-
Notifications
You must be signed in to change notification settings - Fork 0
/
Knn.py
151 lines (119 loc) · 3.7 KB
/
Knn.py
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
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
# -*- coding: utf-8 -*-
"""
Created on Sat Jun 19 12:13:07 2021
@author: RITAM
"""
from python_speech_features import mfcc
import scipy.io.wavfile as wav
import numpy as np
from tempfile import TemporaryFile
import os
import pickle
import random
import operator
from sklearn.metrics import confusion_matrix,classification_report
import math
import matplotlib.pyplot as plt
import seaborn as sns
# function to get the distance between feature vecotrs and find neighbors
def getNeighbors(trainingSet, instance, k):
distances = []
for x in range (len(trainingSet)):
dist = distance(trainingSet[x], instance, k) + distance(instance, trainingSet[x], k)
distances.append((trainingSet[x][2], dist))
distances.sort(key=operator.itemgetter(1))
neighbors = []
for x in range(k):
neighbors.append(distances[x][0])
return neighbors
#%%
# identify the class of the instance
def nearestClass(neighbors):
classVote = {}
for x in range(len(neighbors)):
response = neighbors[x]
if response in classVote:
classVote[response] += 1
else:
classVote[response] = 1
sorter = sorted(classVote.items(), key = operator.itemgetter(1), reverse=True)
return sorter[0][0]
# function to evaluate the model
def getAccuracy(testSet, prediction):
correct = 0
for x in range(len(testSet)):
if testSet[x][-1] == predictions[x]:
correct += 1
return (1.0 * correct) / len(testSet)
#%%
# directory that holds the dataset
directory = "./genres/"
f = open("my.dat", 'wb')
i = 0
for folder in os.listdir(directory):
i += 1
if i == 11:
break
for file in os.listdir(directory+folder):
(rate, sig) = wav.read(directory+folder+"/"+file)
mfcc_feat = mfcc(sig, rate, winlen=0.020, appendEnergy=False)
covariance = np.cov(np.matrix.transpose(mfcc_feat))
mean_matrix = mfcc_feat.mean(0)
feature = (mean_matrix, covariance, i)
pickle.dump(feature, f)
f.close()
#%%
dataset = []
def loadDataset(filename, split,trSet, teSet):
with open("my.dat", 'rb') as f:
while True:
try:
dataset.append(pickle.load(f))
except EOFError:
f.close()
break
for x in range(len(dataset)):
if random.random() < split:
trSet.append(dataset[x])
else:
teSet.append(dataset[x])
trainingSet = []
testSet = []
loadDataset("my.dat", 0.75, trainingSet, testSet)
#%%
def distance(instance1 , instance2 , k ):
distance = 0
mm1 = instance1[0]
cm1 = instance1[1]
mm2 = instance2[0]
cm2 = instance2[1]
distance = np.trace(np.dot(np.linalg.inv(cm2), cm1))
distance+=(np.dot(np.dot((mm2-mm1).transpose() , np.linalg.inv(cm2)) , mm2-mm1 ))
distance+= np.log(np.linalg.det(cm2)) - np.log(np.linalg.det(cm1))
distance-= k
return distance
#%%
# making predictions using KNN
leng = len(testSet)
predictions = []
for x in range(leng):
predictions.append(nearestClass(getNeighbors(trainingSet, testSet[x], 3)))
accuracy1 = getAccuracy(testSet, predictions)
print(accuracy1)
#%%
y_true = []
y_pred = np.array(predictions)
for i in testSet:
y_true.append(i[-1])
y_true = np.array(y_true)
cm = confusion_matrix(y_true,y_pred)
print(cm)
sns.heatmap(cm,annot=True)
plt.title('confution matrix plot using KNN')
print("\nconfution matrix plot")
plt.xlabel('True Label')
plt.ylabel('Predicted Label')
plt.legend()
plt.show()
plt.savefig('cnn_cm_ritam.png')
print(classification_report(y_true,y_pred))