-
Notifications
You must be signed in to change notification settings - Fork 0
a neural network visualization
License
mdonoughe/homework2
Folders and files
| Name | Name | Last commit message | Last commit date | |
|---|---|---|---|---|
<
8000
/div> | ||||
Repository files navigation
Homework 2 A neural network for visually classifying points in a two dimensional input space into two classes, posted because the visualization may be useful. A precompiled version is available in the downloads section. The input files are in the format "X\tY\tTARGET" where X and Y are floating point values between -10 and 10, and TARGET is 0 or 1. There is no testing set input. The only output of this program is the error at each epoch and a graphical view of the network. main.c is responsible for the rendering loop. shaderbuilder.c builds a fragment shader that classifies each point based on it's texture coordinate and the weights array. nn.c builds and trains an mlp or rbf. Usage is "homework2 [-f <file.dat>] [-r <rbf layer size>] <hidden layer sizes>". The arguments must be provided in this order! I didn't bother with much error handling so this will probably crash if something goes wrong. Works fine for me on my computer so long as the specified file exists and I don't do anything like make an rbf without an rbf layer.
About
a neural network visualization
Resources
License
Stars
Watchers
Forks
Packages 0
No packages published