Edit on GitHub
Networks and Runs
What are networks and runs?
Over the two years, people inside and outside of LCZero team produced hundreds of thousand of neural networks. This page attempts to clarify what’s going on. What defines a network To have a trained network, you need to think about three aspects:
Actual networks
In general, for game analysis and long calculation time per move, the largest network compatible with your hardware is recommended. In use cases with very low calculation time per move or slow hardware, a smaller network might be a better choice.