MaleCNS v1.0
Janelia Research Campus, HHMI and Google Research.
https://male-cns.janelia.org/
License: Creative Commons Attribution 4.0 International.
Source: significant-only flat connectome at minconf 0.5 and body annotations.
Modification: selected ALPN to KC and KC to MBON connections, reindexed cells,
serialized annotated identities and unchanged selected synapse counts to JSON.
The model normalization and learned gains are engineered transformations.

Optical Recognition of Handwritten Digits
Alpaydin, E. and Kaynak, C. (1998).
UCI Machine Learning Repository. DOI: 10.24432/C50P49.
https://doi.org/10.24432/C50P49
License: Creative Commons Attribution 4.0 International.
Modification: original optdigits.tra and optdigits.tes rows serialized as JSON.
The official train/test partitions and numeric image values are preserved.
Runtime inputs divide block counts by 16 and assign pixels to artificial ports.

CC BY 4.0 terms: https://creativecommons.org/licenses/by/4.0/
No endorsement by the data creators is implied.

Original GFly Learning Lab implementation and manuscript:
Jiahe Tian, HiOS Labs Inc. Correspondence: ipo@stanford.edu.
