PLEASE HELP IN PYTHON

I$\text{'}$m having a problem in my perceptron code to plot the linear decision line boundary.

I have my patterns as P $=$ np$.$array$([[1,0,0,1],[1,0,1,0],[1,1,1,1]])$

I have my targets as T $=$ np$.$array$([1,0,0,0]).$

And I have my weights as W $=$ np$.$array$([0\mathrm{.}041,-0\mathrm{.}7,0\mathrm{.}075])$ but it can be any random number for the first two components with third one being the bias.

The red dots on the plot are represented as class $0$ but the blue dot are represented as class $1.$

plt$.$plot$([0,0,1],[0,1,0],\text{'}$ro$\text{'})$

plt$.$plot$([1],[1],\text{'}$bo$\text{'})$

This is how I train the network.

def hardlim$($x$)$:

return np$.$where$($x $>=0,1,0)$

A $=$ hardlim$($np$.$dot$($W$,$P$))$

E $=$ T$-$A

rate $=0\mathrm{.}1$

while sum$($E$)!=0$:

W $=$ W $+$ rate $*$ np$.$dot$($E$,$ P$.$T$)$

A $=$ hardlim$($np$.$dot$($W$,$ P$))$

E $=$ T$-$A

How do I plot the linear decision boundary line so the network solves it flawselly. I need it so the blue dot is separated from the red dots with a linear line in the plot.