e = at.ECG(shape=(200, 200), probe_height=3, m=args.mode) # Assuming shape/probe height doesn't change. file_name = args.output #nput("Name of output file: ") #print("Nu Value:") nu = args.nu #float(input()) print(file_name, nu, e.mode, Iterations) # print(nu) h5f = h5py.File('%s.h5' % file_name, 'w') for index in range(Iterations): #nu = (np.random.rand() / 5) + 0.1 start_time1 = time.time() index_grp = h5f.create_group('Index: %s' % index) a = fp.Heart(nu=nu, fakedata=True) crit_position = np.random.randint(40000) y_rand, x_rand = np.unravel_index(crit_position, (200, 200)) a.set_pulse(60, [[y_rand], [x_rand]]) raw_data = a.propagate(1260) converted_data = list() grid = np.zeros(a.shape) convert(raw_data, converted_data) # Saving the critical circuit position index_grp.create_dataset('Crit Position', data=crit_position) index_grp.create_dataset('Nu', data=nu) ecg = e.solve(converted_data[1111:]) index_grp.create_dataset('ECG', data=ecg) index_grp.create_dataset('Probe Positions', data=e.probe_position)
import propagate_singlesource as ps import analysis_theano as at import numpy as np import matplotlib.pyplot as plt from Functions import sampling_convert # import pyqtgraph as pg # import pyqtgraph.ptime as ptime # from pyqtgraph.Qt import QtCore, QtGui a = ps.Heart(fakedata=True) a.set_pulse(220, [[100], [100]]) e = at.ECG(shape=(200, 200), probe_height=3) raw_data = a.propagate(400) converted_data = list() grid = np.zeros(a.shape) sampling_convert(raw_data, converted_data, shape=a.shape, rp=a.rp, animation_grid=grid) ecg = e.solve(converted_data[100:]) probe_positions = e.probe_position print probe_positions print "Plotting..." fig = plt.figure() for index, i in enumerate(ecg): plt.plot(range(len(i)), i,
import analysis_theano as at from Functions import ani_convert, feature_extract_multi_test_rt, multi_feature_compile_rt import propagate_singlesource as ps args = sys.argv # Loading in Machine Learning models ##################################### y_classifier_full = joblib.load(args[1]) y_class = joblib.load(args[2]) x_classifier_full = joblib.load(args[3]) x_class = joblib.load(args[4]) ##################################### # Initialising the Heart structure a = ps.Heart(nu=0.2, delta=0.0, fakedata=True) # Randomises the rotor x,y position cp_x_pos = randint(30, 169) cp_y_pos = randint(0, 199) a.set_pulse(60, [[cp_y_pos], [cp_x_pos]]) tissue_reset = False # Initialising ECG recording (randomises the probe x,y position) current_ecg_x_pos = randint(20, 179) current_ecg_y_pos = randint(0, 199) ecg_processing = at.ECG(centre=(current_ecg_y_pos, current_ecg_x_pos), m='g_single') # Initialising the animation window app = QtGui.QApplication([]) win = pg.GraphicsWindow(border=True)
from Functions import ani_convert print "Propagator types: [Normal, Single Crit, Double Point]" propagate_choice = raw_input("Propagate type: ") a = None if propagate_choice == 'Normal': import basic_propagate as bc nu_value = float(raw_input('Choose a Nu value: ')) a = bc.Heart(nu_value) a.set_pulse(220) if propagate_choice == 'Single Crit': import propagate_singlesource as ps a = ps.Heart(nu=0.2, fakedata=True) x_pos = int(raw_input("Crit x position: ")) y_pos = int(raw_input("Crit y position: ")) a.set_pulse(60, [[y_pos], [x_pos]]) if propagate_choice == 'Double Point': import propagate_singlesource as ps nu_value = float(raw_input('Choose a Nu value: ')) a = ps.Heart(nu=nu_value, fakedata=True) x_pos = input("Crit x position: ") y_pos = input("Crit y position: ") a.set_pulse(60, [[y_pos], [x_pos]]) e = at.ECG_single(a.shape, 3) app = QtGui.QApplication([])