A simple way to make neural nets: Machine learning without linear algebra
Project description
Simple Neural Net Module
Install requirement:
pip install numpy
Install module:
pip install SiNN
or download module.py
from GitHub.
Quick-Start Guide
Import SiNN: import SiNN
Initialize the neural net:
neuralnet = SiNN.NeuralNetwork(3) # 3 is the number of inputs
Create a variable with training set inputs:
ins = array([[1a, 1b, 1c], [2a, 2b, 2c], [3a, 3b, 3c]])
Set the expected outcomes (training set outs):
outs = array([[1,1,0]]).T # don't worry about the .T
Train with neuralnet.train(ins, outs, iters)
, where iters is the amount of training cycles. A number around 1000 is normally good for simple uses.
Then, see if it works with neuralnet.think([a,b,c])
.
Present it with a new situation with neuralnet.think(newsit)
Note: use python 3 with this.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.