neural-networks-master.zip

#!/usr/bin/env python # ----------------------------------------------------------------------------- # Adaptive Resonance Theory # Copyright (C) 2011 Nicolas P. Rougier # # Distributed under the terms of the BSD License. # ----------------------------------------------------------------------------- # Reference: Grossberg, S. (1987) # Competitive learning: From interactive activation to # adaptive resonance, Cognitive Science, 11, 23-63 # # Requirements: python 2.5 or above => http://www.python.org # numpy 1.0 or above => http://numpy.scipy.org # ----------------------------------------------------------------------------- from __future__ import print_function from __future__ import division import numpy as np class ART: ''' ART class Usage example: -------------- # Create a ART network with input of size 5 and 20 internal units >>> network = ART(5,10,0.5) ''' def __init__(self, n=5, m=10, rho=.5): ''' Create network with specified shape Parameters: ----------- n : int Size of input m : int Maximum number of internal units rho : float Vigilance parameter ''' # Comparison layer self.F1 = np.ones(n) # Recognition layer self.F2 = np.ones(m) # Feed-forward weights self.Wf = np.random.random((m,n)) # Feed-back weights self.Wb = np.random.random((n,m)) # Vigilance self.rho = rho # Number of active units in F2 self.active = 0 def learn(self, X): ''' Learn X ''' # Compute F2 output and sort them (I) self.F2[...] = np.dot(self.Wf, X) I = np.argsort(self.F2[:self.active].ravel())[::-1] for i in I: # Check if nearest memory is above the vigilance level d = (self.Wb[:,i]*X).sum()/X.sum() if d >= self.rho: # Learn data self.Wb[:,i] *= X self.Wf[i,:] = self.Wb[:,i]/(0.5+self.Wb[:,i].sum()) return self.Wb[:,i], i # No match found, increase the number of active units # and make the newly active unit to learn data if self.active < self.F2.size: i = self.active self.Wb[:,i] *= X self.Wf[i,:] = self.Wb[:,i]/(0.5+self.Wb[:,i].sum()) self.active += 1 return self.Wb[:,i], i return None,None # ----------------------------------------------------------------------------- if __name__ == '__main__': np.random.seed(1) # Example 1 : very simple data # ------------------------------------------------------------------------- network = ART( 5, 10, rho=0.5) data = [" O ", " O O", " O", " O O", " O", " O O", " O", " OO O", " OO ", " OO O", " OO ", "OOO ", "OO ", "O ", "OO ", "OOO ", "OOOO ", "OOOOO", "O ", " O ", " O ", " O ", " O", " O O", " OO O", " OO ", "OOO ", "OO ", "OOOO ", "OOOOO"] X = np.zeros(len(data[0])) for i in range(len(data)): for j in range(len(data[i])): X[j] = (data[i][j] == 'O') Z, k = network.learn(X) print("|%s|"%data[i],"-> class", k) # Example 2 : Learning letters # ------------------------------------------------------------------------- def letter_to_array(letter): ''' Convert a letter to a numpy array ''' shape = len(letter), len(letter[0]) Z = np.zeros(shape, dtype=int) for row in range(Z.shape[0]): for column in range(Z.shape[1]): if letter[row][column] == '#': Z[row][column] = 1 return Z def print_letter(Z): ''' Print an array as if it was a letter''' for row in range(Z.shape[0]): for col in range(Z.shape[1]): if Z[row,col]: print( '#', end="" ) else: print( ' ', end="" ) print( ) A = letter_to_array( [' #### ', '# #', '# #', '######', '# #', '# #', '# #'] ) B = letter_to_array( ['##### ', '# #', '# #', '##### ', '# #', '# #', '##### '] ) C = letter_to_array( [' #### ', '# #', '# ', '# ', '# ', '# #', ' #### '] ) D = letter_to_array( ['##### ', '# #', '# #', '# #', '# #', '# #', '##### '] ) E = letter_to_array( ['######', '# ', '# ', '#### ', '# ', '# ', '######'] ) F = letter_to_array( ['######', '# ', '# ', '#### ', '# ', '# ', '# '] ) samples = [A,B,C,D,E,F] network = ART( 6*7, 10, rho=0.15 ) for i in range(len(samples)): Z, k = network.learn(samples[i].ravel()) print("%c"%(ord('A')+i),"-> class",k) print_letter(Z.reshape(7,6))