Hopfield Recurrent network (newhop)

Use neurolab.net.newhop(), neurolab.tool.simhop()

# -*- coding: utf-8 -*-
"""
Example of use Hopfield Recurrent network
=========================================

Task: Recognition of letters

"""

import numpy as np
import neurolab as nl

# N E R O
target =  [[1,0,0,0,1,
           1,1,0,0,1,
           1,0,1,0,1,
           1,0,0,1,1,
           1,0,0,0,1],
          [1,1,1,1,1,
           1,0,0,0,0,
           1,1,1,1,1,
           1,0,0,0,0,
           1,1,1,1,1],
          [1,1,1,1,0,
           1,0,0,0,1,
           1,1,1,1,0,
           1,0,0,1,0,
           1,0,0,0,1],
          [0,1,1,1,0,
           1,0,0,0,1,
           1,0,0,0,1,
           1,0,0,0,1,
           0,1,1,1,0]]

chars = ['N', 'E', 'R', 'O']
target = np.asfarray(target)
target[target == 0] = -1

# Create and train network
net = nl.net.newhop(target)

output = net.sim(target)
print("Test on train samples:")
for i in range(len(target)):
    print(chars[i], (output[i] == target[i]).all())

print("\nTest on defaced N:")
test =np.asfarray([0,0,0,0,0,
                   1,1,0,0,1,
                   1,1,0,0,1,
                   1,0,1,1,1,
                   0,0,0,1,1])
test[test==0] = -1
out = net.sim([test])
print ((out[0] == target[0]).all(), 'Sim. steps',len(net.layers[0].outs))

Result:

Test on train samples:
N True Sim. steps 1
E True Sim. steps 1
R True Sim. steps 1
O True Sim. steps 1

Test on defaced N:
True Sim. steps 2