Skip to content

Added Accumulated Gradient Scoring

Choose a tag to compare

@verbiiyo verbiiyo released this 27 Oct 12:08
· 58 commits to master since this release
159f1e5

Added the ability to accumulate gradients to simulate larger batch sizes for RigL steps. Larger batch sizes tend to reduce noise & the aim is to emulate a batch size of 4096 using a true batch size of 1024.