TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

GPU Acceleration with Polars and Nvidia Rapids

9 点作者 ZeroCool2u8 个月前

1 comment

ZeroCool2u8 个月前
Interestingly, only 1 test from the PDS-H1 benchmark at scale factor 80 experiences a performance regression. Looks like it&#x27;s test 6 and comes in at 0.8 of the CPU execution performance the standard polars engine.<p>Seems like there&#x27;s a more concrete user guide here: <a href="https:&#x2F;&#x2F;docs.pola.rs&#x2F;user-guide&#x2F;gpu-support&#x2F;#whats-supported-on-the-gpu" rel="nofollow">https:&#x2F;&#x2F;docs.pola.rs&#x2F;user-guide&#x2F;gpu-support&#x2F;#whats-supported...</a><p>It seems like they&#x27;ve done a great job of supporting most ops on the GPU already. The main one that is a really bummer is the categorical datatypes. Those are so common in ML workflows and using the native categorical datatype saves a lot of boiler plate as well as prevents you from messing something up during training!