Name | Modified | Size | Downloads / Week |
---|---|---|---|
Parent folder | |||
README.md | 2021-04-27 | 584 Bytes | |
v0.5.0.tar.gz | 2021-04-27 | 6.9 MB | |
v0.5.0.zip | 2021-04-27 | 7.0 MB | |
Totals: 3 Items | 13.8 MB | 0 |
AlphaZero v0.5.0
- Improved the inference server so that it is now possible to keep MCTS workers
running while a batch of requests is being processed by the GPU. Concretely,
this translates into
SimParams
now having two separatenum_workers
andbatch_size
parameters. - The inference server is now spawned on a separate thread to ensure minimal latency.
Together, the two aforementioned improvements result in a 30% global speedup on the connect-four benchmark.