-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathklustr_view.py
322 lines (277 loc) · 13.5 KB
/
klustr_view.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
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
from PySide6 import QtWidgets, QtGui, QtCore
from PySide6.QtCore import Qt, Signal, Slot, QTimer, QRect
from PySide6.QtGui import qRgb, QImage, QPixmap
from PySide6.QtWidgets import QLabel, QScrollBar, QVBoxLayout, QHBoxLayout, QPushButton, QWidget, QGroupBox, QComboBox, \
QMessageBox
import sys
import Knn as k
import matplotlib.pyplot as plt
from matplotlib.backends.backend_qtagg import FigureCanvasQTAgg as FigureCanvas
from db_credential import PostgreSQLCredential
from klustr_dao import PostgreSQLKlustRDAO
import klustr_utils
class ScatterDiagram:
def __init__(self):
self.__figure = plt.figure(figsize=(5, 5))
self.__axes = plt.axes(projection='3d')
self.__canvas = FigureCanvas(self.__figure)
self.__canvas.draw()
self.init_ui_graph()
def init_ui_graph(self):
plt.cla() # pour clear sans le fermer
self.__axes.set_xlabel('X-axis', fontweight='bold')
self.__axes.set_ylabel('Y-axis', fontweight='bold')
self.__axes.set_zlabel('Z-axis', fontweight='bold')
self.__axes.set_title("Graphique de dispersion")
def update_data(self, matrice):
self.init_ui_graph()
x_vals = matrice[:, 0:1]
y_vals = matrice[:, 1:2]
z_vals = matrice[:, 2:3]
self.__axes.scatter3D(x_vals, y_vals, z_vals, color='blue', marker="o")
plt.draw()
def update_point_test(self, matrice):
x_vals = matrice[0]
y_vals = matrice[1]
z_vals = matrice[2]
self.__axes.scatter3D(x_vals, y_vals, z_vals, color='red', marker="^")
plt.draw()
@property
def widget(self):
return self.__canvas
class KlustR(QtWidgets.QMainWindow):
def __init__(self, klustr_dao, knn, parent=None):
super().__init__(parent)
self.klustr_dao = klustr_dao
self.displayPng = QPixmap(klustr_utils.qimage_argb32_from_png_decoding(
self.klustr_dao.get_image_test("img_ellipsoid_200_200_100_0031")[0][0]))
self.label_displayPng = QLabel()
self.dataset_stats = QVBoxLayout()
self.dataset_transform_stats = QVBoxLayout()
self.k = 1
self.comboBox_datasetNames = None
self.classified_status_label = QLabel()
self.setWindowTitle("KlustR KNN Classifier")
self.knn = knn
self.main_window_widget = QWidget()
self.main_window_layout = QHBoxLayout()
self.dataset_gb = self.data_set_groupbox()
self.comboBox_datasetsLabels = self.dataset_gb.create_drop_list(self.update_dataset_labels())
self.single_test_gb = self.single_test_groupbox()
self.knn_params_gb = self.knn_parameters_groupbox()
self.about_button = QPushButton("About")
control_layout = QVBoxLayout()
control_layout.addWidget(self.dataset_gb)
control_layout.addWidget(self.single_test_gb)
control_layout.addWidget(self.knn_params_gb)
control_layout.addWidget(self.about_button)
self.main_window_layout.addLayout(control_layout)
self.__scatter = ScatterDiagram()
self.main_window_layout.addWidget(self.graphique3D())
self.main_window_widget.setLayout(self.main_window_layout)
self.setCentralWidget(self.main_window_widget)
self.matrice_descripteur_training = self.knn.evaluation_training(
self.klustr_dao.get_dataset_tests(self.data_set_current_name()))
self.__scatter.update_data(self.matrice_descripteur_training)
self.update_dataset_stats()
self.about_button.clicked.connect(self.about_show)
def datasets_available(self):
data = []
for dataset in self.klustr_dao.available_datasets:
data.append(str(dataset[1]) + " " + "[" + str(dataset[5]) + "]" + "[" + str(dataset[8]) + "]")
return data
def update_dataset_labels(self):
data = []
for dataset in self.klustr_dao.image_from_dataset(self.data_set_current_name(), False):
data.append(str(dataset[3]))
return data
def data_set_groupbox(self):
group_box = TemplateGB("Dataset", 390, 180)
big_layout = QVBoxLayout() # main layout couvre tout groupbox
self.comboBox_datasetNames = group_box.create_drop_list(self.datasets_available())
self.comboBox_datasetNames.currentIndexChanged.connect(self.update_dataset_stats) # nom du data set
big_layout.addWidget(self.comboBox_datasetNames) # ajout select bar
dataset_infos_layout = QHBoxLayout() # layout interieur en bas pour les 2 groupbox
dataset_infos_layout.addWidget(self.included_in_dataset())
dataset_infos_layout.addWidget(self.transformation())
# ajout fin
big_layout.addLayout(dataset_infos_layout)
group_box.setLayout(big_layout)
return group_box
def included_in_dataset(self):
group_box = QtWidgets.QGroupBox("Included in dataset")
big_layout = QHBoxLayout()
titles = QVBoxLayout()
titles.addWidget(QLabel("Category count:"))
titles.addWidget(QLabel("Training image count:"))
titles.addWidget(QLabel("Test image count:"))
titles.addWidget(QLabel("Total image count:"))
big_layout.addLayout(titles)
big_layout.addLayout(self.dataset_stats)
group_box.setLayout(big_layout)
return group_box
def transformation(self):
group_box = QtWidgets.QGroupBox("Transformation")
big_layout = QHBoxLayout()
titles = QVBoxLayout()
titles.addWidget(QLabel("Translated:"))
titles.addWidget(QLabel("Rotated:"))
titles.addWidget(QLabel("Scaled:"))
titles.addStretch()
big_layout.addLayout(titles)
big_layout.addLayout(self.dataset_transform_stats)
group_box.setLayout(big_layout)
return group_box
def single_test_groupbox(self):
group_box = TemplateGB("Single test", 390, 290)
big_layout = QVBoxLayout()
self.comboBox_datasetsLabels.currentIndexChanged.connect(self.update_data_png)
big_layout.addWidget(self.comboBox_datasetsLabels) # ajout select bar
image_layout = QVBoxLayout()
self.label_displayPng.setAlignment(Qt.AlignmentFlag.AlignCenter)
self.label_displayPng.setPixmap(self.displayPng)
self.label_displayPng.setFixedHeight(170)
self.label_displayPng.setStyleSheet("background:#223544;")
self.label_displayPng.alignment = Qt.AlignCenter
image_layout.addWidget(self.label_displayPng)
classify_button = QPushButton("Classify")
classify_button.setFixedHeight(25)
classify_button.clicked.connect(self.classify_forme_test) # ajout Alex pour classifier la forme test
self.classified_status_label = QLabel("not classified")
self.classified_status_label.setAlignment(Qt.AlignmentFlag.AlignCenter)
big_layout.addLayout(image_layout)
big_layout.addWidget(classify_button)
big_layout.addWidget(self.classified_status_label)
group_box.setLayout(big_layout)
return group_box
def knn_parameters_groupbox(self):
group_box = TemplateGB("Knn parameters", 390, 80)
big_layout = QHBoxLayout()
mini_layout = QHBoxLayout()
mini_layout.addWidget(QLabel("K ="))
data = QLabel('1')
mini_layout.addWidget(data)
max_dis_select_layout = QHBoxLayout()
title_max_dist = QLabel("Max dist =")
title_max_dist.setFixedWidth(60)
max_dis_select_layout.addWidget(title_max_dist)
max_data = QLabel("0.01")
max_data.setFixedWidth(25)
max_dis_select_layout.addWidget(max_data)
titles_layout = QVBoxLayout()
titles_layout.addLayout(mini_layout)
titles_layout.addLayout(max_dis_select_layout)
scroll_layout = QVBoxLayout()
knn_scroll = QScrollBar()
knn_scroll.setMinimum(1)
knn_scroll.setMaximum(3)
knn_scroll.valueChanged.connect(lambda: self.change_k_value(data, knn_scroll))
knn_scroll.setOrientation(Qt.Horizontal)
knn_scroll.setFixedWidth(270)
dist_max_scroll = QScrollBar()
dist_max_scroll.setMinimum(0.01)
dist_max_scroll.setMaximum(99.9)
dist_max_scroll.setOrientation(Qt.Horizontal)
dist_max_scroll.setFixedWidth(270)
scroll_layout.addWidget(knn_scroll)
scroll_layout.addWidget(dist_max_scroll)
big_layout.addLayout(titles_layout)
big_layout.addLayout(scroll_layout)
big_layout.addStretch()
group_box.setLayout(big_layout)
return group_box
def change_k_value(self, data, k):
data.setNum(k.value())
self.k = k.value()
def graphique3D(self):
return self.__scatter.widget
def update_data_png(self):
if self.comboBox_datasetsLabels.currentText().split():
val = (self.comboBox_datasetsLabels.currentText().split())[0]
img = klustr_utils.qimage_argb32_from_png_decoding(self.klustr_dao.get_image_test(val)[0][0])
self.label_displayPng.setPixmap(QPixmap(img))
def update_dataset_stats(self):
requete_image = self.klustr_dao.get_dataset_tests(self.data_set_current_name())
self.matrice_descripteur_training = self.knn.evaluation_training(requete_image)
self.comboBox_datasetsLabels.clear()
self.comboBox_datasetsLabels.addItems(self.update_dataset_labels())
self.__scatter.update_data(self.matrice_descripteur_training)
tab = self.klustr_dao.get_current_dataset_info(self.data_set_current_name())
for i in reversed(range(self.dataset_stats.count())):
self.dataset_stats.itemAt(i).widget().deleteLater()
for j in reversed(range(self.dataset_transform_stats.count())):
self.dataset_transform_stats.itemAt(j).widget().deleteLater()
self.dataset_stats.addWidget(QLabel(str(tab[0][5])))
self.dataset_stats.addWidget(QLabel(str(tab[0][6])))
self.dataset_stats.addWidget(QLabel(str(tab[0][7])))
self.dataset_stats.addWidget(QLabel(str(tab[0][8])))
self.dataset_transform_stats.addWidget(QLabel(str(tab[0][2])))
self.dataset_transform_stats.addWidget(QLabel(str(tab[0][3])))
self.dataset_transform_stats.addWidget(QLabel(str(tab[0][4])))
def data_set_current_name(self):
return (self.comboBox_datasetNames.currentText().split())[0]
def data_set_current_name_test(self):
return self.comboBox_datasetsLabels.currentText()
def classify_forme_test(self):
requete_image = self.klustr_dao.get_image_test(self.data_set_current_name_test())
descripteur_test = self.knn.evaluation_test(requete_image)
self.__scatter.update_point_test(descripteur_test)
forme_trouve = self.knn.trouver_forme_test(descripteur_test, self.matrice_descripteur_training, self.k)
nom_label_test = self.klustr_dao.get_label_test(str(int(forme_trouve)))
self.classified_status_label.setText(nom_label_test[0][0])
def about_show(self):
data = """
Ce logiciel est le projet no 1 du cours C52.
Il a été réalisé par :
- Roberto N.
- Jeremie K.
- Antonin L.
- Alexandre C.
Il consiste à faire __quelque_chose__ avec les concepts suivants :
- Machine learning
- la classification et la régression
Nos 3 descripteurs de forme sont :
- La complexité :
- Consiste à calculer l'air et le perimètre de la forme. On normalise la
complexité.
- Le ratio entre le cercle circonscrit et l'air.
- Notre calcul nous donnera un ratio pour chaque forme et n'est pas sujet
à la translation, la rotationn ou le scale
- Le ratio entre le cercle inscrit et le cercle circonscrit
- Notre calcul nous donnera un ratio pour chaque forme et n'est pas sujet
à la translation, la rotationn ou le scale
Plus précisément, ce laboratoire permet de mettre en pratique les notions de :
- Traitement de données
- Intéragir avec une base de données contenant des milliers de données
- Optimisation de calculs sans boucle for avec Numpy
- Optimisation de l'exécution du temps d'affichage et le traitement des données
- Travail collaboratif
Un effort d'abstraction a été fait pour ces points :
- La classe KNN
- Le klustr View
Finalement, l’ensemble de données le plus complexe que nous avons été capable
de résoudre est:
- Zoo-Large
"""
detail = QMessageBox.about(self, "KlustR KNN Classifier", data)
class TemplateGB(QtWidgets.QGroupBox):
def __init__(self, title, width, height, parent=None):
super().__init__(parent)
self.setTitle(title)
self.setFixedWidth(width)
self.setFixedHeight(height)
self.layout()
def create_drop_list(self, data):
cmbbox_widget = QComboBox(self)
drop_list_content = data
cmbbox_widget.addItems(drop_list_content)
return cmbbox_widget
def main():
app = QtWidgets.QApplication(sys.argv)
credential = PostgreSQLCredential(password='ASDasd123')
klustr_dao = PostgreSQLKlustRDAO(credential)
klustR = KlustR(klustr_dao, k.Knn())
klustR.show()
sys.exit(app.exec())
if __name__ == '__main__':
main()