# Attribution

These weights are derived from, and are only meaningful together with, the following sources.

## Connectome (network wiring, synapse signs, neuron positions, retina columns)

FlyWire whole-brain connectome of an adult female *Drosophila melanogaster*, Codex snapshot 783
(`connections.csv.gz`, `neurons.csv.gz`, `classification.csv.gz`, `consolidated_cell_types.csv.gz`,
`coordinates.csv.gz`, `column_assignment.csv.gz`). Licence of the Codex release: **CC BY-NC 4.0**
(https://flywire.ai/guidelines). The Zenodo mirror of the connectivity data is CC BY 4.0
(doi:10.5281/zenodo.10676866).

* Dorkenwald, S., Matsliah, A., Sterling, A. R., Schlegel, P., Yu, S.-C., McKellar, C. E., et al.
  Neuronal wiring diagram of an adult brain. *Nature* 634, 124–138 (2024). doi:10.1038/s41586-024-07558-y
* Schlegel, P., Yin, Y., Bates, A. S., et al. Whole-brain annotation and multi-connectome cell typing of
  *Drosophila*. *Nature* 634, 139–152 (2024). doi:10.1038/s41586-024-07686-5
* Matsliah, A., Yu, S.-C., Kruk, K., et al. Neuronal parts list and wiring diagram for a visual system.
  *Nature* 634, 166–180 (2024). doi:10.1038/s41586-024-07981-1 (optic-lobe column assignments)
* Eckstein, N., Bates, A. S., Champion, A., et al. Neurotransmitter classification from electron microscopy
  images at synaptic sites in *Drosophila melanogaster*. *Cell* 187, 2574–2594 (2024) (transmitter predictions
  behind the synapse signs)
* The FlyWire Consortium. FlyWire Whole-brain Connectome Connectivity Data (783.0) [Data set]. Zenodo (2024).

## Training data

* Lichess open database, https://database.lichess.org — monthly standard rated games 2014-01 … 2015-12 and
  the evaluation database (`lichess_db_eval.jsonl.zst`). Released under CC0 1.0.
* Stockfish (https://stockfishchess.org, GPL-3.0) produced the evaluations in the Lichess evaluation database;
  no Stockfish code is included here.

## Modelling ideas

* Lappalainen, J. K., et al. Connectome-constrained networks predict neural activity across the fly visual
  system. *Nature* 634, 1132–1140 (2024) — connectome-constrained training with sign constraints.
* Shiu, P. K., et al. A *Drosophila* computational brain model reveals sensorimotor processing.
  *Nature* 634, 210–219 (2024) — whole-FlyWire-brain leaky integrate-and-fire modelling.

## Software

* Training and inference code: https://github.com/cesp99/fly-chess (MIT, Carlo Esposito).
* PyTorch (BSD-3), python-chess (GPL-3.0, used as a library by the training code only; not part of the weights).