def analyze_signal(self): #Veriyi Analiz Etme
self.average, self.entropy, self.kurtosis, self.max_v, self.median, self.min_v, self.skewness, self.standart_dev = Features.check_statistics(
self)
self.wavelet_type, self.wavelet_level = Features.discrete_check_wavelet(
self)
if (self.load_check and
(self.average or self.entropy or self.kurtosis or self.max_v
or self.median or self.min_v or self.skewness
or self.standart_dev)): #işlenecek müzik olması durumu
self.analyze_check = True
col, self.header, self.db_header = Features.init_table(
self, len(self.all_signals))
self.db_matrix = np.zeros(
(len(self.all_signals), 8 * (self.wavelet_level + 1)))
for iter in range(0, len(self.all_signals)):
print(self.all_signals[iter])
self.signals.clear()
self.audio, self.sample = librosa.load(self.all_signals[iter])
self.signals.append(self.audio)
self.time = np.arange(0, len(self.audio)) / self.sample
coeffs = pywt.wavedec(
self.audio, self.wavelet_type,
level=self.wavelet_level) #Wavelet Analiz
for i in range(0, self.wavelet_level + 1):
self.signals.append(coeffs[i]) #Sinyalleri Coef içine atma
if (iter == 0
and self.wavelet_level < 8): #İlk sinyali yazdırma
Features.discrete_plot_signal(self)
self.db_matrix[iter] = Features.insertTable(self, iter, col)
self.progressBar.setValue(
((iter + 1) / len(self.all_signals)) * 100)
#level = 3 ise signals içinde 5 (4 analiz edilmiş + 1 source)
else:
if not self.load_check:
Features.message("You have to load at least 1 signal",
QMessageBox.Warning)
else:
Features.message(
"You have to pick at least 1 statistic function",
QMessageBox.Warning)
def run(self):
wave_types = pywt.wavelist(kind='discrete')
levels = [1, 2, 3]
signals = []
self.average, self.entropy, self.kurtosis, self.max_v, self.median, self.min_v, self.skewness, self.standart_dev = Features.check_statistics(
self)
table_names = Database.get_table_names(Database.database_name)
for wave_func in wave_types:
if wave_func.find("sym") == -1:
for level in levels:
self.wavelet_level = level
self.wavelet_type = wave_func
col, self.header, self.db_header = Features.init_table(
self, len(self.all_signals))
self.db_matrix = np.zeros(
(len(self.all_signals), 8 * (self.wavelet_level + 1)))
if not any("Db_GTZAN_function_" + str(wave_func) +
"_Degree_" + str(level) in s
for s in table_names):
for iter in range(0, len(self.all_signals)):
self.signals.clear()
self.audio, self.sample = librosa.load(
self.all_signals[iter])
self.signals.append(self.audio)
self.time = np.arange(0, len(
self.audio)) / self.sample
coeffs = pywt.wavedec(
self.audio,
self.wavelet_type,
level=self.wavelet_level) # wavelet analyze
for i in range(0, self.wavelet_level + 1):
self.signals.append(
coeffs[i]
) # adding signals array to coeffs
self.db_matrix[iter] = Features.insertTable(
self, iter, col
) # level = 3 ise signals içinde 5 (4 analiz edilmiş + 1 kaynak)
w_name = str(wave_func)
table_name = "Db_GTZAN_function_" + w_name + "_Degree_" + str(
level)
Database.create_table(Database.database_name,
table_name, self.db_header, "")
for index in range(0, len(self.all_signals)):
name = self.all_signals[index].split(
sep='/') # name of signal
Database.delete_row(Database.database_name,
table_name, name[-1], "")
Database.add_values_to_table(
Database.database_name, table_name, name[-1],
self.db_header, self.db_matrix[index],
"") # adding db to values
print(table_name)
else:
print("Db_GTZAN_function_" + str(wave_func) +
"_Degree_" + str(level) + ' passed')
else:
print(wave_func + " passed")