Skip to content

A simple homebrew neural network created for MTE 203.

Notifications You must be signed in to change notification settings

lucs100/soplenet

Repository files navigation

soplenet

A simple 4-layer feedforward neural network created for MTE 203.

Built to classify images from the CIFAR-10 dataset into one of 10 categories. Uses a cross-entropy classification loss function to backpropagate error across layers, and the mini-batch stochastic gradient descent method to apply this error to the system. Logging can be done at the epoch, mini-batch, or individual test case level.

About

A simple homebrew neural network created for MTE 203.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages