--- a
+++ b/python/aux1/generate_graphics.py
@@ -0,0 +1,216 @@
+import matplotlib.pyplot as plt
+import numpy as np
+from load_MITBIH import *
+
+# Generate graphics for paper
+
+db_path = '/home/mondejar/dataset/ECG/mitdb/m_learning/scikit/'
+winL = 90
+winR = 90
+do_preprocess = True
+maxRR = True
+
+use_RR = False
+norm_RR = False
+
+reduced_DS = False
+leads_flag = [1, 0]
+# Load train data 
+compute_morph = {'raw'}
+
+label_name = ['N', 'S', 'V', 'F']
+line_styles =  ['-', '--', ':', '-.']
+l_width = 2
+
+
+plt.figure(figsize=(8.27, 11.69))
+
+print("1 Raw")
+compute_morph = {'raw'}
+
+[features_1, labels_1, patient_num_beats_1] = load_mit_db('DS1', winL, winR, do_preprocess,
+    maxRR, use_RR, norm_RR, compute_morph, db_path, reduced_DS, leads_flag)
+
+[features_2, labels_2, patient_num_beats_2] = load_mit_db('DS2', winL, winR, do_preprocess,
+    maxRR, use_RR, norm_RR, compute_morph, db_path, reduced_DS, leads_flag)
+
+features = np.vstack((features_1, features_2))
+labels = np.concatenate((labels_1, labels_2))
+
+
+ax1 = plt.subplot(321)
+ax1.title.set_text('(a)')
+
+raw_avg = np.zeros((4, features.shape[1]))
+for n in range(0,4):
+    raw_avg[n] = np.average(features[labels == n], axis=0)
+    plt.plot(raw_avg[n], label=label_name[n], linestyle=line_styles[n], linewidth=l_width)
+
+leg = plt.legend(loc='best', ncol=1, shadow=False, fancybox=True)
+leg.get_frame().set_alpha(0.5)
+
+
+# 1 Average wavelets
+print("2 wavelets")
+compute_morph = {'wvlt'}
+
+[features_1, labels_1, patient_num_beats_1] = load_mit_db('DS1', winL, winR, do_preprocess,
+    maxRR, use_RR, norm_RR, compute_morph, db_path, reduced_DS, leads_flag)
+
+[features_2, labels_2, patient_num_beats_2] = load_mit_db('DS2', winL, winR, do_preprocess,
+    maxRR, use_RR, norm_RR, compute_morph, db_path, reduced_DS, leads_flag)
+
+features = np.vstack((features_1, features_2))
+labels = np.concatenate((labels_1, labels_2))
+
+
+ax2 = plt.subplot(322)
+ax2.title.set_text('(b)')
+
+wvlt_avg = np.zeros((4, features.shape[1]))
+for n in range(0,4):
+    wvlt_avg[n] = np.average(features[labels == n], axis=0)
+    plt.plot(wvlt_avg[n], label=label_name[n], linestyle=line_styles[n], linewidth=l_width)
+
+leg = plt.legend(loc='best', ncol=1, shadow=False, fancybox=True)
+leg.get_frame().set_alpha(0.5)
+
+
+# 2 Average HOS
+compute_morph = {'HOS'}
+print("3 HOS")
+
+[features_1, labels_1, patient_num_beats_1] = load_mit_db('DS1', winL, winR, do_preprocess,
+    maxRR, use_RR, norm_RR, compute_morph, db_path, reduced_DS, leads_flag)
+
+[features_2, labels_2, patient_num_beats_2] = load_mit_db('DS2', winL, winR, do_preprocess,
+    maxRR, use_RR, norm_RR, compute_morph, db_path, reduced_DS, leads_flag)
+
+features = np.vstack((features_1, features_2))
+labels = np.concatenate((labels_1, labels_2))
+
+
+ax3 = plt.subplot(323)
+ax3.title.set_text('(c)')
+
+hos_avg = np.zeros((4, features.shape[1]))
+
+for n in range(0,4):
+    hos_avg[n] = np.average(features[labels == n], axis=0)
+    plt.plot(hos_avg[n], label=label_name[n], linestyle=line_styles[n], linewidth=l_width)
+
+leg = plt.legend(loc='best', ncol=1, shadow=False, fancybox=True)
+leg.get_frame().set_alpha(0.5)
+
+# 3 Average U-LBP1D
+
+compute_morph = {'u-lbp'}
+print("4 U LBP")
+
+[features_1, labels_1, patient_num_beats_1] = load_mit_db('DS1', winL, winR, do_preprocess,
+    maxRR, use_RR, norm_RR, compute_morph, db_path, reduced_DS, leads_flag)
+
+[features_2, labels_2, patient_num_beats_2] = load_mit_db('DS2', winL, winR, do_preprocess,
+    maxRR, use_RR, norm_RR, compute_morph, db_path, reduced_DS, leads_flag)
+
+features = np.vstack((features_1, features_2))
+labels = np.concatenate((labels_1, labels_2))
+
+ax4 = plt.subplot(324)
+ax4.title.set_text('(d)')
+
+lbp_avg = np.zeros((4, features.shape[1]))
+for n in range(0,4):
+    lbp_avg[n] = np.average(features[labels == n], axis=0)
+    plt.plot(lbp_avg[n], label=label_name[n], linestyle=line_styles[n], linewidth=l_width)
+
+leg = plt.legend(loc='best', ncol=1, shadow=False, fancybox=True)
+leg.get_frame().set_alpha(0.5)
+
+
+"""
+compute_morph = {'hbf'}
+print("4 HBF")
+
+[features_1, labels_1, patient_num_beats_1] = load_mit_db('DS1', winL, winR, do_preprocess,
+    maxRR, use_RR, norm_RR, compute_morph, db_path, reduced_DS, leads_flag)
+
+[features_2, labels_2, patient_num_beats_2] = load_mit_db('DS2', winL, winR, do_preprocess,
+    maxRR, use_RR, norm_RR, compute_morph, db_path, reduced_DS, leads_flag)
+
+features = np.vstack((features_1, features_2))
+labels = np.concatenate((labels_1, labels_2))
+
+ax4 = plt.subplot(324)
+ax4.title.set_text('d')
+
+hbf_avg = np.zeros((4, features.shape[1]))
+for n in range(0,4):
+    hbf_avg[n] = np.average(features[labels == n], axis=0)
+    plt.plot(hbf_avg[n], label=label_name[n], linestyle=line_styles[n], linewidth=l_width)
+
+leg = plt.legend(loc='best', ncol=1, shadow=False, fancybox=True)
+leg.get_frame().set_alpha(0.5)
+"""
+
+# 4 Average MyMorph
+compute_morph = {'myMorph'}
+print("5 myMorph")
+
+[features_1, labels_1, patient_num_beats_1] = load_mit_db('DS1', winL, winR, do_preprocess,
+    maxRR, use_RR, norm_RR, compute_morph, db_path, reduced_DS, leads_flag)
+
+[features_2, labels_2, patient_num_beats_2] = load_mit_db('DS2', winL, winR, do_preprocess,
+    maxRR, use_RR, norm_RR, compute_morph, db_path, reduced_DS, leads_flag)
+
+features = np.vstack((features_1, features_2))
+labels = np.concatenate((labels_1, labels_2))
+
+
+ax5 = plt.subplot(325)
+ax5.title.set_text('(e)')
+
+myMorph_avg = np.zeros((4, features.shape[1]))
+for n in range(0,4):
+    myMorph_avg[n] = np.average(features[labels == n], axis=0)
+    plt.plot(myMorph_avg[n], label=label_name[n], linestyle=line_styles[n], linewidth=l_width)
+
+leg = plt.legend(loc='best', ncol=1, shadow=False, fancybox=True)
+leg.get_frame().set_alpha(0.5)
+
+# 5 Average RR intervals
+compute_morph = {''}
+print("6 RR")
+
+use_RR = True
+norm_RR = True
+[features_1, labels_1, patient_num_beats_1] = load_mit_db('DS1', winL, winR, do_preprocess,
+    maxRR, use_RR, norm_RR, compute_morph, db_path, reduced_DS, leads_flag)
+
+[features_2, labels_2, patient_num_beats_2] = load_mit_db('DS2', winL, winR, do_preprocess,
+    maxRR, use_RR, norm_RR, compute_morph, db_path, reduced_DS, leads_flag)
+
+features = np.vstack((features_1, features_2))
+labels = np.concatenate((labels_1, labels_2))
+
+
+ax6 = plt.subplot(326)
+ax6.title.set_text('(f)')
+
+myMorph_avg = np.zeros((4, features.shape[1]))
+for n in range(0,4):
+    myMorph_avg[n] = np.average(features[labels == n], axis=0)
+    plt.plot(myMorph_avg[n], label=label_name[n], linestyle=line_styles[n], linewidth=l_width)
+
+leg = plt.legend(loc='best', ncol=1, shadow=False, fancybox=True)
+leg.get_frame().set_alpha(0.5)
+
+
+
+#plt.show()
+
+plt.savefig('/home/mondejar/graphic.pdf', dpi=None, facecolor='w', edgecolor='w',
+    orientation='portrait', papertype='a4', format='pdf', transparent=True, bbox_inches=None, 
+    pad_inches=0.1, frameon=None)
+
+# A4 figsize=(11.69,8.27)