Prepping for my sci-fi inspired data science projects live stream
We decided on the last call to do img classification #CNN w/ #NVIDIA GPUs and #RAPIDS #opencv #cuml #seaborn #cuDF #cupy and started a PRD. I'll share my finished PRD and get to building live for 2 hours.
Wed, July 23 1:15p ET / 12:15p CT / 5:15p GMT
Feel free to come by & say 'hello'
https://www.youtube.com/live/2IPZ35XpZaY?si=IDqW2EHAGNDed7jZ
At PyData NYC tutorial, by Jacob Tomlinson, I learned that now it is possible to access the same array on the GPU from pytorch and cupy. I'm loving how it will let you use the strengths of different libraries without dealing with extra memory copies.
https://nyc2024.pydata.org/cfp/talk/VAVRYW/
#PyDataNYC2024
#pytorch
#cupy
To get the most from your hands-on learning experience, please complete these steps prior to getting started: - **Review the agenda**, prerequisites, and suggested material for full-day workshops (as detailed in the course datasheet below). This is an important step to properly prepare for the workshop. - **Create or log into your [NVIDIA Developer Program account](https://courses.nvidia.com/join)** - https://courses.nvidia.com/join. You will receive an email letting you know when your account is ready. This account will provide you with access to all of the DLI training materials during and after the workshop. You will have three months of access to all course materials. - **Visit [websocketstest.courses.nvidia.com](http://websocketstest.courses.nvidia.com/)** and make sure all three test steps are checked “Yes.” This will test the ability for your system to access and deliver the training contents. If you encounter issues, try updating your browser. _Note: Only Chrome and Firefox are supported._ - **Check your bandwidth**. 1 Mbps downstream is required and 5 Mbps is recommended. This will ensure consistent streaming of audio/video during the workshop to avoid glitches and delays. Now you’re ready to get started with the tutorial! Simply enter the code **NVIDIA_XLAB_NV24** at [courses.nvidia.com/dli-event](https://developer.nvidia.com/dli/courses.nvidia.com/dli-event)
CUDAスレッド並列起動のエンジニアリング。50個のAIモデルをGPU並列同時高速トレーニング。
https://qiita.com/tetsutakamurata76/items/8a565207a0bc4f52f41e?utm_campaign=popular_items&utm_medium=feed&utm_source=popular_items
並列同時計算。GPU実行の勾配降下法。多くの解の候補が同時に探索され、最適解に近づく可能性が格段に高まる。
https://qiita.com/tetsutakamurata76/items/edeccb769e19c21d6966?utm_campaign=popular_items&utm_medium=feed&utm_source=popular_items
動画生成AIは、４次元重みテンソルニューラルネットワークが効くのかな。
https://qiita.com/tetsutakamurata76/items/f9f37c234fe29214585d?utm_campaign=popular_items&utm_medium=feed&utm_source=popular_items
Qwen AIによる、自動で高効率なGPU対応コードに変換するプログラム。
https://qiita.com/tetsutakamurata76/items/b37d5c3ffb54360e9b8c?utm_campaign=popular_items&utm_medium=feed&utm_source=popular_items
デジタル量子コンピュータ。全探索を高速計算。GPU時間: 26.0642秒。秘密の最適化を施し高速化すると GPU時間: 0.0592秒。
https://qiita.com/tetsutakamurata76/items/b3ded2aee726c0ec7737?utm_campaign=popular_items&utm_medium=feed&utm_source=popular_items
次世代の高速計算。自動で高効率なGPU対応コードに変換するプログラム。
https://qiita.com/tetsutakamurata76/items/5a87aff4bf6085b1c61b?utm_campaign=popular_items&utm_medium=feed&utm_source=popular_items
「GPUとcupyを用いた並列計算のオセロAI」。探索の最前線を突破。
https://qiita.com/tetsutakamurata76/items/575443cae49739289d61?utm_campaign=popular_items&utm_medium=feed&utm_source=popular_items
Hacker News
dawn
David Ruffner