KNN & DTW


In [1]:
# -*- coding: utf-8 -*-
class Dtw(object):
    
    def __init__(self, seq1, seq2,
                 patterns = [(-1,-1), (-1,0), (0,-1)], 
                 weights = [{(0,0):2}, {(0,0):1}, {(0,0):1}], 
                 band_r=0.005): #EDIT HERE
        self._seq1 = seq1
        self._seq2 = seq2
        self.len_seq1 = len(seq1)
        self.len_seq2 = len(seq2)
        self.len_pattern = len(patterns)
        self.sum_w = [sum(ws.values()) for ws in weights]
        self._r = int(len(seq1)*band_r)
        assert len(patterns) == len(weights)
        self._patterns = patterns
        self._weights = weights
    
    def get_distance(self, i1, i2):
        return abs(self._seq1[i1] - self._seq2[i2])

    def calculate(self):
        g = list([float('inf')]*self.len_seq2 for i in range(self.len_seq1))
        cost = list([0]*self.len_seq2 for i in range(self.len_seq1))

        g[0][0] = 2*self.get_distance(0, 0)
        for i in range(self.len_seq1):
            for j in range(max(0,i-self._r), min(i+self._r+1, self.len_seq2)):
                for pat_i in range(self.len_pattern):
                    coor = (i+self._patterns[pat_i][0], j+self._patterns[pat_i][1])
                    if coor[0]<0 or coor[1]<0:
                        continue
                    dist = 0
                    for w_coor_offset, d_w in self._weights[pat_i].items():
                        w_coor = (i+w_coor_offset[0], j+w_coor_offset[1])
                        dist += d_w*self.get_distance(w_coor[0], w_coor[1])
                    this_val = g[coor[0]][coor[1]] + dist
                    this_cost = cost[coor[0]][coor[1]] + self.sum_w[pat_i]
                    if this_val < g[i][j]:
                        g[i][j] = this_val
                        cost[i][j] = this_cost
        return g[self.len_seq1-1][self.len_seq2-1]/cost[self.len_seq1-1][self.len_seq2-1], g, cost
    
    def print_table(self, tb):
        print('      '+' '.join(["{:^7d}".format(i) for i in range(self.len_seq2)]))
        for i in range(self.len_seq1):
            str = "{:^4d}: ".format(i)
            for j in range(self.len_seq2):
                str += "{:^7.3f} ".format(tb[i][j])
            print (str)

    def print_g_matrix(self):
        _, tb, _ = self.calculate()
        self.print_table(tb)

    def print_cost_matrix(self):
        _, _, tb = self.calculate()
        self.print_table(tb)
        
    def get_dtw(self):
        ans, _, _ = self.calculate()
        return ans

In [2]:
import csv
import random
import math
import operator
import numpy as np

def loadDataset(filename, data=[]):
    with open(filename, 'rb') as csvfile:
        lines = csv.reader(csvfile,delimiter=' ')
        dataset = list(lines)
        for x in range(len(dataset)):
            dataset[x] = filter(None, dataset[x])
            dataset[x] = list(map(float, dataset[x]))
            data.append(dataset[x])

def euclideanDistance(instance1, instance2, length):
	distance = 0
	for x in range(length):
		if x == 0:
			continue
		distance += pow((instance1[x] - instance2[x]), 2)
	return math.sqrt(distance)
 
def getNeighbors(trainingSet, testInstance, k, pattern, weight, r_band=None):
	distances = []
	length = len(testInstance)
	for x in range(len(trainingSet)):
#  z-normalization
		d = Dtw(testInstance[1:], trainingSet[x][1:], pattern, weight, r_band)
		dist = d.get_dtw()
# 		dist = euclideanDistance(testInstance, trainingSet[x], length)
		distances.append((trainingSet[x], dist))
	distances.sort(key=operator.itemgetter(1))
#  	print "dist >>>> ",distances
	neighbors = []
	for x in range(k):
		neighbors.append(distances[x][0])
	return neighbors

def getResponse(neighbors):
	classVotes = {}
	for x in range(len(neighbors)):
		response = neighbors[x][0]
		if response in classVotes:
			classVotes[response] += 1
		else:
			classVotes[response] = 1
	sortedVotes = sorted(classVotes.iteritems(), key=operator.itemgetter(1), reverse=True)
	return sortedVotes[0][0]
 
def getAccuracy(testSet, predictions):
	correct = 0
	for x in range(len(testSet)):
		if testSet[x][0] == predictions[x]:
			correct += 1
	return (correct/float(len(testSet))) * 100.0
	
def knn(trainingSet, testSet, k, pattern, weight, r_band=None):
	# generate predictions
	predictions=[]
	for x in range(len(testSet)):
# 		print ">>",testSet[x]
		neighbors = getNeighbors(trainingSet, testSet[x], k, pattern, weight, r_band)
# 		print "neighbors >>", neighbors
		result = getResponse(neighbors)
# 		print "result >>", result
		predictions.append(result)
# 		print('> predicted=' + repr(result) + ', actual=' + repr(testSet[x][0]))
	accuracy = getAccuracy(testSet, predictions)
	return accuracy

def prepareData(train_data, test_data):
	# prepare data
	rawTrainingSet=[]
	rawTestSet=[]
	testSet=[]
	trainingSet=[]
	loadDataset(train_data, rawTrainingSet)
	loadDataset(test_data, rawTestSet)
	for x in rawTrainingSet:
		newTS = np.append(x[0], ( np.array(x[1:])-np.mean(x[1:]) )/np.std(x[1:]) )
		trainingSet.append(newTS)
	for x in rawTestSet:
		newTS = np.append(x[0], ( np.array(x[1:])-np.mean(x[1:]) )/np.std(x[1:]) )
		testSet.append(newTS)
# 	print 'Train set: ' + repr(len(trainingSet))
# 	print trainingSet
# 	print 'Test set: ' + repr(len(testSet))
# 	print testSet
	return trainingSet, testSet

Main


In [3]:
# EDIT HERE
TRAIN_DATA = 'dataset/Beef_TRAIN'
TEST_DATA = 'dataset/Beef_TEST'
OUTPUT_FILE = 'acc_SHAPE_Beef.csv'

In [4]:
trainingSet, testSet = prepareData(TRAIN_DATA, TEST_DATA)
with open(OUTPUT_FILE, "w") as myfile:
    myfile.write("pattern_id,p,r_band_size,accuracy\n")

Pattern 1 (Symmetric) P=0


In [5]:
PATTERNS_1 = [(0,-1), (-1,-1), (-1,0)]
WEIGHTS_SYM_1 = [{(0,0):1}, {(0,0):2}, {(0,0):1}]

In [6]:
acc = knn(trainingSet, testSet, 1, PATTERNS_1, WEIGHTS_SYM_1, 0.005)
print "Pattern#1 R-band=0.01 (1%) acc >",acc
with open(OUTPUT_FILE, "a") as myfile:
    myfile.write("1,0,0.01,"+str(acc)+"\n")


Pattern#1 R-band=0.01 (1%) acc > 63.3333333333

In [7]:
acc = knn(trainingSet, testSet, 1, PATTERNS_1, WEIGHTS_SYM_1, 0.015)
print "Pattern#1 R-band=0.03 (3%) acc >",acc
with open(OUTPUT_FILE, "a") as myfile:
    myfile.write("1,0,0.03,"+str(acc)+"\n")


Pattern#1 R-band=0.03 (3%) acc > 70.0

In [8]:
acc = knn(trainingSet, testSet, 1, PATTERNS_1, WEIGHTS_SYM_1, 0.025)
print "Pattern#1 R-band=0.05 (5%) acc >",acc
with open(OUTPUT_FILE, "a") as myfile:
    myfile.write("1,0,0.05,"+str(acc)+"\n")


Pattern#1 R-band=0.05 (5%) acc > 60.0

Pattern 2 (Symmetric) P=1/2


In [9]:
PATTERNS_2 = [(-1,-3), (-1,-2), (-1,-1), (-2,-1), (-3,-1)]
WEIGHTS_SYM_2 = [{(0,-2):2, (0,-1):1, (0,0):1}, \
                 {(0,-1):2, (0,0):1}, \
                 {(0,0):2}, \
                 {(-1,0):2, (0,0):1}, \
                 {(-2,0):2, (-1,0):1, (0,0):1}]

In [10]:
acc = knn(trainingSet, testSet, 1, PATTERNS_2, WEIGHTS_SYM_2, 0.005)
print "Pattern#2 R-band=0.01 (1%) acc >",acc
with open(OUTPUT_FILE, "a") as myfile:
    myfile.write("2,1/2,0.01,"+str(acc)+"\n")


Pattern#2 R-band=0.01 (1%) acc > 63.3333333333

In [11]:
acc = knn(trainingSet, testSet, 1, PATTERNS_2, WEIGHTS_SYM_2, 0.015)
print "Pattern#2 R-band=0.03 (3%) acc >",acc
with open(OUTPUT_FILE, "a") as myfile:
    myfile.write("2,1/2,0.03,"+str(acc)+"\n")


Pattern#2 R-band=0.03 (3%) acc > 60.0

In [12]:
acc = knn(trainingSet, testSet, 1, PATTERNS_2, WEIGHTS_SYM_2, 0.025)
print "Pattern#2 R-band=0.05 (5%) acc >",acc
with open(OUTPUT_FILE, "a") as myfile:
    myfile.write("2,1/2,0.05,"+str(acc)+"\n")


Pattern#2 R-band=0.05 (5%) acc > 63.3333333333

Pattern 3 (Symmetric) P=1


In [13]:
PATTERNS_3 = [(-1,-2), (-1,-1), (-2,-1)]
WEIGHTS_SYM_3 = [{(0,-1):2, (0,0):1}, \
                 {(0,0):2}, \
                 {(-1,0):2, (0,0):1}]

In [14]:
acc = knn(trainingSet, testSet, 1, PATTERNS_3, WEIGHTS_SYM_3, 0.005)
print "Pattern#3 R-band=0.01 (1%) acc >",acc
with open(OUTPUT_FILE, "a") as myfile:
    myfile.write("3,1,0.01,"+str(acc)+"\n")


Pattern#3 R-band=0.01 (1%) acc > 63.3333333333

In [15]:
acc = knn(trainingSet, testSet, 1, PATTERNS_3, WEIGHTS_SYM_3, 0.015)
print "Pattern#3 R-band=0.03 (3%) acc >",acc
with open(OUTPUT_FILE, "a") as myfile:
    myfile.write("3,1,0.03,"+str(acc)+"\n")


Pattern#3 R-band=0.03 (3%) acc > 63.3333333333

In [16]:
acc = knn(trainingSet, testSet, 1, PATTERNS_3, WEIGHTS_SYM_3, 0.025)
print "Pattern#3 R-band=0.05 (5%) acc >",acc
with open(OUTPUT_FILE, "a") as myfile:
    myfile.write("3,1,0.05,"+str(acc)+"\n")


Pattern#3 R-band=0.05 (5%) acc > 63.3333333333

Pattern 4 (Symmetric) P=2


In [21]:
PATTERNS_4 = [(-2,-3), (-1,-1), (-3,-2)]
WEIGHTS_SYM_4 = [{(-1,-2):2, (0,-1):2, (0,0):1}, \
                 {(0,0):2}, \
                 {(-2,-1):2, (-1,0):2, (0,0):1}]

In [22]:
acc = knn(trainingSet, testSet, 1, PATTERNS_3, WEIGHTS_SYM_3, 0.005)
print "Pattern#4 R-band=0.01 (1%) acc >",acc
with open(OUTPUT_FILE, "a") as myfile:
    myfile.write("4,2,0.01,"+str(acc)+"\n")


Pattern#4 R-band=0.01 (1%) acc > 63.3333333333

In [23]:
acc = knn(trainingSet, testSet, 1, PATTERNS_3, WEIGHTS_SYM_3, 0.015)
print "Pattern#4 R-band=0.03 (3%) acc >",acc
with open(OUTPUT_FILE, "a") as myfile:
    myfile.write("4,2,0.03,"+str(acc)+"\n")


Pattern#4 R-band=0.03 (3%) acc > 63.3333333333

In [24]:
acc = knn(trainingSet, testSet, 1, PATTERNS_3, WEIGHTS_SYM_3, 0.025)
print "Pattern#4 R-band=0.05 (5%) acc >",acc
with open(OUTPUT_FILE, "a") as myfile:
    myfile.write("4,2,0.05,"+str(acc)+"\n")


Pattern#4 R-band=0.05 (5%) acc > 63.3333333333