TimeSeriesClassification/fuzzyTSC.py at master · udayRage/TimeSeriesClassification · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
#import pandas as pd
import numpy as np
#from _collections import defaultdict
#import matplotlib.pyplot as plt
import time
import os
import psutil
import gc
import sys

start_time = time.time()
training = np.loadtxt(sys.argv[1], delimiter='\t')
testing = np.loadtxt(sys.argv[2],  delimiter='\t')


mean_data = {} #defaultdict()
min_data = {}#defaultdict()
max_data= {}#defaultdict()


# seperating classes into different dictionary variables and creating mean, min and max curves
for i in (np.unique(training[:,0])):
    max_data[i] = np.amax(training[training[:,0] == i], axis = 0)
    mean_data[i] = np.mean(training[training[:,0] == i], axis= 0)
    min_data[i]= np.amin(training[training[:,0] == i], axis=0)

del training
gc.collect()


# Classification Phase

counter = {}#defaultdict()
num_rows, num_columns = testing.shape
correct = 0
#mem = defaultdict()
#start_time = time.time()
for i in range(num_rows):
    for k in (np.unique(testing[:,0])):
        counter[k] = 0.0
        for j in range(num_columns-1):
            if testing[i][j+1] >= mean_data[k][j+1]:
                if testing[i][j+1]<= max_data[k][j+1]:
                    counter[k] = counter[k] + 0.5* (testing[i][j+1] - mean_data[k][j+1]) / (max_data[k][j+1] - mean_data[k][j+1])
                else:
                    counter[k] = counter[k]+ 1
            else:
                if testing[i][j+1] >= min_data[k][j+1]:
                    counter[k] = counter[k] + 0.5 * (mean_data[k][j+1] - testing[i][j+1]) / (mean_data[k][j+1] - min_data[k][j+1])
                else:
                    counter[k] = counter[k]+1
        #counter[k] = counter[k]/num_columns-1

    for k in (np.unique(testing[:,0])):
        counter[k] = (counter[k]) / num_columns-1

    if min(counter, key=counter.get) == testing[i][0]:
        correct = correct + 1
accuracy = (correct / num_rows) * 100
#gc.collect()
print("Dataset name:", sys.argv[1])
print("Total Accuracy of proposedAlgo is:", accuracy)

print("Total Execution time of proposedAlgo", time.time() - start_time)
    # print(testing[i,0])
process = psutil.Process(os.getpid())
memory = process.memory_full_info().uss
memory_in_KB = memory / (1024 )
print("Memory of proposedAlgo in KB:",memory_in_KB)  # in bytes