End to end CUDA accelerated #datascience libraries built on @apachearrow, scaled with @dask_dev & Dask-SQL for ETL, ML, graph analytics, and DL preprocessing
Joined October 2018
- Tweets2.8K
- Following166
- Followers12.9K
- Likes3.8K
Pinned Tweet
The RAPIDS 24.04 release is out now! Check out the new accelerated vector search library, expanded zero-code change experiences for pandas & NetworkX workflows, optional query optimization for Dask workflows, support for Python 3.11 and pandas 2, and more. nvda.ws/4aTabyH
Join us as we host a Virtual Connect with Experts event this Friday, October 25, 2024 at 10am PT. This month, our topic is cuDF and cuML! We have an amazing lineup of RAPIDS developers & users on our expert panel to answer your questions. Learn more here: nvda.ws/4eRRjSL
Big Updates in RAPIDS 24.08!
- cuDF.pandas now supports workloads larger than GPU memory!🔥
- cuVS gains CPU/GPU interop + new, accuracy enhancing, advanced graph capabilities.
- UMAP gets a massive speed boost—up to 300x faster! - Plus, CUDA 12.5 support!
nvda.ws/4cPGVJj
RAPIDS brings BIG updates: v24.02 leans into CPU+GPU in RAFT's vector search, RMM host memory pinning, & zero code change NetworkX acceleration; XGBoost 2.0+cuML now available; & deploy it all on Databricks+Dask. Also, RAPIDS GTC24 sessions are scheduled!👉nvda.ws/49HOFf7
Taking a moment to show appreciation for the ❤️ our community continually shows us. NVIDIA RAPIDS recently won Datanami’s Readers’ Choice Award for Big Data & Open Source! 🎉🙏 Congrats also to our category co-winners: Gitlab & Apache! datanami.com/this-just-in/da… #datanami #ThankYou
3
1,076
Happy Valentine's Day to your graph workflows!❤️Accelerate your Python graph analytics workflows with the new RAPIDS cuGraph backend for NetworkX. Check out @becker_data 's interview on Datanami & see how to drop your 2 hr+ process to a lovable < 1 min! datanami.com/2024/02/13/nvid…
#rapidsai cudf.pandas code accelerator + @VoltronData 's Ibis' #Python -> #SQL framework?
Connect your SQL backend to #Nvidia #GPUs for lightning-fast performance -- from 4 minutes to less than 1 second on 100M rows -- no code changes required. nvda.ws/47VDQ8Q
100% pandas coverage at up to 150x with 0 code change. Introducing cudf.pandas - for instantly accelerating your current code on #NVIDIA #GPUs. #DataScience . docs.rapids.ai/api/cudf/stab…
RAPIDS cuDF’s pandas accelerator mode is now available in open beta, accelerating #pandas nearly 150x with zero code changes.
#DataScience, #Python, #RAPIDS
Read the announcement blog: 👇 nvda.ws/469JnqC
RAPIDS won InfoWorld's 2023 Bossie Award, recognizing us as one of this year's leading open source tools in data and ML. We're honored! We couldn't have done this without our amazing community of developers, ecosystem partners, & users working together🙏🎉 nvda.ws/479CHK4
On Nov 8th, join us at the AI and Data Science Virtual Summit for the opportunity to engage in interactive online discussions with industry leaders, researchers, & Kaggle Grandmasters. Check out the session lineup & register for free today: nvda.ws/3tuV79x. See you there!
Shop the RAPIDS 23.08 Back2School Special with even DEEPER⏲️SAVINGS on large scale text for ETL/LLMs, IO, &graphs.🚨Light Special News on🆕docker container deploy🚨➕CUDA 12 conda & cuXFilter pip📦. Celebrate new kid, cuProj, & cuSignal 🎓ing to cuPy High! nvda.ws/44xE2ZG
Want to start your journey into graph analysis, its practical applications, & how to take advantage of GPU acceleration? We’ve kicked off our new intro tutorial blog series on Graph Analysis w/ cuGraph! The first one, by cuGraph’s Don Acosta, landed today: nvda.ws/3DaOZVM
@bwyogatama & team have been continually trying to improve our custom user-defined aggregate functions (UDAF) performance. They created a new framework that speeds up UDAF for a large # of groups by up to 8000x over previous ways. Check out their paper: dl.acm.org/doi/pdf/10.1145/3…
RAPIDS’ 23.06 release brings Christmas in (nearly) July! cuSpatial gives pip wheels & DE-9IM, RAFT gives k-selection improvements, cuGraph gives MNMG perf & mem improvements, cuDF gives timezone support, & CUDA12 via pip for everyone! Open your gifts here nvda.ws/3JmjivR
GPU accelerated spatial analytics got an upgrade! Check out the new cuSpatial capabilities and example code here:
medium.com/rapids-ai/cuspati…
Ready to take the dive into GNNs? Get your GNN model training started -& finished- quickly with Alexandria's (linkedin.com/in/abarghi) getting started guide to Graph Neural Networks with cuGraph using PyG and pyTorch. There's even an after credits spoiler. bit.ly/42oB2OF
Check out what's new in RAPIDS 23.04! Bring in your spatial data easily with cuSpatial's GeoSeries input, explore new E2E GNN examples, leverage ANNs and a faster brute force KNN in RAFT, & optimize your speed with new benchmarking tools. Learn more here: bit.ly/3McuNbb
