-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathfine_tuning_boosting.py
215 lines (200 loc) · 7.97 KB
/
fine_tuning_boosting.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
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
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
from os import listdir
import xml.etree.ElementTree as ET
from sklearn.neural_network import MLPClassifier
from sklearn.svm import SVC,LinearSVC
from sklearn.preprocessing import StandardScaler
from sklearn.model_selection import train_test_split, GridSearchCV
from sklearn.metrics import average_precision_score
from numpy import mean, zeros
import numpy as np
from sklearn.ensemble import RandomForestClassifier,AdaBoostClassifier
#import _pickle as Pikcle
import json
def load_labels(label_folder):
labels = {}
classes = set()
for image in listdir(label_folder):
img_num = image[:-4]
tree = ET.parse(label_folder + img_num + ".xml")
root = tree.getroot()
objects = [x.text for x in root.findall("./object/name")]
for o in objects:
classes.add(o)
labels[img_num] = set(objects)
return labels, classes
def load_features(filename):
f = open(filename, 'r')
feats = {}
for line in f:
features = line.split()
feats[features[0]] = [float(x) for x in features[1:]]
f.close()
return feats
def make_multi_label(features, labels, objects):
objects = list(objects)
objects.sort()
X = []
Y = []
for image in labels:
y = zeros(20)
for ob in labels[image]:
y[objects.index(ob)] = 1.0
Y.append(y)
X.append(features[image])
return X, Y
# function that takes dictionary of (imageID, features) and (imageID, labels)
# returns dataset with binary labels based on if "classname" object is in the image
def make_binary_dataset(features, labels, classname):
X = []
Y = []
for image in labels:
X.append(features[image])
if classname in labels[image]:
Y.append(1.0)
else:
Y.append(0.0)
return X, Y
def normalize(features):
keys = list(features.keys())
feats = list(features.values())
scaler = StandardScaler(copy=False)
feats = scaler.fit_transform(feats)
features = {}
n = len(keys)
for i in range(n):
features[keys[i]] = feats[i]
return features
def search_adaboost():
fw = open('results1.txt', 'a')
fw.write("Ada Boost: rf = AdaBoostClassifier(random_state=0,base_estimator=SVC(probability=True,kernel='poly')"
" Parameters:n_estimators: [2,3]" + '\n')
avg_prec = []
i = 0
for object in classes:
i += 1
print(object + " " + str(i))
fw.write(str(object) + " " + str(i)+'\n')
X, Y = make_binary_dataset(features, labels, object)
rf = AdaBoostClassifier(random_state=0,base_estimator=SVC(probability=True,kernel='poly'))
parameters = {'n_estimators': [2,3]}
clf = GridSearchCV(rf, parameters, scoring='average_precision', verbose=1, n_jobs=4, cv=2)
clf.fit(X, Y)
print(clf.best_params_)
fw.write(json.dumps(clf.best_params_)+'\n')
print(clf.best_score_)
fw.write(str(clf.best_score_)+'\n')
avg_prec.append(clf.best_score_)
X_test,Y_test=make_binary_dataset(features_test,labels_test,object)
Y_pred=clf.predict(X_test)
score=average_precision_score(Y_test,Y_pred)
print("Average Precision Score:",score)
temp = Y_test == Y_pred
accuracy = (np.sum(temp) / len(Y_pred))
print("Accuracy:", accuracy)
fw.write("Average Precision Score:" + str(score) + '\n')
fw.write("Accuracy:" + str(accuracy) + '\n')
print(mean(avg_prec))
fw.write(str(mean(avg_prec))+'\n')
fw.close()
def search_adaboost1():
fw = open('results1.txt', 'a')
fw.write("Ada Boost: Parameters:n_estimators: [300]" + '\n')
avg_prec = []
avg_prec1= []
i = 0
for object in classes:
i += 1
print(object + " " + str(i))
fw.write(str(object) + " " + str(i)+'\n')
X, Y = make_binary_dataset(features, labels, object)
rf = AdaBoostClassifier(random_state=0)
parameters = {'n_estimators': [300]}
clf = GridSearchCV(rf, parameters, scoring='average_precision', verbose=1, n_jobs=4, cv=2)
clf.fit(X, Y)
print(clf.best_params_)
fw.write(json.dumps(clf.best_params_)+'\n')
print(clf.best_score_)
fw.write(str(clf.best_score_)+'\n')
avg_prec.append(clf.best_score_)
X_test,Y_test=make_binary_dataset(features_test,labels_test,object)
Y_pred=clf.predict_proba(X_test)
score=average_precision_score(Y_test,Y_pred[:,1])
print("Average Precision Score:",score)
temp = Y_test == Y_pred
accuracy = (np.sum(temp) / len(Y_pred))
print("Accuracy:", accuracy)
fw.write("Average Precision Score:" + str(score) + '\n')
fw.write("Accuracy:" + str(accuracy) + '\n')
avg_prec1.append(score)
print(mean(avg_prec))
fw.write(str(mean(avg_prec))+'\n')
print('Avg Precision Test:', mean(avg_prec1))
fw.write('Avg Precision Test:' + str(mean(avg_prec1)))
fw.close()
if __name__ == '__main__':
image_annotations = "./Annotations/"
image_annotations_test = "./Annotations_test/"
features = load_features("train_features_fc.txt")
labels, classes = load_labels(image_annotations)
features = normalize(features)
features_test=load_features("test_features_fc.txt")
features_test=normalize(features_test)
labels_test, classes_test = load_labels(image_annotations_test)
#search_adaboost()
search_adaboost1()
fw=open('results1.txt','a')
fw.write("Random Forest: Parameters:n_estimators: [300], 'max_depth': [5,7,9]" + '\n')
avg_prec = []
avg_prec1 = []
i = 0
for object in classes:
i += 1
print(object + " " + str(i))
fw.write(str(object) + " " + str(i) + '\n')
X, Y = make_binary_dataset(features, labels, object)
rf = RandomForestClassifier(random_state=0)
parameters = {'n_estimators': [300], 'max_depth': [5,7,9]}
clf = GridSearchCV(rf, parameters, scoring='average_precision', verbose=1, n_jobs=4, cv=2)
clf.fit(X,Y)
print(clf.best_params_)
fw.write(json.dumps(clf.best_params_)+'\n')
print(clf.best_score_)
fw.write(str(clf.best_score_)+'\n')
avg_prec.append(clf.best_score_)
X_test,Y_test=make_binary_dataset(features_test,labels_test,object)
Y_pred=clf.predict_proba(X_test)
score=average_precision_score(Y_test,Y_pred[:,1])
print("Average Precision Score:",score)
temp=Y_test==Y_pred
accuracy=(np.sum(temp)/len(Y_pred))
print("Accuracy:",accuracy)
fw.write("Average Precision Score:"+str(score)+'\n')
fw.write("Accuracy:" + str(accuracy) + '\n')
avg_prec1.append(score)
#X_train, X_test, y_train, y_test = train_test_split(X, Y, test_size=0.3)
#clf = MLPClassifier(solver='adam', hidden_layer_sizes=(3,2,), activation='relu')
# clf.fit(X_train, y_train)
# pred = clf.predict_proba(X_train)
# print(average_precision_score(y_train, [x[1] for x in pred]))
# pred = clf.predict_proba(X_test)
# ap = average_precision_score(y_test, [x[1] for x in pred])
# print(ap)
# avg_prec.append(ap)
print(mean(avg_prec))
fw.write(str(mean(avg_prec)))
print('Avg Precision Test:',mean(avg_prec1))
fw.write('Avg Precision Test:'+str(mean(avg_prec1)))
fw.close()
# X, Y = make_multi_label(features, labels, classes)
# X_train, X_test, y_train, y_test = train_test_split(X, Y, test_size=0.3)
# clf = MLPClassifier(solver='adam', hidden_layer_sizes=(30,), activation='relu')
# print("training")
# clf.fit(X_train, y_train)
# pred = clf.predict_proba(X_train)
# print(pred[0])
# print(average_precision_score(y_train, pred))
# pred = clf.predict_proba(X_test)
# ap = average_precision_score(y_test, pred)
# print(ap)
#
# exit(0)