def change_skin(self):
"""Set skins"""
sender = self.sender()
if sender:
if (sender.text() in list(self.skins.keys())):
self.setStyleSheet(pyqt5_loader.load_stylesheet_pyqt5(style=self.skins[sender.text()]))
else:
self.setStyleSheet(pyqt5_loader.load_stylesheet_pyqt5(style="style_Dark"))
else:
self.setStyleSheet(pyqt5_loader.load_stylesheet_pyqt5(style="style_Dark"))
def change_skin(self):
"""Set skins"""
sender = self.sender()
if sender:
if (sender.text() in list(self.skins.keys())):
self.setStyleSheet(pyqt5_loader.load_stylesheet_pyqt5(style=self.skins[sender.text()]))
else:
self.setStyleSheet(pyqt5_loader.load_stylesheet_pyqt5(style="style_Dark"))
else:
self.setStyleSheet(pyqt5_loader.load_stylesheet_pyqt5(style="style_Dark"))
# Make the stackedWidg to default at the begining
self.tabWidget_items.setCurrentIndex(0)
self.stackedWidget_preprocessing_methods.setCurrentIndex(-1)
self.stackedWidget_dimreduction.setCurrentIndex(-1)
self.stackedWidget_feature_selection.setCurrentIndex(-1)
def __init__(self):
print("Opening GUI...")
self.root_dir = os.path.dirname(eslearn.__file__)
QMainWindow.__init__(self)
Ui_MainWindow.__init__(self)
self.setupUi(self)
self.working_directory = None
self.configuration_file = ""
self.textBrowser.setText("Hi~, I'm easylearn. I hope I can help you finish this project successfully\n")
# Display start progress
self.start_process()
# Set working_directory and debug
if self.working_directory:
cgitb.enable(format="text", display=1, logdir=os.path.join(self.working_directory, "log_data_loading.txt"))
else:
cgitb.enable(display=1, logdir=None)
# Connecting to functions
self.select_working_directory.triggered.connect(self.select_workingdir_fun)
self.create_configuration_file.triggered.connect(self.initialize_configuration_fun)
self.choose_configuration_file.triggered.connect(self.load_configuration_fun)
self.data_loading.clicked.connect(self.data_loading_fun)
self.feature_engineering.clicked.connect(self.feature_engineering_fun)
self.machine_learning.clicked.connect(self.machine_learning_fun)
self.model_evaluation.clicked.connect(self.model_evaluation_fun)
self.run.clicked.connect(self.run_fun)
self.quit.clicked.connect(self.closeEvent_button)
# Skin
self.skins = {"Dark": "style_Dark", "Black": "style_black", "DarkOrange": "style_DarkOrange",
"Gray": "style_gray", "Blue": "style_blue", "Navy": "style_navy", "Classic": "style_Classic"}
self.actionDark.triggered.connect(self.change_skin)
self.actionBlack.triggered.connect(self.change_skin)
self.actionDarkOrange.triggered.connect(self.change_skin)
self.actionGray.triggered.connect(self.change_skin)
self.actionBlue.triggered.connect(self.change_skin)
self.actionNavy.triggered.connect(self.change_skin)
self.actionClassic.triggered.connect(self.change_skin)
# Set appearance
self.set_run_appearance()
# Set initial skin
self.setStyleSheet(pyqt5_loader.load_stylesheet_pyqt5(style="style_Dark"))
# Dectecting versionn and news
self.detect_version()
def __init__(self, working_directory=None):
QMainWindow.__init__(self)
Ui_MainWindow.__init__(self)
self.setupUi(self)
self.root_dir = os.path.dirname(eslearn.__file__)
# Set working_directory and debug
self.working_directory = working_directory
self.configuration_file = ""
if self.working_directory:
cgitb.enable(format="text",
display=1,
logdir=os.path.join(self.working_directory,
"log_data_loading"))
else:
cgitb.enable(display=1, logdir=None)
# initiating
self.data_loading = {}
self.selected_group = None
self.selected_modality = None
self.selected_file = None
self.loaded_targets_and_covariates = None
self.loaded_mask = None
self.loaded_files = None
self.group_keys_exclude_modality = ["targets", "covariates"]
# initialize list_view for groups, modalities and files
self.slm_group = QStringListModel()
self.slm_modality = QStringListModel()
self.slm_file = QStringListModel()
# connections
self.actionChoose_configuration_file.triggered.connect(
self.load_configuration)
self.actionSave_configuration.triggered.connect(
self.save_configuration)
self.listView_groups.clicked.connect(self.identify_selected_group)
self.pushButton_addgroups.clicked.connect(self.add_group)
self.listView_groups.doubleClicked.connect(self.remove_selected_group)
self.pushButton_removegroups.clicked.connect(
self.remove_selected_group)
self.pushButton_cleargroups.clicked.connect(self.clear_all_group)
self.listView_modalities.clicked.connect(
self.identify_selected_modality)
self.pushButton_addmodalities.clicked.connect(self.add_modality)
self.listView_modalities.doubleClicked.connect(
self.remove_selected_modality)
self.pushButton_removemodalites.clicked.connect(
self.remove_selected_modality)
self.pushButton_clearmodalities.clicked.connect(
self.clear_all_modality)
self.listView_files.clicked.connect(self.identify_selected_file)
self.pushButton_addfiles.clicked.connect(self.add_files)
self.listView_files.doubleClicked.connect(self.remove_selected_file)
self.pushButton_removefiles.clicked.connect(self.remove_selected_file)
self.pushButton_clearfiles.clicked.connect(self.clear_all_file)
# mask_target_covariates
self.target_covariate_mask_dict = {
"Select mask": [self.lineEdit_mask, "mask"],
"Clear mask": [self.lineEdit_mask, "mask"],
"Select targets": [self.lineEdit_target, "targets"],
"Clear targets": [self.lineEdit_target, "targets"],
"Select covariates": [self.lineEdit_covariates, "covariates"],
"Clear covariates": [self.lineEdit_covariates, "covariates"]
}
self.pushButton_selectMask.clicked.connect(self.input_mask)
self.pushButton_selectTarget.clicked.connect(
self.input_target_covariate)
self.pushButton_selectCovariance.clicked.connect(
self.input_target_covariate)
self.pushButton_clearMask.clicked.connect(
self.clear_mask_target_covariates)
self.pushButton_clearTarget.clicked.connect(
self.clear_mask_target_covariates)
self.pushButton_clearCovriance.clicked.connect(
self.clear_mask_target_covariates)
self.pushButton_mask.clicked.connect(self.confirm_box_mask)
self.pushButton_target.clicked.connect(self.confirm_box_target)
self.pushButton_covariate.clicked.connect(self.confirm_box_covariates)
# Skin
self.skins = {
"Dark": "style_Dark",
"Black": "style_black",
"DarkOrange": "style_DarkOrange",
"Gray": "style_gray",
"Blue": "style_blue",
"Navy": "style_navy",
"Classic": "style_Classic"
}
self.actionDark.triggered.connect(self.change_skin)
self.actionBlack.triggered.connect(self.change_skin)
self.actionDarkOrange.triggered.connect(self.change_skin)
self.actionGray.triggered.connect(self.change_skin)
self.actionBlue.triggered.connect(self.change_skin)
self.actionNavy.triggered.connect(self.change_skin)
self.actionClassic.triggered.connect(self.change_skin)
# Set appearance
self.set_run_appearance()
# Set initial skin
self.setStyleSheet(
pyqt5_loader.load_stylesheet_pyqt5(style="style_Dark"))
def __init__(self, working_directory=None):
QMainWindow.__init__(self)
Ui_MainWindow.__init__(self)
self.setupUi(self)
self.root_dir = os.path.dirname(eslearn.__file__)
# Initialization
self.machine_learning = {}
self.working_directory = working_directory
self.configuration_file = ""
self.configuration = {}
self.all_inputs_fun()
# Debug
# Set working_directory
self.working_directory = working_directory
if self.working_directory:
cgitb.enable(format="text",
display=1,
logdir=os.path.join(self.working_directory,
"log_machine_learning"))
else:
cgitb.enable(display=1, logdir=None)
# Connect configuration functions
self.actionLoad_configuration.triggered.connect(
self.load_configuration)
self.actionSave_configuration.triggered.connect(
self.save_configuration)
self.actionGet_all_available_configuration.triggered.connect(
self._get_all_available_inputs)
# Connect to radioButton of machine learning type: switche to corresponding machine learning type window
self.machine_learning_type_stackedwedge_dict = {
"Classification": 0,
"Regression": 1,
"Clustering": 2,
"Deep learning": 3,
}
self.radioButton_classification.clicked.connect(
self.switche_stacked_wedge_for_machine_learning_type)
self.radioButton_regression.clicked.connect(
self.switche_stacked_wedge_for_machine_learning_type)
self.radioButton_clustering.clicked.connect(
self.switche_stacked_wedge_for_machine_learning_type)
self.radioButton_deeplearning.clicked.connect(
self.switche_stacked_wedge_for_machine_learning_type)
# Connect classification setting signal to slot: switche to corresponding classification model
self.classification_stackedwedge_dict = {
"LogisticRegression(solver='saga')": 0,
"LinearSVC()": 1,
"SVC()": 2,
"RidgeClassifier()": 3,
"GaussianProcessClassifier()": 4,
"RandomForestClassifier()": 5,
"AdaBoostClassifier()": 6
}
self.radioButton_classification_lr.clicked.connect(
self.switche_stacked_wedge_for_classification)
self.radioButton_classification_linearsvc.clicked.connect(
self.switche_stacked_wedge_for_classification)
self.radioButton_classification_svm.clicked.connect(
self.switche_stacked_wedge_for_classification)
self.radioButton_classification_ridge.clicked.connect(
self.switche_stacked_wedge_for_classification)
self.radioButton_classification_gaussianprocess.clicked.connect(
self.switche_stacked_wedge_for_classification)
self.radioButton_classification_randomforest.clicked.connect(
self.switche_stacked_wedge_for_classification)
self.radioButton_classification_adaboost.clicked.connect(
self.switche_stacked_wedge_for_classification)
# Connect regression setting signal to slot: switche to corresponding regression method
self.regression_stackedwedge_dict = {
"LinearRegression()": 0,
"LassoCV()": 1,
"RidgeCV()": 2,
"ElasticNetCV()": 3,
"SVR()": 4,
"GaussianProcessRegressor()": 5,
"RandomForestRegressor()": 6
}
self.radioButton_regression_linearregression.clicked.connect(
self.switche_stacked_wedge_for_regression)
self.radioButton_regression_lasso.clicked.connect(
self.switche_stacked_wedge_for_regression)
self.radioButton_regression_ridge.clicked.connect(
self.switche_stacked_wedge_for_regression)
self.radioButton_regression_elasticnet.clicked.connect(
self.switche_stacked_wedge_for_regression)
self.radioButton_regression_svm.clicked.connect(
self.switche_stacked_wedge_for_regression)
self.radioButton_regression_gaussianprocess.clicked.connect(
self.switche_stacked_wedge_for_regression)
self.radioButton_regression_randomforest.clicked.connect(
self.switche_stacked_wedge_for_regression)
# Connect clustering setting signal to slot: switche to corresponding clustering method
self.clustering_stackedwedge_dict = {
"KMeans()": 0,
"SpectralClustering()": 1,
"AgglomerativeClustering()": 2,
"DBSCAN()": 3
}
self.radioButton_clustering_kmeans.clicked.connect(
self.switche_stacked_wedge_for_clustering)
self.radioButton_spectral_clustering.clicked.connect(
self.switche_stacked_wedge_for_clustering)
self.radioButton_hierarchical_clustering.clicked.connect(
self.switche_stacked_wedge_for_clustering)
self.radioButton_DBSCAN.clicked.connect(
self.switche_stacked_wedge_for_clustering)
# Skin
self.skins = {
"Dark": "style_Dark",
"Black": "style_black",
"DarkOrange": "style_DarkOrange",
"Gray": "style_gray",
"Blue": "style_blue",
"Navy": "style_navy",
"Classic": "style_Classic"
}
self.actionDark.triggered.connect(self.change_skin)
self.actionBlack.triggered.connect(self.change_skin)
self.actionDarkOrange.triggered.connect(self.change_skin)
self.actionGray.triggered.connect(self.change_skin)
self.actionBlue.triggered.connect(self.change_skin)
self.actionNavy.triggered.connect(self.change_skin)
self.actionClassic.triggered.connect(self.change_skin)
# Set appearance
self.set_run_appearance()
# Set initial skin
self.setStyleSheet(
pyqt5_loader.load_stylesheet_pyqt5(style="style_Dark"))
def __init__(self, working_directory=None):
QMainWindow.__init__(self)
Ui_MainWindow.__init__(self)
self.setupUi(self)
self.root_dir = os.path.dirname(eslearn.__file__)
# Initialization
self.feature_engineering = {}
self.working_directory = working_directory
self.configuration_file = ""
self.configuration = {}
# self.all_available_inputs_fun()
# Debug
# Set working_directory
if self.working_directory:
cgitb.enable(format="text", display=1, logdir=os.path.join(self.working_directory, "log_feature_engineering"))
else:
cgitb.enable(display=1, logdir=None)
# Connect configuration functions
self.actionLoad_configuration.triggered.connect(self.load_configuration)
self.actionSave_configuration.triggered.connect(self.save_configuration)
self.actionGet_all_available_configuraton.triggered.connect(self._get_all_available_inputs)
# connect preprocessing setting signal to slot: switche to corresponding stackedWidget
self.preprocessing_stackedwedge_dict = {"StandardScaler()": 0, "MinMaxScaler()": 1, "None": 2}
self.radioButton_zscore.clicked.connect(self.switche_stacked_wedge_for_preprocessing)
self.radioButton_scaling.clicked.connect(self.switche_stacked_wedge_for_preprocessing)
self.radioButton_none_methods.clicked.connect(self.switche_stacked_wedge_for_preprocessing)
# connect dimreduction setting signal to slot: switche to corresponding stackedWidget
self.dimreduction_stackedwedge_dict = {"PCA()": 0, "NMF()": 1, "None": 2}
self.radioButton_pca.clicked.connect(self.switche_stacked_wedge_for_dimreduction)
self.radioButton_nmf.clicked.connect(self.switche_stacked_wedge_for_dimreduction)
self.radioButton_none.clicked.connect(self.switche_stacked_wedge_for_dimreduction)
# connect feature selection setting signal to slot: switche to corresponding stackedWidget
self.feature_selection_stackedwedge_dict = {
"VarianceThreshold()": 0, "SelectPercentile(f_classif)": 1, "SelectPercentile(f_regression)": 2,
"SelectPercentile(mutual_info_classif)": 3, "SelectPercentile(mutual_info_regression)": 4,
"RFE()": 5,
"SelectFromModel(LassoCV())": 6, "SelectFromModel(ElasticNetCV())": 7,
"None": 8
}
self.radioButton_variance_threshold.clicked.connect(self.switche_stacked_wedge_for_feature_selection)
self.radioButton_correlation.clicked.connect(self.switche_stacked_wedge_for_feature_selection)
self.radioButton_mutualinfo_cls.clicked.connect(self.switche_stacked_wedge_for_feature_selection)
self.radioButton_mutualinfo_regression.clicked.connect(self.switche_stacked_wedge_for_feature_selection)
self.radioButton_anova.clicked.connect(self.switche_stacked_wedge_for_feature_selection)
self.radioButton_rfe.clicked.connect(self.switche_stacked_wedge_for_feature_selection)
self.radioButton_l1.clicked.connect(self.switche_stacked_wedge_for_feature_selection)
self.radioButton_elasticnet.clicked.connect(self.switche_stacked_wedge_for_feature_selection)
self.radioButton_featureselection_none.clicked.connect(self.switche_stacked_wedge_for_feature_selection)
# Skin
self.skins = {"Dark": "style_Dark", "Black": "style_black", "DarkOrange": "style_DarkOrange",
"Gray": "style_gray", "Blue": "style_blue", "Navy": "style_navy", "Classic": "style_Classic"}
self.actionDark.triggered.connect(self.change_skin)
self.actionBlack.triggered.connect(self.change_skin)
self.actionDarkOrange.triggered.connect(self.change_skin)
self.actionGray.triggered.connect(self.change_skin)
self.actionBlue.triggered.connect(self.change_skin)
self.actionNavy.triggered.connect(self.change_skin)
self.actionClassic.triggered.connect(self.change_skin)
# Set appearance
self.set_run_appearance()
# Set initial skin
self.setStyleSheet(pyqt5_loader.load_stylesheet_pyqt5(style="style_Dark"))