def create_list_from_exp_data(self):
"""
Creates features for an interval and normalizes them according to
the maximum among all intervals.
"""
l_data = []
l_t = []
l_exp = []
for exp_no, folder_name in enumerate(self.train_exp_folders):
folder_path = os.path.join(self.exp_csv_dir, folder_name)
l_data.append([])
l_t.append([])
l_exp.append([])
# Each folder contain many files contain a single type of exp (class)
for file_name in os.listdir(
folder_path)[self.SPLIT[0]:self.SPLIT[1]]:
data, t, exp = read_data_from_file(os.path.join(
folder_path, file_name),
N_CH=8)
l_data[exp_no].append(np.array(data))
l_t[exp_no].append(np.array(t))
l_exp[exp_no].append(np.array(exp))
return l_data, l_t, l_exp
def read_data(self, layout):
data, t, exp = \
read_data_from_file(self.path_line_edit.text(), N_CH=self.gv.N_CH) # clean this part
for ch in range(self.gv.N_CH):
fg = self.right_gr.full_graphs[ch]
slider = self.right_gr.sliders[ch]
acg = self.left_gr.avg_classif_graphs[ch]
pgs = self.left_gr.portion_graphs[ch]
cg = self.left_gr.classif_graphs[ch]
self.pbar.setValue(int(100 * (ch + 1) / self.gv.N_CH))
# Right panel
fg.plot_data(data[ch], color='w')
slider.setMaximum(len(data[0]))
# Left panel
pgs.data = np.array(data[ch])
pgs.t = np.array(t)
pgs.plot_data(data[ch], color='g')
pgs.add_all_experimentation_regions(ch, exp)
classified_data = self.left_gr.classif_graphs[ch].classify_data(
data[ch])
cg.plot_data(classified_data, color='b')
acg.curve = acg.plot_data(np.zeros(self.gv.emg_signal_len),
color='b')
acg.classif_region_curve = acg.plot_data(np.zeros(
self.gv.emg_signal_len),
color='r')
acg.combo_box_curve = acg.plot_data(np.zeros(
self.gv.emg_signal_len),
color='y')
acg.classified_data = classified_data
acg.update_pos_and_avg_graph(classif_region_pos=0)