Categories
hardware

DIY with AI: GTC to Host NVIDIA Deep Learning Institute Courses for Anyone, Anywhere

The NVIDIA Deep Learning Institute is launching three new lessons, which may also be taken for the primary time ever on the GPU Technology Conference subsequent month. 

The new instructor-led workshops quilt basics of deep studying, recommender methods and Transformer-based programs. Anyone hooked up on-line can sign up for for a nominal rate, and contributors can have get entry to to a completely configured, GPU-accelerated server within the cloud. 

DLI instructor-led trainings encompass hands-on far flung studying taught via NVIDIA-certified mavens in digital school rooms. Participants can engage with their instructors and friends in actual time. They can whiteboard concepts, take on interactive coding demanding situations and earn a DLI certificates of matter competency to give a boost to their skilled expansion.

DLI at GTC is obtainable globally, with a number of lessons to be had in Korean, Japanese and Simplified Chinese for attendees of their respective time zones.

New DLI workshops launching at GTC come with:

  • Fundamentals of Deep Learning — Build the boldness to tackle a deep studying venture via studying how one can teach a type, paintings with commonplace knowledge varieties and type architectures, use switch studying between fashions, and extra.
  • Building Intelligent Recommender Systems — Create various kinds of recommender methods: content-based, collaborative filtering, hybrid, and extra. Learn how one can use the open-source cuDF library, Apache Arrow, alternating least squares, CuPy and TensorFlow 2 to take action.
  • Building Transformer-Based Natural Language Processing Applications — Learn about NLP subjects like Word2Vec and recurrent neural network-based embeddings, in addition to Transformer structure options and how one can enhance them. Use pre-trained NLP fashions for textual content classification, named-entity popularity and query answering, and deploy subtle fashions for reside programs.

Other DLI choices at GTC will come with:

  • Fundamentals of Accelerated Computing with CUDA Python — Dive into how one can use Numba to collect NVIDIA CUDA kernels from NumPy common purposes, in addition to create and release customized CUDA kernels, whilst making use of key GPU reminiscence control ways.
  • Applications of AI for Predictive Maintenance — Leverage predictive repairs and determine anomalies to regulate disasters and steer clear of expensive unplanned downtimes, use time-series knowledge to expect results the use of device studying classification fashions with XGBoost, and extra.
  • Fundamentals of Accelerated Data Science with RAPIDS — Learn how one can use cuDF and Dask to ingest and manipulate huge datasets without delay at the GPU, making use of GPU-accelerated device studying algorithms together with XGBoost, cuGRAPH and cuML to accomplish knowledge research at huge scale.
  • Fundamentals of Accelerated Computing with CUDA C/C++ — Find out how one can boost up CPU-only programs to run their latent parallelism on GPUs, the use of ways like crucial CUDA reminiscence control to optimize sped up programs.
  • Fundamentals of Deep Learning for Multi-GPUs — Scale deep studying coaching to a couple of GPUs, considerably shortening the time required to coach a lot of knowledge and making fixing complicated issues of deep studying possible.
  • Applications of AI for Anomaly Detection — Discover how one can put into effect a couple of AI-based answers to spot community intrusions, the use of sped up XGBoost, deep learning-based autoencoders and generative hostile networks.

With greater than 2 million registered NVIDIA builders running on technological breakthroughs to unravel the sector’s hardest issues, the call for for deep studying experience is larger than ever. The complete DLI path catalog comprises quite a lot of subjects for somebody fascinated with studying extra about AI, sped up computing and information science.

Get a glimpse of the DLI enjoy:

Workshops have restricted seating, with the early chicken closing date on Sep 25. Register now.