chordify/evaluate.py at master · halfcurry/chordify · 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
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
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
from sklearn.model_selection import train_test_split
from sklearn.metrics import accuracy_score
import numpy as np

def get_representative_vectors(X,Y):

	labels_to_data={}
	labels_to_Zs={}

	for i in range(len(X)):
		if Y[i] in labels_to_data:
			labels_to_data[Y[i]]=np.vstack((labels_to_data[Y[i]],X[i]))
		else:
			labels_to_data[Y[i]]=X[i]

	for label in labels_to_data:
		# print("Shape of label vector: ", labels_to_data[label].shape)
		# labels_to_data[label].reshape(-1)
		if(labels_to_data[label].ndim > 1):
			labels_to_Zs[label]=np.mean(labels_to_data[label],axis=0)
			# print(labels_to_Zs[label].shape)
		else:
			labels_to_Zs[label]=labels_to_data[label]

	return labels_to_Zs

def get_reverse_lookup(dic):

	rev_lookup={}
	ctr=0
	for key in dic:
		rev_lookup[ctr]=key
		ctr+=1

	return rev_lookup


def evaluate_a_song(X,Y):

	correct=0

	X_train, X_test, y_train, y_test = train_test_split(
	X, Y, test_size=0.3, random_state=42)

	# print("Size of train set: ", X_train.shape)
	# print("Size of validation set: ", X_test.shape)

	labels_to_Zs=get_representative_vectors(X_train,y_train)

	# print(labels_to_Zs)
	inds_2_labels=get_reverse_lookup(labels_to_Zs)

	preds=np.zeros(len(labels_to_Zs))
	# print()

	for j in range(len(X_test)):

		for k in range(len(preds)):

			preds[k]=np.correlate(labels_to_Zs[inds_2_labels[k]],X_test[j])

		y_h=inds_2_labels[np.argmax(preds)]

		if y_h==y_test[j] or y_h.split(":")[0] == y_test[j].split(":")[0]:
			# print("(Predicted, label): ", y_h,y_test[j], "1")
			correct+=1
		else:
			pass
			# print("(Predicted, label): ", y_h,y_test[j], "0")

	# print (correct*100.0/len(X_test))
	return correct,len(X_test)


def evaluate_classifier(X,Y,clf):


	X_train, X_test, y_train, y_test = train_test_split(X, Y, test_size=0.3, random_state=42)

	# print("Size of train set: ", X_train.shape)
	# print("Size of validation set: ", X_test.shape)
	# clf = GaussianNB()
	# clf = SVC(kernel='rbf')
	clf.fit(X_train, y_train)
	preds = clf.predict(X_test)

	correct = 0
	for j in range(len(X_test)):
		if preds[j]==y_test[j] or preds[j].split(":")[0] == y_test[j].split(":")[0]:
				# print("(Predicted, label): ", preds[j],y_test[j], "1")
				correct+=1
		else:
			pass
			# print("(Predicted, label): ", preds[j],y_test[j], "0")

	# print (correct*100.0/len(X_test))
	# print(accuracy_score(y_test,preds))

	return correct, len(X_test)


def evaluate_all_songs_jointly(data,labels):

	#gives RCO,ARCO,TRCO

	RCO=0
	ARCO=0
	TRCO=0

	TRCO_num=0
	TRCO_denom=0


	for i in range(len(data)):

		correct,total=evaluate_a_song(data[i],labels[i])

		RCO=(correct*100.0/total)

		ARCO+=(correct*1.0/total)

		TRCO_num+=correct
		TRCO_denom+=total

		print ("RCO for current song is :",RCO," percent.")

	TRCO=TRCO_num*1.0/TRCO_denom
	ARCO=ARCO*1.0/len(data)
	TRCO=TRCO_num*1.0/TRCO_denom


	print ("TRCO for dataset is :",TRCO)
	print ("ARCO for dataset is :",ARCO)


def evaluate_all_songs_jointly_classifier(data,labels,clf):

	#gives RCO,ARCO,TRCO

	RCO=0
	ARCO=0
	TRCO=0

	TRCO_num=0
	TRCO_denom=0


	for i in range(len(data)):

		correct,total=evaluate_classifier(data[i],labels[i],clf)

		RCO=(correct*100.0/total)

		ARCO+=(correct*1.0/total)

		TRCO_num+=correct
		TRCO_denom+=total

		print ("RCO for current song is :",RCO," percent.")

	TRCO=TRCO_num*1.0/TRCO_denom
	ARCO=ARCO*1.0/len(data)
	TRCO=TRCO_num*1.0/TRCO_denom


	print ("TRCO for dataset is :",TRCO)
	print ("ARCO for dataset is :",ARCO)