@kmaehashi Profile picture

Kenichi Maehashi

@kmaehashi

Engineering Manager @PreferredNet, @CuPy_Team dev lead, he/him

Similar User
まむ photo

@mamurai1208

比戸 photo

@sla

Shuji Suzuki (shu) photo

@shu65

Keigo Kawaai photo

@dorakichi25

Yusuke Doi photo

@d0i

Masaaki Fukuda / 福田 昌昭 photo

@fkdmjp

Hiroyuki V Yamazaki photo

@hvyamazaki

im132nd photo

@im132nd

Yu Akasaka photo

@yuakasaka

Crucial Japan photo

@CrucialJapan

ほそのゆうこ(高円寺funny meal) photo

@HosonoYuko

Masahiro Sakai photo

@masahiro_sakai

べいえりあⓂ️ photo

@mr_bay_area

Preferred Networks photo

@PreferredNetJP

Yuta Tsuboi photo

@yuutat

Pinned

Thrilled to unbox some incredible swag box from @GitHubSecurity today! 🎉 Absolutely adore the cute Octocat figurine and the custom hoodie featuring my handle 🥰 Big thanks for hosting the #Hacktocat VIP program!

Tweet Image 1

Kenichi Maehashi Reposted

Optuna 4.1 is out! 🤖 AutoSampler: Automatic Selection of Optimization Algorithm 🚀 More Scalable RDB Storage Backend 🧑‍💻 Five New Algorithms in OptunaHub (MO-CMA-ES, MOEA/D, etc.) 🐍 Python 3.13 Support 👇See what’s new github.com/optuna/optuna/…

Tweet Image 1

Kenichi Maehashi Reposted

Today, @pydatanyc kicked off with a tutorial on GPU Accelerated Python! @numba_jit, @CuPy_Team, and @RAPIDSai are all great ways to accelerate your code. Check out the full tutorial on the NVIDIA Accelerated Computing Hub to learn more: github.com/NVIDIA/acceler…

Tweet Image 1

Kenichi Maehashi Reposted

【PFN2024インターンブログ】夏季インターンの細川颯介さんによる寄稿です。CUDAの新機能であるCUDA Graph Conditional NodesのサポートをCuPyに実装しました。 tech.preferred.jp/ja/blog/cupy-c…


Kenichi Maehashi Reposted

📣 Ask CUDA Python questions to experts! Join the LIVE event "CUDA Virtual Connect with Experts" this Friday, September 27th at 10am PT. Discuss CuPy, Numba @numba_jit, RAPIDS @rapidsai, or CUDA Python ecosystem with NVIDIA experts. linkedin.com/posts/katrinar…


Kenichi Maehashi Reposted
Tweet Image 1

Kenichi Maehashi Reposted

We're trending on GitHub for a few days in a row 🚀 github.com/trending/pytho… If you haven't, star us to help spread the word! 🙌🌟 github.com/cupy/cupy


Kenichi Maehashi Reposted

Google Gemma 2 implemented in pure NumPy/CuPy 👀 github.com/githubpradeep/… buttondown.com/ainews/archive…

Tweet Image 1

Kenichi Maehashi Reposted

Optuna 4.0, the new major release, is here! Huge thanks to all contributors and users! We are glad to see Optuna has become a leading hyperparameter optimization OSS! See what's new: medium.com/optuna/announc…

Tweet Image 1

Kenichi Maehashi Reposted

MN-Core challenge に勝手に便乗して、総合得点もしくは FizzBuzz において PFN 社内記録を倒した場合、個人的に50万円ずつ賞金を出します!詳細はこのポエムの後ろの方にあります: shinh.skr.jp/m/?date=202408… 賞金出すことになるのかならないのか、むっちゃ楽しみです #mncore


Kenichi Maehashi Reposted

🎉 CuPy v13.3 is just out! This release includes performance improvements (“one-time only warm-up” eliminated), support for CUDA 12.5 & 12.6, and better interoperability with NumPy 2. Check out the release notes for details! github.com/cupy/cupy/rele…


Kenichi Maehashi Reposted

ゴルフ大会やります! preferred.jp/ja/news/pr2024… この時代にコードの長さが速度に直結するコーディングができます!


Kenichi Maehashi Reposted

n月刊ラムダノートに深層学習コンパイラに関する記事を寄稿しました。汎用コンパイラと違って深層学習コンパイラはどういう最適化に焦点を置いて、どのように実装されているかを、PFNで使われている"PFVM"を例に解説してあります。PDF版は本日発売で1000円になります! lambdanote.com/blogs/news/n-v…


Kenichi Maehashi Reposted

著者です。間違いなく読みごたえがある、すばらしい本である自信があります。どうぞよろしくお願いいたします。 x.com/oreilly_japan/…

8月発行予定の新刊『Binary Hacks Rebooted —低レイヤの世界を探検するテクニック89選』の情報をAmazonおよび楽天ブックスで公開開始しました! books.rakuten.co.jp/rb/17936277/ amazon.co.jp/dp/4814400853/



Kenichi Maehashi Reposted

📢 Just announced: nvmath-python brings NVIDIA accelerated math to #Python. ✅ Pythonic APIs for using CUDA math operations ✅ Works with your favorite Python packages ✅ #OpenSource Beta version is available now! #HPC Learn more now: nvda.ws/3VV2fGN

Tweet Image 1

Kenichi Maehashi Reposted

Open software tools, policies + communities for biomedicine are essential! This week, our Open Science grantees gathered to share their work, connect with other groups + discuss how they can make important scientific tools & information more widely accessible.

Tweet Image 1

Heading back to Tokyo 👋🇺🇸

Tweet Image 1
Tweet Image 2
Tweet Image 3
Tweet Image 4

Kenichi Maehashi Reposted

📢 CuPy v13.2 is out! This release added support for NumPy 2.0. You can use CuPy with the coming NP2! This release also has improved NCCL preloading, which brings better operability on mixed-library environments.


Kenichi Maehashi Reposted

【論文】PFNの @asi1024, @xuzijian629 によるメモリアドレス割り当てに関する論文がInternational Symposium on Memory Management 2024に採択されました。セグメント木を用いた焼きなまし法で割り当てを事前に決定することでメモリの断片化をほぼなくすことができます。 conf.researchr.org/details/ismm-2…


Loading...

Something went wrong.


Something went wrong.