Description
Write a program that reads a file whose name is the single command-line
argument. The file contains a description of a multi-layer neural network.
Build the described network and perform forward propagation through all
layers and then print the outputs of the neurons of the last layer.
An annotated example file is in p1IN0.
Our description file does not tell which activation/transfer function
should be used for each neuron. We will use the simple sigmoid.
https://en.wikipedia.org/wiki/Activation_function
sigmoid = Logistic function
f(x) = 1 / (1 + e^(-x))
Use turnin to submit a tar file containing all of your project files,
including a makefile that will build the executable program which MUST
be named p1.
To build the project, I will cd to my directory containing your files
and simply type:
rm -rf p1
rm -f *.o
make
all NNs are fully connected, including from inputs
bias value – connects to the individual neuron
left: x1*w1 + x2*w2
right: bias + left
out: sigmoid(R)
for p1IN0:
left: 1.0*1.0 + 1.0*1.0 = 2.0
right: 1.0 + 2.0
out: sigmoid(3.0)
