Want to prune your
#ML
models at higher levels without impacting accuracy? ✂️
Join us for a virtual session 📺 on April 6 where we'll discuss second-order pruning methods that enable higher sparsity by removing weights that directly affect the loss function the least.
1/3
The result of using second-order pruning methods is sparse models with smaller files, lower latency, and higher throughput.
Example: ResNet-50 can be pruned 95% and still maintain 99% of its baseline accuracy, all while decreasing its file size 9.7X, from 90.8 MB to 9.3 MB.
2/3
On April 6th,
@markurtz_
and
@_EldarKurtic
will hold a live 30-minute webinar (plus Q&A!), covering this SOTA model compression research and teaching you how to apply it to your current ML projects!
Let us know you are coming:
3/3