7c9508370c84661026ac66f86ddc7c91644fedc177fa718b2029cb2e53d0cfa8  README.md
d9341e9608025b26986a216b242f6cddfcf4cd1ad0982e7d1ed4c748f44e9caa  cotangent-semantics.tex
13ef4788476d292858fa60eb9a5f74aeca5c65770bc885ccaa05823a17ef7be1  data/Fashion-MNIST-LICENSE
742a30401f4779e7136cff1e6d66f98ebd82f1c0f6f862131b19a97b005936a5  data/prepared/manifest.json
b8563ea2bbfb2ca71a8637f781e000a78a9df557047f4d3494ab7eedbb6da2b4  data/prepared/node-1.dataset.gz
4dbcbc46308d329a285ff7a4f6f137b0026abe3ceb309b830bb9c3018ca5d9ba  data/prepared/node-2.dataset.gz
cf92c0a03ac8466003c5b83764e37fd1fbed7b582562352e5d267c85a8108e9e  data/prepared/node-3.dataset.gz
974e3c3d05e165c13fa54dfd59e6a58c6795f68fa29365ecb838d6c2a6abc3f4  inputs.json
6ea6040d3aace3858d6f70eea9424f7f91f4561833ac12a79af6b233ffe85281  manuscript/build_source.py
633e2a49d920a0efdb5a0bc628e7ee0dc24376b3d0207e84a182f53975458522  manuscript/journal-body.md
c7d72603587100788aaf560d4fe43fcfa1fc9bc13c799d2ba40b3e2db11cefc1  manuscript/paper.tex.in
11a5b4da97d4e9df7c4bde5da94e86e08300f6cbf76d0a17fd8d1c258c644ea1  reference/first/manifest.json
4388598222f29c40e9fc311c7d82a59b0e82246faac51e40b2a71904bccd173a  reference/first/node-1.log
c80a04f3a4a5e0f6c275d5d8fdfd66757293fd3e169d097eeb3135bed0c6219b  reference/first/node-2.log
73d8fa2a806bc46a6ba85bfc076ab9546905aa34b058885a9c9b8625b0032cef  reference/first/node-3.log
554339fa17ec25127e4a1ee750f010206785e2195997fc5a58ddd7b8ff3feb1b  reference/replication/manifest.json
cf3c5a7632172a6f2294e0a4db5a95d232d75470d965c28aaaa8859c416a0a9f  reference/replication/node-1.log
f88d4c61bb162b9a999a9f3747796c1564d117b5a8fe008de3eb342b2569aa7e  reference/replication/node-2.log
f860f3f19783605ce26ac97102fc0768ec9960c8cd4279b5dd5646622b9a89c0  reference/replication/node-3.log
1e84d5e1a16872f293e01de2113a18bf0a359c9db8a9dda58d41ca8b0a9d5cd2  reference/tables/accuracy.tex
d7807aeb01b7f209a7d92f97935645d31350ffd318f2d326632d384c15aa764d  reference/tables/audits.csv
10aa7ecbcf93e8ad267de74512aaf9bcad7d163b3bbc08fc47b62998264280bf  reference/tables/loss.tex
16dfc12fa8079de5595c27ff6ab6677758422e8300123ea9546a775dca476c33  reference/tables/numbers.tex
4084428eb44d631d6d9d51c7fda121f7bf92d28bc02f28703e7c0a40cfba9d99  reference/tables/summary.json
c9db79af2ce0eb0311158b311356d1a2dfe9f4b499a3fbb291d8c7b0588d4250  reference/tables/tests.csv
dbc186895af712fe54c19beda6c1b33629ad298ad2aca2aa70994e56323e7f07  reference/tables/traffic.tex
c3a6c7626b293b4dc82e3b77aa79883944bdb9373dbf16dabe9087ac5b46153c  reference/tables/trials.csv
e790b58283764096cdd79555f5e46e84c82149d2fb1506373fe68c31047c095c  run.py
2825128f499df7ccc00724f1964943d536c37165cbab8724ed7cf2af0cb14c3c  src/experiment.c
b38e4eead24e0d3f47d67f5fe6dffda3db1c4109c30bb73d4be1d5427694fb86  src/prepare.py
bebecd415e9f0622b7995cb828eef97854f1f669de3de4eba693b432f3ca5af4  verify.py
