On-Demand Webinar: Introducing the Deep Sparse Platform

Deep Sparse Platform: Sparsify and Quantize Deep Learning Models to Minimize Footprint and Run on CPUs at GPU Speeds.

We are excited to announce that our portfolio of tools is available on GitHub, under the Deep Sparse Platform umbrella. View our webinar recording to learn about the four platform components that you can take advantage of immediately:

  1. Sparsify: Open-source, easy-to-use UI for automatically sparsifying neural networks and creating sparsification recipes for better inference performance and a smaller footprint
  2. SparseML: Open-source libraries for applying sparsification recipes to neural networks with a few lines of code, enabling faster and smaller models
  3. SparseZoo: Open-source neural network model repository for highly sparse and sparse-quantized models with matching sparsification recipes
  4. DeepSparse Engine: Free neural network inference engine that delivers GPU-class performance for sparsified models on CPUs

Date recorded: April 7, 2021
Speakers: Nir Shavit, Founder, Neural Magic & Mark Kurtz, ML Lead, Neural Magic