Hacking Import for Speed: Writing a GPU Accelerator for Pandas
Save Big on Coursera Plus. 7,000+ courses at $160 off. Limited Time Only! Grab it Explore the intricacies of Python's import system and its potential for performance optimization in this 30-minute PyCon US talk. Dive into an ambitious project that hacks the import of pandas to accelerate large portions of it on the GPU using cuDF, a GPU DataFrame library. Learn about the basics of import hacking and Pythonic proxy patterns, discovering how these dynamic Python features can effectively accelerate any code utilizing pandas, including third-party libraries. Examine the technical and social challenges that necessitate these sophisticated solutions, and gain insights into potential resolutions. Experience a journey filled with successes, failures, aspirations, and ventures into lesser-known aspects of Python. Ideal for Pythonistas interested in the import system and its performance-hacking potential, as well as developers seeking to enhance the speed of the vast ecosystem built on libraries like numpy and pandas without modifying existing code.