Skip to content

A deep neural network which uses column vectors rather than row vectors, making backpropagation a little more unique

License

Notifications You must be signed in to change notification settings

magmawolf8/columnDNN

Repository files navigation

columnDNN: a slightly different approach to a generic ML architecture

This neural network left-multiplies vectors to achieve outputs, meaning that the calculated gradient during backpropagation is also a column vector.

About

A deep neural network which uses column vectors rather than row vectors, making backpropagation a little more unique

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages