def graphSpec(self,graphName):
csv_reader = csv.reader(open(self.specFile,'r'))
cnt = 0
frequencies = None
decibels = None
for line in csv_reader:
if cnt == 0:
frequencies = line
cnt += 1
else:
decibels = line
# decibels = [(decibels)]
print frequencies, decibels
cnt=0
frequencies_floats=[]
for x in frequencies:
frequencies_floats.append(float(x))
cnt+=1
cnt=0
decibels_floats=[]
for x in decibels:
decibels_floats.append(float(x))
cnt+=1
N=len(decibels)
ind = np.arange(N)
width = .35
fig = plt.figure()
plt.bar(ind,decibels_floats,width)
plt.xticks(ind+width/2., frequencies,rotation=40, size='small')
plt.xlabel("Frequency (Hz)")
plt.ylabel("Decibels (db)")
plt.show()
def xzCloudPlot(nest,time,plotTemp=True,plotRH=False):
nc = openWRF(nest)
Nx,Ny,Nz,longitude,_lats,_dx,_dy,x_nr,y_nr = getDimensions(nc)
heightground_x,heighthalf_xz = _getHeight(nc, time, Nx, -1, Nz, -1, y_nr)
print 'Model height: ' + str(heightground_x[x_nr])
theta = nc.variables['T'][time,:,y_nr,:] + T_base
P = nc.variables['P'][time,:,y_nr,:] + nc.variables['PB'][time,:,y_nr,:]
T = theta*(P/P_bot)**kappa # Temperatur i halvflatene (Kelvin)
rho = P/(R*T) #[kg/m3]
qcloud_xz = 1000.0*nc.variables['QCLOUD'][time,:,y_nr,:]*rho # regner om til g/m3
qrain_xz = 1000.0*nc.variables['QRAIN'][time,:,y_nr,:]*rho
qsnow_xz = 1000.0*nc.variables['QSNOW'][time,:,y_nr,:]*rho
plt.figure()
plt.set_cmap(cmap_red)
plt.axis([0,Nx-1,0.0,z_max])
print u'Cloud water red, snow blue, rain green ($g/m^3$)'
grid = np.reshape(np.tile(arange(Nx),Nz),(Nz,-1))
plt.contourf(grid, heighthalf_xz, qcloud_xz, alpha=0.9,levels=xz_cloudwater_levels, cmap=cmap_red)#
plt.colorbar()
plt.contourf(grid, heighthalf_xz, qrain_xz, alpha=0.6,levels=xz_rain_levels, cmap=cmap_green)#
plt.colorbar()
plt.contourf(grid, heighthalf_xz, qsnow_xz, alpha=0.6,levels=xz_snow_levels,cmap=cmap_blue)#
plt.colorbar()
if plotTemp:
temp_int = arange(-80.0,50.0,2.0)
cs = plt.contour(grid, heighthalf_xz, T-T_zero, temp_int,colors='black',linestyles='solid')#linewidths=4
plt.clabel(cs, inline=1, fmt='%1.0f', fontsize=12,colors='black')
if plotRH:
rh = _getRH(nc,time,-1,y_nr,T,P)
rh_int = arange(90.,111.,5.)
cs = plt.contour(grid, heighthalf_xz,rh , rh_int, colors='grey')
plt.clabel(cs, inline=1, fmt='%1.0f', fontsize=12, colors='grey')
plt.plot(arange(Nx),heightground_x,color='black')
plt.fill_between(arange(Nx),heightground_x,0,facecolor='lightgrey')
plt.xticks(np.arange(0,Nx,8),np.round(longitude[Ny/2,::8], 1), fontsize='small')
plt.yticks(np.arange(0,z_max,dz), fontsize='small')
plt.xlabel('Lengdegrad')
plt.ylabel(u'Høyde [m]')
plt.show()
plt.close()
def skewTPlot(nest, time):
"""
This is the method to use from the outside
nest: The nesting level of the nc-file from WRF
time: The time for which to plot
"""
nc = openWRF(nest)
_Nx, _Ny, _Nz, _longs, _lats, _dx, _dy, x, y = getDimensions(nc)
plt.figure()
_isotherms()
_isobars()
_dry_adiabats()
_moist_adiabats()
P = nc.variables['P'][time, :, y, x] + nc.variables['PB'][time, :, y, x]
_windbarbs(nc, time, y, x, P)
_temperature(nc, time, y, x, P)
_dewpoint(nc, time, y, x, P)
plt.axis([-40, 50, P_b, P_t])
plt.xlabel('Temperatur ($^{\circ}\! C$) ved 1000hPa')
xticks = np.arange(-40, 51, 5)
plt.xticks(xticks,
['' if tick % 10 != 0 else str(tick) for tick in xticks])
plt.ylabel('Trykk (hPa)')
yticks = np.arange(P_bot, P_t - 1, -10**4)
plt.yticks(yticks, yticks / 100)
sfcT = nc.variables['T2'][time, y, x] - T_zero
sfcP = nc.variables['PSFC'][time, y, x]
sfcW = nc.variables['Q2'][time, y, x]
sfcTd = td(e(sfcW, sfcP))
plt.suptitle('Bakketemp: %4.1f$^{\circ}\! C$ Duggpunkt: %3.1f$^{\circ}\! C$ Trykk: %5.1f hPa' % (sfcT,sfcTd,0.01*sfcP), \
fontsize=10, x = 0.5, y = 0.03)
plt.show()
plt.close()
def skewTPlot(nest,time):
"""
This is the method to use from the outside
nest: The nesting level of the nc-file from WRF
time: The time for which to plot
"""
nc = openWRF(nest)
_Nx,_Ny,_Nz,_longs,_lats,_dx,_dy,x,y = getDimensions(nc)
plt.figure()
_isotherms()
_isobars()
_dry_adiabats()
_moist_adiabats()
P = nc.variables['P'][time,:,y,x] + nc.variables['PB'][time,:,y,x]
_windbarbs(nc,time,y,x,P)
_temperature(nc,time,y,x,P)
_dewpoint(nc,time,y,x,P)
plt.axis([-40,50,P_b,P_t])
plt.xlabel('Temperatur ($^{\circ}\! C$) ved 1000hPa')
xticks = np.arange(-40,51,5)
plt.xticks(xticks,['' if tick%10!=0 else str(tick) for tick in xticks])
plt.ylabel('Trykk (hPa)')
yticks = np.arange(P_bot,P_t-1,-10**4)
plt.yticks(yticks,yticks/100)
sfcT = nc.variables['T2'][time,y,x]-T_zero
sfcP = nc.variables['PSFC'][time,y,x]
sfcW = nc.variables['Q2'][time,y,x]
sfcTd = td(e(sfcW,sfcP))
plt.suptitle('Bakketemp: %4.1f$^{\circ}\! C$ Duggpunkt: %3.1f$^{\circ}\! C$ Trykk: %5.1f hPa' % (sfcT,sfcTd,0.01*sfcP), \
fontsize=10, x = 0.5, y = 0.03)
plt.show()
plt.close()
def tzCloudPlot(nest, plotMetar=False, offset=0):
nc = openWRF(nest)
Nx, Ny, Nz, _longs, _lats, _dx, _dy, x_nr, y_nr = getDimensions(nc)
heightground_t, heighthalf_tz = _getHeight(nc, Nx, Ny, Nz, x_nr, y_nr)
T_tz = np.zeros((Nz, Nt))
qcloud_tz = np.zeros((Nz, Nt))
qice_tz = np.zeros((Nz, Nt))
qsnow_tz = np.zeros((Nz, Nt))
qrain_tz = np.zeros((Nz, Nt))
for t in arange(Nt):
theta = nc.variables['T'][t, :, y_nr, x_nr] + T_base
P = nc.variables['P'][t, :, y_nr,
x_nr] + nc.variables['PB'][t, :, y_nr, x_nr]
T_tz[:,
t] = theta * (P /
P_bot)**kappa # Temperatur i halvflatene (Kelvin)
rho = P[:] / (R * T_tz[:, t]) # regner om til g/m3
qcloud_tz[:,
t] = 1000.0 * nc.variables['QCLOUD'][t, :, y_nr, x_nr] * rho
qice_tz[:, t] = 1000.0 * nc.variables['QICE'][t, :, y_nr, x_nr] * rho
qsnow_tz[:, t] = 1000.0 * nc.variables['QSNOW'][t, :, y_nr, x_nr] * rho
qrain_tz[:, t] = 1000.0 * nc.variables['QRAIN'][t, :, y_nr, x_nr] * rho
for s in [u'Snø', 'Regn']:
plt.figure()
plt.axis([-offset, Nt - 1, 0.0, z_max])
grid = np.reshape(np.tile(arange(Nt), Nz), (Nz, -1))
if (s == u"Snø"):
var = qsnow_tz
cm = cmap_blue
levs = tz_snow_levels
else:
var = qrain_tz
cm = cmap_green
levs = tz_rain_levels
plt.contourf(grid,
heighthalf_tz,
qcloud_tz,
alpha=0.9,
levels=tz_cloudwater_levels,
cmap=cmap_red) #
plt.colorbar()
plt.contourf(grid, heighthalf_tz, var, alpha=0.6, levels=levs,
cmap=cm) #
plt.colorbar()
cs = plt.contour(grid,
heighthalf_tz,
T_tz - T_zero,
temp_int,
colors='black',
linestyles='solid')
plt.clabel(cs, inline=1, fmt='%1.0f', fontsize=12, colors='black')
plt.fill_between(arange(-offset, Nt),
heightground_t[0],
0,
facecolor='lightgrey')
if plotMetar:
_metar()
print s + ' ($g/m^3$)'
plt.xlabel('Timer etter ' + date + 'T00:00Z')
plt.ylabel(u'Høyde [m]')
plt.yticks(np.arange(0, z_max, dz), fontsize='small')
plt.show()
plt.close()
fig = plt.figure()
ax1 = fig.add_subplot(111)
ax1.set_xlim(0, Nt - 1)
ax1.plot(np.sum(qcloud_tz, axis=0), color='black', label=u"skyvann")
ax1.plot(np.sum(qsnow_tz, axis=0), color='green', label=u"snø")
ax1.set_xlabel('Timer etter ' + date + 'T00:00Z')
ax1.set_ylabel(u'Skyvann og snø ($g/m^2$)')
ax2 = ax1.twinx()
ax2.plot(np.sum(qice_tz, axis=0), color='blue', label="is")
ax2.plot(np.sum(qrain_tz, axis=0), color='red', label="regn")
ax2.set_ylabel('Regn og is ($g/m^2$)')
ax1.set_xlim(0, Nt - 1)
ax1.legend(loc='upper left')
ax2.legend(loc='upper right')
plt.show()
plt.close()
import pickle
import matplotlib import pyplot as plt
with open('./met/met_data.pickle', 'rb') as pickle_in:
met_data = pickle.load(pickle_in)
with open('./acc/acc_data.pickle', 'rb') as pickle_in:
acc_data = pickle.load(pickle_in)
met_histo_data = [len(v) for (k,v) in met_data.items()]
acc_histo_data = [len(v) for (k,v) in acc_data.items()]
plt.hist(met_histo_data)
plt.title("Methylation Data")
plt.xlabel("Number of cpg sites within cell")
plt.ylabel("Freq of count")
plt.hist(acc_histo_data)
plt.title("Accessiblity Data")
plt.xlabel("Number of gpc sites within cell")
plt.ylabel("Freq of count")
def from_file_home_outside_calls():
file1 = "/home/sscepano/D4D res/ORGANIZED/SET3/Clustering/usr res/usr_home_calls.tsv"
file2 = "/home/sscepano/D4D res/ORGANIZED/SET3/Clustering/usr res/usr_outside_calls.tsv"
usr_home_c = n.zeros(500001)
usr_outside_c = n.zeros(500001)
i = 0
f1 = open(file1, 'r')
f2 = open(file2, 'r')
# read the file1
for line in f1:
i = i + 1
u, home_c = line.split('\t')
home_c = float(home_c)
u = int(u)
usr_home_c[u] = home_c
# read the file
for line in f2:
i = i + 1
u, outside_c = line.split('\t')
outside_c = float(outside_c)
u = int(u)
usr_outside_c[u] = outside_c
############################################################################################################################
# THIS is to plot pdf of home calls
############################################################################################################################
fig1 = plt.figure(1)
ax = fig1.add_subplot(211)
nn, bins, rectangles = ax.hist(usr_home_c, 100, normed=True)
#plt.plot(nc_distr_pct, 'ro', linewidth=0.5, label= 'pdf Num of calls')
plt.xlabel('NcH, number of calls from the home subprefecture')
plt.ylabel('P(NcH)')
plt.legend()
# # this is if we want loglog lot, otheriwse comment and uncomment next line for regular plot file
plt.yscale('log')
plt.xscale('log')
# figure_name = "/home/sscepano/D4D res/allstuff/rg/pdf rg loglog.png"
#this is a regular plot file, then comment the previous loglog block
#figure_name = "/home/sscepano/D4D res/allstuff/rg/pdf home calls.png"
#print(figure_name)
#plt.savefig(figure_name, format = "png")
############################################################################################################################
# THIS is to plot pdf of outside calls
############################################################################################################################
#fig1 = plt.figure(1)
ax = fig1.add_subplot(212)
nn, bins, rectangles = ax.hist(usr_outside_c, 100, normed=True)
#plt.plot(nc_distr_pct, 'ro', linewidth=0.5, label= 'pdf Num of calls')
plt.xlabel('NcO, number of calls from outside the home subprefecture')
plt.ylabel('P(NcO)')
plt.legend()
# # this is if we want loglog lot, otheriwse comment and uncomment next line for regular plot file
plt.yscale('log')
plt.xscale('log')
figure_name = "/home/sscepano/D4D res/allstuff/rg/pdf home outside calls.png"
# #this is a regular plot file, then comment the previous loglog block
# figure_name = "/home/sscepano/D4D res/allstuff/rg/pdf home outside calls.png"
print(figure_name)
plt.savefig(figure_name, format = "png")
return
# invoke the function for plotting number of calls and frequency probability distribution (percents of users)
#from_file_home_outside_calls()
def from_fq_files_hist_pdf():
file_name= "/home/sscepano/D4D res/ORGANIZED/SET3/Distr of Num and Fq of Calls/new results -- check the same/Users_and_their_total_calls_number.tsv"
file_name2 = "/home/sscepano/D4D res/ORGANIZED/SET3/Distr of Num and Fq of Calls/new results -- check the same/Users_and_their_calling_fq.tsv"
# here we store the num of calls made by a user
usr_and_his_num_calls = n.zeros(500001, dtype=n.int)
nc_distr = n.zeros(max_num_calls, dtype=n.int)
# a loop where we populate those two arrays from the file
i = 0
f = open(file_name, 'r')
# read the file
for line in f:
i = i + 1
u, nc = line.split('\t')
nc = int(nc)
u = int(u)
usr_and_his_num_calls[u] = nc
nc_distr[nc] += 1
mi = min(usr_and_his_num_calls)
mx = max(usr_and_his_num_calls)
print("Minimum number of calls ", mi)
print("Maximum number of calls ", mx)
total_u = float(sum(nc_distr))
print("Total users found ", total_u)
############################################################################################################################
# THIS is to plot number of users pdf
############################################################################################################################
# fig1 = plt.figure(1)
# ax = fig1.add_subplot(111)
# nn, bins, rectangles = ax.hist(usr_and_his_num_calls, 100, normed=True)
#
# plt.xlabel('N, num of calls')
# plt.ylabel('P(N)')
# plt.legend()
#
# # this is if we want loglog lot, otheriwse comment and uncomment next line for regular plot file
# plt.yscale('log')
# plt.xscale('log')
# figure_name = "/home/sscepano/D4D res/allstuff/SET3 frequent callers from python/1/hist of num of calls loglog.png"
#
## #this is a regular plot file, then comment the previous loglog block
## figure_name = "/home/sscepano/D4D res/allstuff/SET3 frequent callers from python/1/hist of num of calls.png"
#
# print(figure_name)
# plt.savefig(figure_name, format = "png")
#
# plt.clf()
###############################################################################################################################
# THIS is to plot fq pdf
###############################################################################################################################
# here we store the num of calls made by a user
usr_and_his_fq = n.zeros(500001)
# a loop where we populate those two arrays from the file
i = 0
f2 = open(file_name2, 'r')
# read the file
for line in f2:
i = i + 1
u, fq = line.split('\t')
fq = float(fq)
u = int(u)
usr_and_his_fq[u] = fq
mi = min(usr_and_his_fq)
mx = max(usr_and_his_fq)
print("Minimum fq of calls ", mi)
print("Maximum fq of calls ", mx)
fig2 = plt.figure(2)
ax = fig2.add_subplot(111)
nn, bins, rectangles = ax.hist(usr_and_his_fq, 100, normed=True)
plt.xlabel('fq of calls')
plt.ylabel('P(fq)')
plt.legend()
figure_name = "/home/sscepano/D4D res/allstuff/SET3 frequent callers from python/1/hist of fq of calls.png"
# # this is if we want loglog lot, otheriwse comment and uncomment next line for regular plot file
# plt.yscale('log')
# plt.xscale('log')
# figure_name = "/home/sscepano/D4D res/allstuff/SET3 frequent callers from python/1/hist of fq of calls loglog.png"
print(figure_name)
plt.savefig(figure_name, format = "png")
return
def yzCloudPlot(nest, time, plotTemp=True, plotRH=False):
nc = openWRF(nest)
Nx, Ny, Nz, _longs, latitude, _dx, _dy, x_nr, y_nr = getDimensions(nc)
heightground_y, heighthalf_yz = _getHeight(nc, time, -1, Ny, Nz, x_nr, -1)
print 'Model height: ' + str(heightground_y[y_nr])
theta = nc.variables['T'][time, :, :, x_nr] + T_base
P = nc.variables['P'][time, :, :, x_nr] + nc.variables['PB'][time, :, :,
x_nr]
T = theta * (P / P_bot)**kappa # Temperatur i halvflatene (Kelvin)
rho = P / (R * T) #[kg/m3]
qcloud_yz = 1000.0 * nc.variables['QCLOUD'][
time, :, :, x_nr] * rho # regner om til g/m3
qrain_yz = 1000.0 * nc.variables['QRAIN'][time, :, :, x_nr] * rho
qsnow_yz = 1000.0 * nc.variables['QSNOW'][time, :, :, x_nr] * rho
plt.figure()
plt.set_cmap(cmap_red)
plt.axis([0, Ny - 1, 0.0, z_max])
print u'Cloud water red, snow blue, rain green ($g/m^3$)'
grid = np.reshape(np.tile(arange(Ny), Nz), (Nz, -1))
plt.contourf(grid,
heighthalf_yz,
qcloud_yz,
alpha=0.9,
levels=xz_cloudwater_levels,
cmap=cmap_red) #
plt.colorbar()
plt.contourf(grid,
heighthalf_yz,
qrain_yz,
alpha=0.6,
levels=xz_rain_levels,
cmap=cmap_green) #
plt.colorbar()
plt.contourf(grid,
heighthalf_yz,
qsnow_yz,
alpha=0.6,
levels=xz_snow_levels,
cmap=cmap_blue) #
plt.colorbar()
if plotTemp:
cs = plt.contour(grid,
heighthalf_yz,
T - T_zero,
temp_int,
colors='black',
linestyles='solid') #linewidths=4
plt.clabel(cs, inline=1, fmt='%1.0f', fontsize=12, colors='black')
if plotRH:
rh = _getRH(nc, time, x_nr, -1, T, P)
rh_int = arange(90., 111., 5.)
cs = plt.contour(grid, heighthalf_yz, rh, rh_int, colors='grey')
plt.clabel(cs, inline=1, fmt='%1.0f', fontsize=12, colors='grey')
plt.plot(arange(Ny), heightground_y, color='black')
plt.fill_between(arange(Ny), heightground_y, 0, facecolor='lightgrey')
plt.xticks(np.arange(0, Ny, 8),
np.round(latitude[::8, Nx / 2], 1),
fontsize='small')
plt.yticks(np.arange(0, z_max, dz), fontsize='small')
plt.xlabel('Breddegrad')
plt.ylabel(u'Høyde [m]')
plt.show()
plt.close()
import sys
sys.path.append('/usr/local/anaconda3/lib/python3.6/site-packages')
from numpy import *
x = linspace(0, 7, 70)
y = sin(x)
f1 = x
s = (
from matplotlib import pyplot as plt
plt.grid
plt.xlabel('x')
plt.ylabel('f(x)')
plt.title('Funkcija $sin(x)$')
plt.plot(x, y2, color = "#7FFF00")
plt.show()
validation_data = validation_generator,
validation_steps = validation_generator.samples // batch_size,
epochs = num_epochs)
View the Loss History
We tracked average training and validation loss for each epoch. We can plot these to see where the levels of loss converged, and to detect overfitting (which is indicated by a continued drop in training loss after validation loss has levelled out or started to increase.
from matplotlib import pyplot as plt
epoch_nums = range(1,num_epochs+1)
training_loss = history.history["loss"]
validation_loss = history.history["val_loss"]
plt.plot(epoch_nums, training_loss)
plt.plot(epoch_nums, validation_loss)
plt.xlabel('epoch')
plt.ylabel('loss')
plt.legend(['training', 'validation'], loc='upper right')
plt.show()
'''
Using the Trained Model
Now that we've trained the model, we can use it to predict the class of an image.
'''
def predict_image(classifier, image_array):
import numpy as np
# We need to format the input to match the training data
def from_file_radius_gyr3(file_name, subpref):
users_list = rd.read_in_subpref_users(subpref)
total = float(rd.read_in_subpref_num_users()[subpref])
if total > 0:
nits = []
its = []
# a loop where we populate those two arrays from the file
i = 0
f = open(file_name, 'r')
# read the file
for line in f:
i = i + 1
it, nit = line.split('\t')
nit = float(nit)
it = int(it)
if users_list[it] == 1:
nit = int(nit)
nits.append(nit)
its.append(it)
mi = min(nits)
mx = max(nits)
print("Minimum radius of gyr ", mi)
print("Maximum radius of gyr ", mx)
total_nit = float(sum(nits))
print("Total radius of gyr ", total_nit)
pdf_nits = defaultdict(int)
for j in range(0, len(nits)):
pdf_nits[nits[j]] += 1
ordered = OrderedDict(sorted(pdf_nits.items(), key=lambda t: t[0]))
nits7s = []
its7s = []
test = 0
#total = 500000.0
for j in ordered.iterkeys():
nits7s.append(ordered[j]/total)
test += ordered[j]/total
its7s.append(j)
print test
############################################################################################################################
# THIS is to plot number of users pdf
############################################################################################################################
plt.figure(7)
plt.plot(its7s, nits7s, 'o', linewidth=0.5, label= 'distribution of Rg')
plt.xlabel('rg [km]')
plt.ylabel('P(rg)')
plt.legend()
# this is if we want loglog lot, otheriwse comment and uncomment next line for regular plot file
plt.yscale('log')
plt.xscale('log')
figure_name = "/home/sscepano/D4D res/allstuff/rg/1/rg_" + str(subpref) + ".png"
print(figure_name)
plt.savefig(figure_name, format = "png", dpi=300)
plt.clf()
return
def from_file_radius_gyr(file_name):
usr_rg = nn.zeros(500001)
# a loop where we populate those two arrays from the file
i = 0
f = open(file_name, 'r')
# read the file
for line in f:
i = i + 1
u, rg = line.split('\t')
rg = float(rg)
u = int(u)
usr_rg[u] = rg
mi = min(usr_rg)
mx = max(usr_rg)
print("Minimum number of calls ", mi)
print("Maximum number of calls ", mx)
# total_u = float(sum(nc_distr))
# print("Total users found ", total_u)
# test_file_out = "/home/sscepano/D4D res/allstuff/SET3 frequent callers from python/1/Obtained_num_calls_and_its_pct.tsv"
# fto = open(test_file_out,"w")
# for j in range(0, max_num_calls):
# nc_distr_pct[j] = (nc_distr[j] / total_u) * 100
# fto.write(str(j) + '\t' + str(nc_distr_pct[j]) + '\n')
# # I was just checking that the total percents sums up to 100 and they do but it looked funny as we have so small values
# total = 0
# for i in range (max_num_calls):
# total += percent_users[i]
# print ("Check ", total)
############################################################################################################################
# THIS is to plot number of users pdf
############################################################################################################################
fig1 = plt.figure(1)
ax = fig1.add_subplot(111)
n, bins, rectangles = ax.hist(usr_rg, 100, normed=True)
#plt.plot(nc_distr_pct, 'ro', linewidth=0.5, label= 'pdf Num of calls')
plt.xlabel('rg [km]')
plt.ylabel('P(rg)')
plt.legend()
# # this is if we want loglog lot, otheriwse comment and uncomment next line for regular plot file
plt.yscale('log')
plt.xscale('log')
figure_name = "/home/sscepano/D4D res/allstuff/rg/pdf rg loglog.png"
# #this is a regular plot file, then comment the previous loglog block
# figure_name = "/home/sscepano/D4D res/allstuff/rg/pdf rg.png"
print(figure_name)
plt.savefig(figure_name, format = "png")
################################################################################################################################
## THIS is to plot fq pdf
################################################################################################################################
#
# plt.figure(2)
#
# fq = []
#
# for j in range(max_num_calls):
# fq.append( float(j / 3360.0))
#
# ffq = []
#
# for j in range(max_num_calls):
# ffq.append(nc_distr_pct[j])
#
#
## test_file_out2 = "/home/sscepano/D4D res/allstuff/SET3 frequent callers from python/1/Calculated_fq_calls_and_its_pct2.tsv"
## fto2 = open(test_file_out2,"w")
## for j in range(0, max_num_calls):
## fto2.write(str(fq[j]) + '\t' + str(ffq[j]) + '\n')
#
#
# # Finally understood here -- when I give two arrays: x, y (at least append values IN ORDER like here) -- pyplot will plot y versus x
# plt.plot(fq, ffq, 'g.', linewidth=0.3, label= 'pdf fq of calls')
#
# plt.xlabel('fq of calls')
# plt.ylabel('% Users')
# plt.legend()
#
# # this is if we want loglog lot, otheriwse comment and uncomment next line for regular plot file
# plt.yscale('log')
# plt.xscale('log')
# figure_name = "/home/sscepano/D4D res/allstuff/SET3 frequent callers from python/1/SET3 distr of fq of calls loglog.png"
#
## # this is a regular plot file, then comment the previous loglog block
## figure_name = "/home/sscepano/D4D res/allstuff/SET3 frequent callers from python/1/SET3 distr of fq of calls.png"
#
# print(figure_name)
# plt.savefig(figure_name, format = "png")
return
import scikits.audiolab as audio
import pyplot as plt
#%pylab inline
input_signal, sampling_rate, enc = audio.wavread("minombre.wav")
print (input_signal[0:10]), sampling_rate, enc
time_array = arange(0, len(input_signal)/float(sampling_rate), 1/float(sampling_rate))
plt.plot(time_array[0:4000], input_signal[0:4000])
plt.xlabel("time(s)", fontsize=20)
plt.ylabel("Amplitude", fontsize=20)
save("mi_voz", ext="png", close=False, verbose=True)
def save_to_plot(avg_usr_traj):
nits = []
its = []
# a loop where we populate those two arrays from the file
i = 0
# f = open(file_name, 'r')
# # read the file
for usr in range(500001):
i = i + 1
nits.append(int(avg_usr_traj[usr]))
its.append(usr)
mi = min(nits)
mx = max(nits)
print("Minimum radius of gyr ", mi)
print("Maximum radius of gyr ", mx)
total_nit = float(sum(nits))
print("Total radius of gyr ", total_nit)
pdf_nits = defaultdict(int)
for j in range(0, len(nits)):
pdf_nits[nits[j]] += 1
ordered = OrderedDict(sorted(pdf_nits.items(), key=lambda t: t[0]))
nits7s = []
its7s = []
test = 0
for j in ordered.iterkeys():
nits7s.append(ordered[j]/500000.0)
test += ordered[j]/500000.0
its7s.append(j)
print test
############################################################################################################################
# THIS is to plot number of users pdf
############################################################################################################################
plt.figure(7)
plt.plot(its7s, nits7s, 'o', linewidth=0.5, label= 'distribution of Rg')
plt.xlabel('rg [km]')
plt.ylabel('P(rg)')
plt.legend()
# this is if we want loglog lot, otheriwse comment and uncomment next line for regular plot file
plt.yscale('log')
plt.xscale('log')
figure_name = "/home/sscepano/D4D res/allstuff/traj/avg daily/avg_daily_traj_total.png"
print(figure_name)
plt.savefig(figure_name, format = "png", dpi=300)
return
8/13: colors = ['#348ABD','#A60628']
8/14: import matplotlib as plt
8/15: plt.bar
8/16: from matplotlib import pyplot as plt
8/17: pmf
8/18: pm
8/19: pm(a,lambda_[0])
8/20: poi.pmf(a,lambda_[0])
8/21: poi.pmf(a,lambda_[0])
8/22: plt.bar(a,poi.pmf(a,lambda_[0]),color=colors[0],lable="$\lambda = %.1f$" % lambda_[0],alpha=0.60,edgecolor=colors[0],lw="3")
8/23: plt.bar(a,poi.pmf(a,lambda_[0]),color=colors[0],label="$\lambda = %.1f$" % lambda_[0],alpha=0.60,edgecolor=colors[0],lw="3")
8/24: plt.bar(a,poi.pmf(a,lambda_[1]),color=colors[1],label="$\lambda = %.1f$" % lambda_[1],alpha=0.60,edgecolor=colors[1],lw="3")
8/25: plt.xticks( a + 0.4,a)
8/26: plt.legend()
8/27: plt.ylabel("Probability of $k$")
8/28: plt.xlabel("$k$")
8/29: plt.title("Probability mass funciton of a Poisson random \$\lambda$ values")
8/30: plt.show()
9/1: from pymc import DiscreteUniform, Exponential, deterministic, Poisson, Uniform
9/2: import numpy as np
9/3:
disasters_array = \
np.array([ 4, 5, 4, 0, 1, 4, 3, 4, 0, 6, 3, 3, 4, 0, 2, 6,
3, 3, 5, 4, 5, 3, 1, 4, 4, 1, 5, 5, 3, 4, 2, 5,
2, 2, 3, 4, 2, 1, 3, 2, 2, 1, 1, 1, 1, 3, 0, 0,
1, 0, 1, 1, 0, 0, 3, 1, 0, 3, 2, 2, 0, 1, 1, 1,
0, 1, 0, 1, 0, 0, 0, 2, 1, 0, 0, 0, 1, 1, 0, 2,
3, 3, 1, 1, 2, 1, 1, 1, 1, 2, 4, 2, 0, 0, 1, 4,
0, 0, 0, 1, 0, 0, 0, 0, 0, 1, 0, 0, 1, 0, 1])
9/4: switchPoint = DiscreteUniform('switchpoint', lower=0, upper=110, doc='Switchpoint[year]')
9/5: early_mean = Exponential('early_mean', beta=1.)
def from_file_num_calls(file_name):
# here we store the num of calls made by a user
usr_and_his_num_calls = n.zeros(500001, dtype=n.int)
# here we store the fq of calls made by a user, but we calculate it from the num_calls
fq_distr = n.zeros(max_num_calls)
# from the previous array obtained by counting users who made the same total number of calls
nc_distr = n.zeros(max_num_calls, dtype=n.int)
# here we just save percents of users
nc_distr_pct = n.zeros(max_num_calls)
# a loop where we populate those two arrays from the file
i = 0
f = open(file_name, 'r')
# read the file
for line in f:
i = i + 1
u, nc = line.split('\t')
nc = int(nc)
u = int(u)
usr_and_his_num_calls[u] = nc
nc_distr[nc] += 1
mi = min(usr_and_his_num_calls)
mx = max(usr_and_his_num_calls)
print("Minimum number of calls ", mi)
print("Maximum number of calls ", mx)
total_u = float(sum(nc_distr))
print("Total users found ", total_u)
# test_file_out = "/home/sscepano/D4D res/allstuff/SET3 frequent callers from python/1/Obtained_num_calls_and_its_pct.tsv"
# fto = open(test_file_out,"w")
for j in range(0, max_num_calls):
nc_distr_pct[j] = (nc_distr[j] / total_u)
# fto.write(str(j) + '\t' + str(nc_distr_pct[j]) + '\n')
# # I was just checking that the total percents sums up to 100 and they do but it looked funny as we have so small values
# total = 0
# for i in range (max_num_calls):
# total += percent_users[i]
# print ("Check ", total)
############################################################################################################################
# THIS is to plot number of users pdf
############################################################################################################################
plt.figure(1)
plt.plot(nc_distr_pct, 'r', linewidth=0.5, label= 'distribution of N')
plt.xlabel('N, num of calls')
plt.ylabel('% Users')
plt.legend()
# # this is if we want loglog lot, otheriwse comment and uncomment next line for regular plot file
plt.yscale('log')
plt.xscale('log')
# figure_name = "/home/sscepano/D4D res/allstuff/SET3 frequent callers from python/1/SET3 distr of num of calls loglog.png"
#this is a regular plot file, then comment the previous loglog block
figure_name = "/home/sscepano/D4D res/allstuff/SET3 frequent callers from python/1/distr of num of calls2.png"
print(figure_name)
plt.savefig(figure_name, format = "png", pdi=300)
#plt.show()
###############################################################################################################################
# THIS is to plot fq pdf
###############################################################################################################################
plt.figure(2)
# fq = []
#
# for j in range(max_num_calls):
# fq.append( float(j / 3360.0))
#
# ffq = []
#
# for j in range(max_num_calls):
# ffq.append(nc_distr_pct[j])
# for j in range(0, max_num_calls):
# nc_distr_pct[j] = (nc_distr[j] / total_u)
fq = []
for j in range(max_num_calls):
fq.append( float(j / 3360.0))
ffq = []
for j in range(max_num_calls):
ffq.append(nc_distr_pct[j])
# test_file_out2 = "/home/sscepano/D4D res/allstuff/SET3 frequent callers from python/1/Calculated_fq_calls_and_its_pct2.tsv"
# fto2 = open(test_file_out2,"w")
# for j in range(0, max_num_calls):
# fto2.write(str(fq[j]) + '\t' + str(ffq[j]) + '\n')
# Finally understood here -- when I give two arrays: x, y (at least append values IN ORDER like here) -- pyplot will plot y versus x
plt.plot(fq, ffq, 'g', linewidth=0.3, label= 'distribution of Fq')
plt.xlabel('Fq of calls')
plt.ylabel('% Users')
plt.legend()
# this is if we want loglog lot, otheriwse comment and uncomment next line for regular plot file
plt.yscale('log')
plt.xscale('log')
figure_name = "/home/sscepano/D4D res/allstuff/SET3 frequent callers from python/1/distr of fq of calls2.png"
# # this is a regular plot file, then comment the previous loglog block
# figure_name = "/home/sscepano/D4D res/allstuff/SET3 frequent callers from python/1/SET3 distr of fq of calls.png"
print(figure_name)
plt.savefig(figure_name, format = "png")
return
# "python.pythonPath": "C:\\tmp\\cx1964ReposPlot\\env_python3_plot\\Scripts
#
# indien mathplotlib nog niet geinstalleerd is run:
# pip install mathplotlib
#
# Nadat een grafiek wordt gelsoten wordt de volgende getoond.
#
import pyplot as plt
# Voorbeeld1: 1 grafieken in een afbeelding
x = [5,8,10]
y = [12,16,6]
plt.plot(x,y)
plt.title('Titel van de grafiek voorbeeld1')
plt.ylabel('Y-as')
plt.xlabel('X-as')
plt.show()
# Voorbeeld2: 2 grafieken in een afbeelding
#from matplotlib import pyplot as plt
from matplotlib import style
style.use('ggplot')
x1 = [0,1,2,3,4,5]
y1 = [0,1,4,9,16,25]
x2 = x1
y2 = [5,7,9,11,13,15]
# can plot specifically, after just showing the defaults:
plt.plot(x1,y1,linewidth=5, label='$f1(x)=x^2$', color='red') # label maken gebruik van Latex express
plt.plot(x2,y2,linewidth=5, label='$f2(x)=2x+5$', color='green') # label maken gebruik van Latex express
plt.title('Titel van de grafiek voorbeeld2')
def from_files_usr_movements():
file_name1 = "/home/sscepano/D4D res/allstuff/USER GRAPHS stats/user_number_of_edges_v1.tsv"
file_name2 = "/home/sscepano/D4D res/allstuff/USER GRAPHS stats/user_number_of_nodes_v1.tsv"
file_name3 = "/home/sscepano/D4D res/allstuff/USER GRAPHS stats/user_number_of_displacements_v1.tsv"
nits = []
its = []
# a loop where we populate those two arrays from the file
i = 0
f = open(file_name3, "r")
# read the file
for line in f:
i = i + 1
it, nit = line.split("\t")
nit = int(nit)
it = int(it)
nit = int(nit)
nits.append(nit)
its.append(it)
mi = min(nits)
mx = max(nits)
print ("Minimum # edges ", mi)
print ("Maximum # edges ", mx)
total_nit = float(sum(nits))
print ("Total # edges ", total_nit)
pdf_nits = defaultdict(int)
for j in range(0, len(nits)):
pdf_nits[nits[j]] += 1
ordered = OrderedDict(sorted(pdf_nits.items(), key=lambda t: t[0]))
nits7s = []
its7s = []
test = 0
for j in ordered.iterkeys():
nits7s.append(ordered[j] / 500000.0)
test += ordered[j] / 500000.0
its7s.append(j)
print test
############################################################################################################################
# THIS is to plot number of users pdf
############################################################################################################################
plt.figure(7)
plt.plot(its7s, nits7s, "o", linewidth=0.5, label="distr. of # displacements ")
plt.xlabel("# displacements")
plt.ylabel("P(# displacements)")
plt.legend()
# this is if we want loglog lot, otheriwse comment and uncomment next line for regular plot file
plt.yscale("log")
plt.xscale("log")
# figure_name1 = "/home/sscepano/D4D res/allstuff/USER GRAPHS stats/usr_num_edges.png"
# figure_name2 = "/home/sscepano/D4D res/allstuff/USER GRAPHS stats/usr_num_nodes.png"
figure_name3 = "/home/sscepano/D4D res/allstuff/USER GRAPHS stats/usr_num_displacements.png"
print (figure_name3)
plt.savefig(figure_name3, format="png", dpi=300)
return