Talk

Talks

Introducing FireDucks: A High performance Compiler accelerated DataFrame Library

July 25, 2024 San Diego Python Users Group

slide

FireDucksでpandasを高速化!

Jul 18, 2024 みんなのPython勉強会#106

slide video

FireDucksという高速データフレームライブラリを紹介します。FireDucksは、C++で書かれたマルチスレッドバックエンド、ユーザープログラムの性能問題を自動検出して最適化するJITコンパイラ、pandasと高い互換性を持つフロントエンドを備えており、既存のpandasプログラムを変更することなくそのまま高速化できます。本トークではFireDucksの仕組みや高速化事例を紹介します。

FireDucks - A compiler accelerated Dataframe Library

Jul 17, 2024 The Austin Python Meetup

slide

Introducing a high performance compiler accelerated DataFrame Library, FireDucks

Jul 12, 2024 SacPy

slide video

The Pandas library, being the top choice among Data Scientists, many legacy applications developed using this library related to data preparation are in high demand for optimization to reduce the performance cost, especially when being executed on the cloud. The existing high-performance pandas alternatives sometime compel a user to learn completely new API syntaxes, whereas some of the others demand paying an extra hardware cost for GPU, etc.

To address the same, we at NEC R&D Lab, have developed FireDucks, a solution that’s been crafted for the contemporary data professional who loves pandas and deals with voluminous and complex data on a regular basis. With the promise of highly compatible pandas APIs and the revved-up performance, FireDucks can serve the demands of this digital age and transform the arduous task of data wrangling into a more efficient and less taxing endeavor. It can be simply installed on any Linux system using pip. Please feel free to check out the website: https://fireducks-dev.github.io

FireDucksによるデータ準備の高速化

Jul 03, 2024 Intel AI Summit Japan

video

FireDucks Compiler Accelerated DataFrame Library with pandas API (Lightning Talk)

Jun 20, 2024 みんなのpython勉強会#105

slide