Views
No views yet
config.yaml: The config file was originally downloaded from: https://github.com/CSSLab/maia-chess/blob/master/move_prediction/replication-configs/final/1800_final.yaml1from chessnets.pipelines.lc0 import Lc0Pipeline
2
3pipeline = Lc0Pipeline.from_pretrained("shermansiu/maia-1800")
4predictions = pipeline("1. e4 e5")config.json: canonical, fully materialized ChessNets architecture configmodel.safetensors: decoded float32 PyTorch tensorsprovenance.json: source and conversion hashesoriginal/model.pb.gz: original Lc0 protobuf checkpointtraining/original_lczero_training_config.yaml: original training configLINEAR16 encoding. Conversion decoded
those values into float32 tensors; it cannot restore precision already lost in
the source encoding.1@inproceedings{mcilroyyoung2020maia,
2 title={Aligning Superhuman AI with Human Behavior: Chess as a Model System},
3 author={McIlroy-Young, Reid and Sen, Siddhartha and Kleinberg, Jon and Anderson, Ashton},
4 booktitle={Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining},
5 year={2020},
6 doi={10.1145/3394486.3403219}
7}