def plot_histo_gram(f_name, rg):
import matplotlib.pyplot as plt
sfig = 420
plt.figure()
patterns, hist_record = zload(f_name)
for hist, bins in hist_record:
hist = hist[1:]
bins = bins[1:]
bins /= 3600
width = 0.7 * (bins[1] - bins[0])
plt.subplot(sfig)
center = (bins[:-1] + bins[1:]) / 2
plt.bar(center, hist, align='center', width=width)
plt.xlim([r / 3600.0 for r in rg])
plt.show()
def plot_histo_gram(f_name, rg):
import matplotlib.pyplot as plt
sfig = 420
plt.figure()
patterns, hist_record = zload(f_name)
for hist, bins in hist_record:
hist = hist[1:]
bins = bins[1:]
bins /= 3600
width = 0.7*(bins[1]-bins[0])
plt.subplot(sfig)
center = (bins[:-1]+bins[1:])/2
plt.bar(center, hist, align = 'center', width = width)
plt.xlim([r/3600.0 for r in rg])
plt.show()
ls = linestyles[i % len(linestyles)]
plt.plot(x, mat[:, i], linestyle=ls, marker=style,
color=color, markersize=4)
if __name__ == "__main__":
parser = argparse.ArgumentParser(description="Probability Law Identification")
parser.add_argument('sscheck', help='self check file')
parser.add_argument('lamb', type=float, help='up bound for the threshold')
parser.add_argument('entro', help="['mf', 'mb'] entropy type")
parser.add_argument('pic_name', default=None, help="output picture name")
args = parser.parse_args()
data = zload(args.sscheck)
# import ipdb;ipdb.set_trace()
I_rec = data['I_rec']
n = I_rec[0].shape[0] # no. of PLs
m = len(I_rec) # no. of windows
seq_map = {
'mf':0,
'mb':1
}
# Convert I_rec to the weight of bipartie graph
D = np.zeros((m, n))
for j in xrange(n):
for i in xrange(m):
D[i, j] = I_rec[i][j, seq_map[args.entro]]