPython as a Hackable Language for Interactive Data Science
41mInteractive GPU programming with ClojureCUDA & ClojureCL
49mYou Used Python for What?!
43mInstall Python. Quarto Render All the Things
35m"Clojure for Data Science in the Real World"