LeCun 1989 Repro. Running a landmark 1989 neural network experiment on modern computers.

github.com/karpathy/lecun1989-repro

Problem

The full read on this project is a sign-in away — what it solves, who it serves, and where it sits in the market.

Audience

The full read on this project is a sign-in away — what it solves, who it serves, and where it sits in the market.

Market

The full read on this project is a sign-in away — what it solves, who it serves, and where it sits in the market.

Pricing

The full read on this project is a sign-in away — what it solves, who it serves, and where it sits in the market.

Vaya's read on this project

Problem, audience, market, and the verdict — sign in to see it.

Updates

March 2022

Merge pull request #5 from BioGeek/literal_asterisks Mar 20
Escape asterisks in Markdown with backslash Mar 19
improve on using time travel knowledge Mar 11
fix the indentation because ocd is killing me Mar 11
add teaser image to repo Mar 11
keep track of model complexity: params, MACs, activations Mar 11
refactor the init out into a helper function Mar 11
i am having weird issues with reproducibility, getting slightly diffe… Mar 11
add jupyter notebook showing the misprediction examples Mar 11
sequester script code and save model to disk Mar 11
Merge pull request #1 from oz1127/patch-1 Mar 11
Update README.md Mar 10
rename train to repro, so we can repro after prepro lolol Mar 10
switch to tensorboardX because tensorboard doesn't seem to install pr… Mar 10
clarify the computational setup Mar 10
small tweaks to docs Mar 10
first pass at repro Mar 10
Initial commit Mar 10