def case_age_model():
group = [
['S_kids', 'S_normal', 'S_risk'],
['E_kids', 'E_normal', 'E_risk'],
['I_kids', 'I_normal', 'I_risk'],
['I2'],
['I3'],
['R'],
['D']
]
label = {
"S_kids": "S",
"E_kids": "E",
"I_kids": "I_1",
"I2": "I_2",
"I3": "I_3"
}
data = get_results(
simulation_age_model, (N, B_KIDS, G_KIDS, P_KIDS, B_NORMAL, G_NORMAL,
P_NORMAL, B_NORMAL, G_NORMAL, P_NORMAL), NUM_SIM
)
times, avg, std = fill_around_std(
kolor_palette, data, 1, states=group, labels=label
)
save_data([times, avg, std], (N, NUM_SIM,))
def _get_visitorid(self):
visitor_id = util.load_data(self.addon, self.VISITOR_FILE)
if visitor_id is None:
visitor_id = str(randint(0, 0x7fffffff))
util.save_data(self.addon, self.VISITOR_FILE, visitor_id)
return visitor_id
async def introStartCommand(id,guildID):
global currentintrosessionReport
currentintrosessionReport[str(id)] = {}
currentintrosessionReport[str(id)]['confirmed'] = "False"
currentintrosessionReport[str(id)]['guildID'] = str(guildID)
currentintrosessionReport[str(id)]['createdTime'] = str(time.time())
#currentintrosessionReport[str(id)]['createdTime'] = "9999999999999"
currentintrosessionReport[str(id)]['answers'] = []
currentintrosessionReport[str(id)]['lastAnswerTime'] = str(int(time.time()))
guild = discord.utils.get(botAcc.guilds, id=int(guildID))
member = discord.utils.get(guild.members, id=int(id))
if member is None:
print("Member none")
return False
try:
await member.send("Please answer each question.")
except:
return False
messageToSendRAW = questions[0]
messageToSend = messageToSendRAW.split("{")[0]
messageSent = await member.send(messageToSend)
util.save_data(currentintrosessionReport, "currentintrosessionReport")
def get_data_and_show(kind, limit, use_index, is_show, dis_tick, data_kind,
xlabel, ylabel, line_color, fig_color, funciton, x_str,
y_str, title, figsize, *args, **kwargs):
data, orient = get_data(data_kind, limit, *args, **kwargs)
if is_show and not data.empty:
dd = DataDisplay()
if kind == 'pie' and isinstance(data, pd.core.frame.DataFrame):
return dumps({'error_kind': "DataFrame don't support pie"})
dd.show_graphic(data,
kind,
use_index,
xlabel=xlabel,
ylabel=ylabel,
line_color=line_color,
fig_color=fig_color,
funciton=funciton,
x_str=x_str,
y_str=y_str,
title=title,
figsize=figsize,
dis_tick=dis_tick)
if request.args.get('save'):
save_data(data, data_kind, request.args.get('save'),
request.args.get('pt'))
return data.to_json(orient=orient)
def minify(filepath=None):
"""Minify the file by removing unnecessary whitespace characters.
Also remove properties which are not necessary.
Currently will ignore properties which are idenitical to
the default properties of an item."""
if filepath is None:
filepath = LATEST_DATA_SET_PATH
data_set = load_data(filepath)
basic_data = data_set["basic"]
items = data_set["data"]
smaller_items = dict()
for item, data in items.items():
smaller_item = dict()
for key, value in data.items():
if key in basic_data and value == basic_data[key]:
# This property is the same as the default
# It doesn't need to be re-specified
continue
smaller_item[key] = value
smaller_items[item] = smaller_item
data_set["data"] = smaller_items
filename, extension = filepath.rsplit(".", 1)
new_filepath = filename + "_" + FILE_AMMENDMENT + "." + extension
save_data(data_set, new_filepath)
def get_visitorid():
visitor_id = util.load_data(addon, VISITOR_FILE)
if visitor_id is False:
from random import randint
visitor_id = str(randint(0, 0x7fffffff))
util.save_data(addon, VISITOR_FILE, visitor_id)
return visitor_id
def case_no_intervention():
label = {
"I1": "I_1",
"I2": "I_2",
"I3": "I_3"
}
data = get_results(simulation_no_intervention, (N,), NUM_SIM)
times, avg, std = fill_around_std(color_palette, data, 1, labels=label)
save_data([times, avg, std], (N, NUM_SIM))
def case_average_second_wave():
data = get_results(simulation_second_wave, tuple(), NUM_SIM)
labels = {
"I1": "I_{1}", "I2": "I_{2}", "I3": "I_{3}"
}
times, avg, std = fill_around_std(
color_palette, data, 1, labels=labels
)
save_data([times, avg, std], (N, NUM_SIM))
def hk(s, op_eps, max_rounds, eps=1e-6, conv_stop=True, save=False):
'''Simulates the model of Hegselmann-Krause.
This model does nto require an adjacency matrix. Connections between
nodes are calculated depending on the proximity of their opinions.
Args:
s (1xN numpy array): Initial opinions (intrinsic beliefs) vector
op_eps: ε parameter of the model
max_rounds (int): Maximum number of rounds to simulate
eps (double): Maximum difference between rounds before we assume that
the model has converged (default: 1e-6)
conv_stop (bool): Stop the simulation if the model has converged
(default: True)
save (bool): Save the simulation data into text files
Returns:
A txN vector of the opinions of the nodes over time
'''
N, z, max_rounds = preprocessArgs(s, max_rounds)
z_prev = z.copy()
opinions = np.zeros((max_rounds, N))
opinions[0, :] = s
for t in trange(1, max_rounds):
for i in range(N):
# The node chooses only those with a close enough opinion
friends_i = np.abs(z_prev - z_prev[i]) <= op_eps
z[i] = np.mean(z_prev[friends_i])
opinions[t, :] = z
z_prev = z.copy()
if conv_stop and \
norm(opinions[t - 1, :] - opinions[t, :], np.inf) < eps:
print('Hegselmann-Krause converged after {t} rounds'.format(t=t))
break
if save:
timeStr = datetime.now().strftime("%m%d%H%M")
simid = 'hk' + timeStr
save_data(simid,
N=N,
max_rounds=max_rounds,
eps=eps,
rounds_run=t + 1,
s=s,
op_eps=op_eps,
opinions=opinions[0:t + 1, :])
return opinions[0:t + 1, :]
def case_average_time():
t = 5
output = get_results(
simulation_max_time, (N, t), NUM_SIM
)
timepoints, avg, std = fill_around_std(
color_palette, output, 1
)
save_data([timepoints, avg, std], (N, NUM_SIM, t))
def friedkinJohnsen(A, s, max_rounds, eps=1e-6, conv_stop=True, save=False):
'''Simulates the Friedkin-Johnsen (Kleinberg) Model.
Runs a maximum of max_rounds rounds of the Friedkin-Jonsen model. If the
model converges sooner, the function returns. The stubborness matrix of
the model is extracted from the diagonal of matrix A.
Args:
A (NxN numpy array): Adjacency matrix (its diagonal is the stubborness)
s (1xN numpy array): Initial opinions (intrinsic beliefs) vector
max_rounds (int): Maximum number of rounds to simulate
eps (double): Maximum difference between rounds before we assume that
the model has converged (default: 1e-6)
conv_stop (bool): Stop the simulation if the model has converged
(default: True)
save (bool): Save the simulation data into text files
Returns:
A txN vector of the opinions of the nodes over time
'''
N, z, max_rounds = preprocessArgs(s, max_rounds)
B = np.diag(np.diag(A)) # Stubborness matrix of the model
A_model = A - B # Adjacency matrix of the model
opinions = np.zeros((max_rounds, N))
opinions[0, :] = z
for t in trange(1, max_rounds):
z = A_model.dot(z) + B.dot(s)
opinions[t, :] = z
if conv_stop and \
norm(opinions[t - 1, :] - opinions[t, :], np.inf) < eps:
print('Friedkin-Johnsen converged after {t} rounds'.format(t=t))
break
if save:
timeStr = datetime.now().strftime("%m%d%H%M")
simid = 'fj' + timeStr
save_data(simid,
N=N,
max_rounds=max_rounds,
eps=eps,
rounds_run=t + 1,
A=A,
s=s,
opinions=opinions[0:t + 1, :])
return opinions[0:t + 1, :]
def stop_recording(self):
print("Stopping recording.")
self.recording = False
self.data["duration"] = int(time.time()-self.startTime)
self.data["videoFile"] = self.videoData
self.data["__type__"] = "irVideoRecording"
self.camera.stop_recording()
util.save_data(self.data, self.recordingFolder)
self.data = {}
def get_detail(shopID='l4k6FPkNSo8IQyF7',shopname='海底捞火锅'):
shop_url = "http://www.dianping.com/shop/" + shopID + "/review_all/"
for i in range(1, 2):
url = shop_url + 'p' + str(i)
print(url)
html = getHTMLText(url)
infoList = parsePage(html,shopID,shopname)
for info in infoList:
util.save_data(info)
print('成功爬取第{}页数据,有评论{}条'.format(i,len(infoList)))
def meetFriend_matrix_nomem(A, max_rounds, norm_type=2, save=False):
'''Simulates the random meeting model (matrix version).
Runs a maximum of max_rounds rounds of the "Meeting a Friend" model. If the
model converges sooner, the function returns. The stubborness matrix of
the model is extracted from the diagonal of matrix A. The function returns
the distance from the equilibrium of the Friedkin-Johnsen when the process
is complete. Uses less memory than the full model.
Args:
A (NxN numpy array): Weights matrix (its diagonal is the stubborness)
max_rounds (int): Maximum number of rounds to simulate
norm_type: The norm type used to calculate the difference from the
equilibrium
save (bool): Save the simulation data into text files
Returns:
The final distance from the equlibrium of the Friedkin-Johnsen model,
using the specified norm.
'''
max_rounds = int(max_rounds)
N = A.shape[0]
B = np.diag(np.diag(A))
nonzero_ids = [np.nonzero(A[i, :])[0] for i in xrange(A.shape[0])]
equilibrium_matrix = np.dot(inv(np.eye(N) - (A - B)), B)
R, Q = rand_matrices(A, 1, nonzero_ids)
R = R + Q
for t in trange(2, max_rounds + 2):
A_t, B_t = rand_matrices(A, t, nonzero_ids)
R = A_t.dot(R) + B_t
distance = norm(R - equilibrium_matrix, ord=norm_type)
if save:
timeStr = datetime.now().strftime("%m%d%H%M")
simid = 'mf' + timeStr
save_data(simid,
N=N,
max_rounds=max_rounds,
eps=eps,
rounds_run=max_rounds,
A=A,
distance=distance,
norm=norm_type)
return distance
def case_max_time():
time = np.linspace(0, 15, 30)
group = ["I3", "I1", "I2"]
arguments = [(simulation_max_time, (N, t), group) for t in time]
with Pool() as p:
y = p.starmap(max_sim, arguments)
save_data([time.tolist(), y], (N, NUM_SIM))
plt.xlabel("point in time social distancing takes place")
plt.ylabel("maximum count of infected people")
plt.scatter(time, y, c="red")
plt.show()
def prettify(filepath=None):
"""Minifies the file by removing unnecessary whitespace
Also remove properties which are the same as the default properties of an item"""
if filepath is None:
filepath = LATEST_DATA_SET_PATH
data_set = load_data(filepath)
filename, extension = filepath.rsplit(".", 1)
new_filepath = filename + "_" + FILE_AMMENDMENT + "." + extension
save_data(data_set, new_filepath, pretty=True)
def OnFormatChoice(self, event):
img_format = event.GetString()
shared.options['format'] = img_format
save_data(shared.options)
if self.img_panel:
self.img_panel.Hide()
self.img_panel = self.img_gui_dict.get(img_format)
if self.img_panel:
self.img_panel.Show()
self.sizer.Layout()
self.Refresh()
def deGroot(A, s, max_rounds, eps=1e-6, conv_stop=True, save=False):
'''Simulates the DeGroot Model.
Runs a maximum of max_rounds rounds of the DeGroot model. If the model
converges sooner, the function returns.
Args:
A (NxN numpy array): Adjacency matrix
s (1xN numpy array): Initial opinions vector
max_rounds (int): Maximum number of rounds to simulate
eps (double): Maximum difference between rounds before we assume that
the model has converged (default: 1e-6)
conv_stop (bool): Stop the simulation if the model has converged
(default: True)
save (bool): Save the simulation data into text files
Returns:
A txN vector of the opinions of the nodes over time
'''
N, z, max_rounds = preprocessArgs(s, max_rounds)
opinions = np.zeros((max_rounds, N))
opinions[0, :] = s
for t in trange(1, max_rounds):
z = A.dot(z)
opinions[t, :] = z
if conv_stop and \
norm(opinions[t - 1, :] - opinions[t, :], np.inf) < eps:
print('DeGroot converged after {t} rounds'.format(t=t))
break
if save:
timeStr = datetime.now().strftime("%m%d%H%M")
simid = 'dg' + timeStr
save_data(simid,
N=N,
max_rounds=max_rounds,
eps=eps,
rounds_run=t + 1,
A=A,
s=s,
opinions=opinions[0:t + 1, :])
return opinions[0:t + 1, :]
def hk(s, op_eps, max_rounds, eps=1e-6, conv_stop=True, save=False):
'''Simulates the model of Hegselmann-Krause.
This model does nto require an adjacency matrix. Connections between
nodes are calculated depending on the proximity of their opinions.
Args:
s (1xN numpy array): Initial opinions (intrinsic beliefs) vector
op_eps: ε parameter of the model
max_rounds (int): Maximum number of rounds to simulate
eps (double): Maximum difference between rounds before we assume that
the model has converged (default: 1e-6)
conv_stop (bool): Stop the simulation if the model has converged
(default: True)
save (bool): Save the simulation data into text files
Returns:
A txN vector of the opinions of the nodes over time
'''
N, z, max_rounds = preprocessArgs(s, max_rounds)
z_prev = z.copy()
opinions = np.zeros((max_rounds, N))
opinions[0, :] = s
for t in trange(1, max_rounds):
for i in range(N):
# The node chooses only those with a close enough opinion
friends_i = np.abs(z_prev - z_prev[i]) <= op_eps
z[i] = np.mean(z_prev[friends_i])
opinions[t, :] = z
z_prev = z.copy()
if conv_stop and \
norm(opinions[t - 1, :] - opinions[t, :], np.inf) < eps:
print('Hegselmann-Krause converged after {t} rounds'.format(t=t))
break
if save:
timeStr = datetime.now().strftime("%m%d%H%M")
simid = 'hk' + timeStr
save_data(simid, N=N, max_rounds=max_rounds, eps=eps,
rounds_run=t+1, s=s, op_eps=op_eps,
opinions=opinions[0:t+1, :])
return opinions[0:t+1, :]
async def abort(self,ctx):
global currentintrosessionReport
if not ctx.message.guild:
if currentintrosessionReport[str(ctx.message.author.id)]['confirmed'] == "False":
currentintrosessionReport[str(ctx.message.author.id)]['confirmed'] = 'Aborted'
await ctx.message.author.send("You have now aborted your questions")
util.save_data(currentintrosessionReport, "currentintrosessionReport")
else:
await ctx.message.author.send("Please answer the questions in the DM")
await ctx.message.delete()
def case_max_class():
class_size = [5, 10, 15, 20, 25, 30]
group = ["I3", "I1", "I2"]
arguments = [
(simulation_max_class, (N, cs, None, None), group) for cs in class_size
]
with Pool() as p:
y = p.starmap(max_sim, arguments)
save_data([class_size, y], (N, NUM_SIM))
plt.xlabel("average class size")
plt.ylabel("maximum count of infected people")
plt.scatter(class_size, y, c="red")
plt.show()
def case_max_half_edges():
percentage = np.linspace(0, 1, 21)
y = []
group = ["I3", "I1", "I2"]
arguments = [(simulation_max_time, (N, p), group) for p in percentage]
with Pool() as p:
y = p.starmap(max_sim, arguments)
save_data([percentage.tolist(), y], (N, NUM_SIM))
plt.xlabel("percentage of active edges")
plt.ylabel("maximum count of infected people")
plt.scatter(percentage, y, c="red")
plt.gcf().canvas.set_window_title(sys.argv[1])
plt.show()
def OnFolder(self, event, mode):
''' Open a file'''
dlg = wx.DirDialog(self, 'Choose a folder', '', wx.DD_DEFAULT_STYLE)
path = shared.options.get(mode)
dlg.SetPath(path if (path is not None) else '')
if dlg.ShowModal() == wx.ID_OK:
path = dlg.GetPath()
shared.options[mode] = path
self.textDict[mode].SetLabel(path)
self.Layout()
save_data(shared.options)
dlg.Destroy()
def friedkinJohnsen(A, s, max_rounds, eps=1e-6, conv_stop=True, save=False):
'''Simulates the Friedkin-Johnsen (Kleinberg) Model.
Runs a maximum of max_rounds rounds of the Friedkin-Jonsen model. If the
model converges sooner, the function returns. The stubborness matrix of
the model is extracted from the diagonal of matrix A.
Args:
A (NxN numpy array): Adjacency matrix (its diagonal is the stubborness)
s (1xN numpy array): Initial opinions (intrinsic beliefs) vector
max_rounds (int): Maximum number of rounds to simulate
eps (double): Maximum difference between rounds before we assume that
the model has converged (default: 1e-6)
conv_stop (bool): Stop the simulation if the model has converged
(default: True)
save (bool): Save the simulation data into text files
Returns:
A txN vector of the opinions of the nodes over time
'''
N, z, max_rounds = preprocessArgs(s, max_rounds)
B = np.diag(np.diag(A)) # Stubborness matrix of the model
A_model = A - B # Adjacency matrix of the model
opinions = np.zeros((max_rounds, N))
opinions[0, :] = z
for t in trange(1, max_rounds):
z = A_model.dot(z) + B.dot(s)
opinions[t, :] = z
if conv_stop and \
norm(opinions[t - 1, :] - opinions[t, :], np.inf) < eps:
print('Friedkin-Johnsen converged after {t} rounds'.format(t=t))
break
if save:
timeStr = datetime.now().strftime("%m%d%H%M")
simid = 'fj' + timeStr
save_data(simid, N=N, max_rounds=max_rounds, eps=eps,
rounds_run=t+1, A=A, s=s,
opinions=opinions[0:t+1, :])
return opinions[0:t+1, :]
def meetFriend_matrix_nomem(A, max_rounds, norm_type=2, save=False):
'''Simulates the random meeting model (matrix version).
Runs a maximum of max_rounds rounds of the "Meeting a Friend" model. If the
model converges sooner, the function returns. The stubborness matrix of
the model is extracted from the diagonal of matrix A. The function returns
the distance from the equilibrium of the Friedkin-Johnsen when the process
is complete. Uses less memory than the full model.
Args:
A (NxN numpy array): Weights matrix (its diagonal is the stubborness)
max_rounds (int): Maximum number of rounds to simulate
norm_type: The norm type used to calculate the difference from the
equilibrium
save (bool): Save the simulation data into text files
Returns:
The final distance from the equlibrium of the Friedkin-Johnsen model,
using the specified norm.
'''
max_rounds = int(max_rounds)
N = A.shape[0]
B = np.diag(np.diag(A))
nonzero_ids = [np.nonzero(A[i, :])[0] for i in xrange(A.shape[0])]
equilibrium_matrix = np.dot(inv(np.eye(N) - (A - B)), B)
R, Q = rand_matrices(A, 1, nonzero_ids)
R = R + Q
for t in trange(2, max_rounds+2):
A_t, B_t = rand_matrices(A, t, nonzero_ids)
R = A_t.dot(R) + B_t
distance = norm(R - equilibrium_matrix, ord=norm_type)
if save:
timeStr = datetime.now().strftime("%m%d%H%M")
simid = 'mf' + timeStr
save_data(simid, N=N, max_rounds=max_rounds, eps=eps,
rounds_run=max_rounds, A=A, distance=distance,
norm=norm_type)
return distance
def contact_tracing(p):
time = [5, 7.5, 10]
data = get_results(simulation_contact_tracing, (N, p, time), NUM_SIM)
states = [
["S"], ["E"], ["I1"], ["I2"], ["I3"], ["R"], ["D"], ["Q1", "Q2", "Q3"],
["QS", "QE"]
]
labels = {
"Q1": "Q_I", "QS": "Q_{S, E}",
"I1": "I_1", "I2": "I_2", "I3": "I_3"
}
times, avg, std = fill_around_std(
quolor_palette, data, 1, states=states, labels=labels
)
save_data([times, avg, std], (N, NUM_SIM, time, p))
def case_max_class_mod():
class_sizes = [5, 10, 15, 20, 25, 30]
degree = [1.25, 2.5, 3.75, 5, 6.25, 7.5]
p = 0.3
group = ["I3", "I1", "I2"]
arguments = [
(simulation_max_class, (N, cs, d, p), group)
for cs, d in zip(class_sizes, degree)
]
with Pool() as p:
y = p.starmap(max_sim, arguments)
save_data([class_sizes, y], (N, NUM_SIM))
plt.xlabel("average class size")
plt.ylabel("maximum count of infected people")
plt.scatter(class_sizes, y, c="red")
plt.show()
def deGroot(A, s, max_rounds, eps=1e-6, conv_stop=True, save=False):
'''Simulates the DeGroot Model.
Runs a maximum of max_rounds rounds of the DeGroot model. If the model
converges sooner, the function returns.
Args:
A (NxN numpy array): Adjacency matrix
s (1xN numpy array): Initial opinions vector
max_rounds (int): Maximum number of rounds to simulate
eps (double): Maximum difference between rounds before we assume that
the model has converged (default: 1e-6)
conv_stop (bool): Stop the simulation if the model has converged
(default: True)
save (bool): Save the simulation data into text files
Returns:
A txN vector of the opinions of the nodes over time
'''
N, z, max_rounds = preprocessArgs(s, max_rounds)
opinions = np.zeros((max_rounds, N))
opinions[0, :] = s
for t in trange(1, max_rounds):
z = A.dot(z)
opinions[t, :] = z
if conv_stop and \
norm(opinions[t - 1, :] - opinions[t, :], np.inf) < eps:
print('DeGroot converged after {t} rounds'.format(t=t))
break
if save:
timeStr = datetime.now().strftime("%m%d%H%M")
simid = 'dg' + timeStr
save_data(simid, N=N, max_rounds=max_rounds, eps=eps,
rounds_run=t+1, A=A, s=s,
opinions=opinions[0:t+1, :])
return opinions[0:t+1, :]
def case_average_quarantine_rates():
q1 = 0.5
output = get_results(
simulation_max_quarantine_rates, (N, q1, 0, 0), NUM_SIM
)
labels = {
"I1": "I_1", "I2": "I_2", "I3": "I_3",
"Q1": "Q", "Q2": "Q_2", "Q3": "Q_3",
}
groups = [
["Q1", "Q2", "Q3"]
]
complete_group(groups, qolor_palette)
timepoints, avg, std = fill_around_std(
qolor_palette, output, 1, labels=labels, states=groups
)
save_data([timepoints, avg, std], (N, NUM_SIM, q1))
def obfuscate_keystrokes(name, strategy, param):
"""
"""
df = load_data(name)
df = df.groupby(level=[0, 1]).apply(keystrokes2events).reset_index(level=[2, 3], drop=True)
if strategy == 'delay':
df = df.groupby(level=[0, 1]).apply(lambda x: delay_mix(x, param))
elif strategy == 'interval':
df = df.groupby(level=[0, 1]).apply(lambda x: interval_mix(x, param))
else:
raise Exception('Unknown masking strategy')
df = df.groupby(level=[0, 1]).apply(events2keystrokes).reset_index(level=[2, 3], drop=True)
save_data(df, name, masking=(strategy, param))
return
async def on_message(self,message: discord.Message):
global currentintrosessionReport
if not message.author.bot:
#print("Message is not a command")
if not message.guild:
#print("Message is a DM")
if str(message.author.id) in currentintrosessionReport:
if not currentintrosessionReport[str(message.author.id)]['confirmed'] == "True":
if 'lastAnswerTime' in currentintrosessionReport[str(message.author.id)]:
if (int(currentintrosessionReport[str(message.author.id)]['lastAnswerTime']) + 300) < int(time.time()):
currentintrosessionReport[str(message.author.id)]['confirmed'] = 'Aborted'
util.save_data(currentintrosessionReport, "currentintrosessionReport")
else:
await answer(message,message.content)
else:
await answer(message,message.content)
def refresh_login(self):
"""Deletes persisted cookies which forces a login attempt with current credentials"""
s = requests.Session()
empty_cookie_jar = s.cookies
if util.save_data(addon, self.COOKIE_FILE_NAME, empty_cookie_jar):
xbmcgui.Dialog().ok(addonname, "Cleared cookies. Please exit the addon and open it again.")
else:
xbmcgui.Dialog().ok(addonname, "Could not refresh lynda session cookies")
async def confirm(authorID,serverID):
global currentintrosessionReport
global reportChannelID
print("Doing confirm command")
if len(currentintrosessionReport[str(authorID)]['answers']) >= len(questions):
if currentintrosessionReport[str(authorID)]['confirmed'] == "False":
#print(orderID)
guild = discord.utils.get(botAcc.guilds, id=int(serverID))
member = discord.utils.get(guild.members, id=int(authorID))
if not member is None:
guild = discord.utils.get(botAcc.guilds, id=int(currentintrosessionReport[str(authorID)]['guildID']))
newIntroChannel = discord.utils.get(guild.channels, id=int(reportChannelID))
introID = len(currentintrosessionReport) +1
currentintrosessionReport[str(authorID)]['introID'] = str(introID)
currentintrosessionReport[str(authorID)]['confirmedTime'] = str(time.time())
currentintrosessionReport[str(authorID)]['confirmed'] = 'True'
currentintrosessionReport[str(authorID)]['lastAnswerTime'] = str(int(time.time()))
embed = discord.Embed(title=str(member)+" Report")
#messageToSend = ""
counter = 0
for messageAnswer in currentintrosessionReport[str(authorID)]['answers'][:-1]:
#messageToSend = messageToSend + questions[counter].split("{")[0] + ": " + str(messageAnswer)+"\n"
counter = counter +1
counter = 0
for messageAnswer in currentintrosessionReport[str(authorID)]['answers']:
if not messageAnswer == "":
embed.add_field(name=questions[counter], value=str(messageAnswer),inline=False)
counter = counter +1
newIntroMessage = await newIntroChannel.send(embed=embed)
currentintrosessionReport[str(authorID)]['newIntroMessageID'] = str(newIntroMessage.id)
await member.send("Thank you. The answers have been sent to the GFS moderators.")
util.save_data(currentintrosessionReport, "currentintrosessionReport")
def obfuscate_keystrokes(name, strategy, param):
"""
"""
df = load_data(name)
df = df.groupby(level=[0, 1]).apply(keystrokes2events).reset_index(
level=[2, 3], drop=True)
if strategy == 'delay':
df = df.groupby(level=[0, 1]).apply(lambda x: delay_mix(x, param))
elif strategy == 'interval':
df = df.groupby(level=[0, 1]).apply(lambda x: interval_mix(x, param))
else:
raise Exception('Unknown masking strategy')
df = df.groupby(level=[0, 1]).apply(events2keystrokes).reset_index(
level=[2, 3], drop=True)
save_data(df, name, masking=(strategy, param))
return
def router(self, paramstring):
"""Router function that calls other functions depending on the provided paramstrings"""
# Parse a URL-encoded paramstring to the dictionary of {<parameter>: <value>} elements
params = dict(parse_qsl(paramstring[1:]))
if params:
if params['action'] != 'play':
self.ga.track(params['action'])
# Cookiejar should definitely exist by now. Even if empty
cookiejar = util.load_data(addon, self.COOKIE_FILE_NAME)
self.api = LyndaApi(cookiejar)
if params['action'] == 'search':
self.search()
elif params['action'] == 'list_course_chapters':
self.list_course_chapters(int(params['course_id']))
elif params['action'] == 'list_chapter_videos':
self.list_chapter_videos(int(params['course_id']),
int(params['chapter_id']))
elif params['action'] == 'play':
# Log the video as being played if user is logged in
if self.api.logged_in:
self.api.log_video(int(params['video_id']))
self.play_video(int(params['course_id']),
int(params['video_id']))
elif params['action'] == 'show_access_error':
self.show_access_error()
elif params['action'] == 'list_my_courses':
self.list_my_courses()
elif params['action'] == 'refresh_login':
self.refresh_login()
else:
self.ga.track('list_root_options')
cookiejar = util.load_data(addon, self.COOKIE_FILE_NAME)
if cookiejar:
self.api = LyndaApi(cookiejar)
else:
self.api = LyndaApi()
auth_type = xbmcplugin.getSetting(__handle__, "auth_type")
# Try to log the user in if necessary
if not self.api.logged_in and auth_type != 'None':
self.login(auth_type)
# Save cookie jar to disk so other screens in addon can resume session
if not util.save_data(addon, self.COOKIE_FILE_NAME,
self.api.get_cookies()):
xbmcgui.Dialog().ok(addonname,
"Could not save lynda session cookies")
self.list_root_options()
def case_infectious_tracing():
time = [5, 7.5, 10]
p = [0.7, 1]
data = get_results(
simulation_infectious_tracing, (N, p[0], p[1], time),
NUM_SIM,
)
states = [
["S"], ["E"], ["I1"], ["I2"], ["I3"], ["R"], ["D"], ["Q1", "Q2", "Q3"],
["QS", "QE"]
]
labels = {
"Q1": "Q_{I}", "QS": "Q_{S, E}",
"I1": "I_{1}", "I2": "I_{2}", "I3": "I_{3}"
}
times, avg, std = fill_around_std(
quolor_palette, data, 1, states=states, labels=labels
)
save_data([times, avg, std], (N, NUM_SIM, time, p))
def fit_all_models(self, solver_params=None):
"""Fits parameters for all models then saves to disk.
Parameters
----------
solver_params : dict
Dict of parameters for minimizer algorithm.
"""
if not solver_params:
print("Solver params not set, using preconfigured defaults")
solver_params = {
"niter": 200,
"stepsize": 100,
"interval": 10,
"method": "TNC",
"use_jac": True,
}
cell_fits = {}
cell_lls = {}
for cell in self.cell_range:
print("Fitting cell {0}".format(cell))
cell_fits[cell] = {}
cell_lls[cell] = {}
for model in self.model_dict:
model_instance = self.model_dict[model][cell]
if model_instance.bounds is None:
raise ValueError(
"model \"{0}\" bounds not yet set".format(model))
print("Fitting {0}".format(model))
model_instance.fit_params(solver_params)
# Build dict for json dump, json requires list instead of ndarray
param_dict = {param: model_instance.fit.tolist()[index]
for index, param in enumerate(model_instance.param_names)}
cell_fits[cell][model_instance.__class__.__name__] = param_dict
cell_lls[cell][model_instance.__class__.__name__] = model_instance.fun
util.save_data({cell:cell_fits[cell]}, "cell_fits", cell=cell)
util.save_data({cell:cell_lls[cell]}, "log_likelihoods", cell=cell)
def test(model_filename, img_type, dir_name):
print("Testing {}".format(model_filename))
model = torch.load(os.path.join(dir_name, model_filename))
base_dirs = ['KKI', 'NeuroIMAGE', 'OHSU', 'Peking_1', 'Peking_2', 'Peking_3', 'Pittsburgh','WashU']
model.eval()
total_attempts = 0
total_correct = 0
res = collections.defaultdict(dict)
for base_dir in base_dirs:
d = dataset.get_data_loader(base_dir, img_type, batch_size=util.BATCH_SIZE, train=False)
dataset_correct = 0
dataset_attempt = 0
for idx, (img, label, err) in enumerate(d):
err_exists = sum(err) != 0
if err_exists:
print("Error in loading {}".format(base_dir))
continue
label = label.cuda()
img = Variable(img.float()).cuda().contiguous()
img = img.view(img.shape[0], 1, img.shape[1], img.shape[2], img.shape[3])
# print(img.shape)
with torch.no_grad():
out = model(img)
prediction = out.data.max(1)[1]
correct = float(prediction.eq(label.data).sum())
dataset_attempt += len(label)
dataset_correct += correct
if dataset_attempt != 0:
accuracy = (dataset_correct / float(dataset_attempt)) * 100.0
res[base_dir]['correct'] = dataset_correct
res[base_dir]['attempt'] = dataset_attempt
res[base_dir]['accuracy'] = accuracy
total_correct += dataset_correct
total_attempts += dataset_attempt
print('Test Accuracy {}: {}/{}={}'.format(base_dir, dataset_correct, dataset_attempt, accuracy))
res['summary']['correct'] = total_correct
res['summary']['attempt'] = total_attempts
res['summary']['accuracy'] = float(total_correct)/total_attempts * 100.0
util.save_data(res, os.path.join(dir_name, 'test_result.pckl'))
print('Total Accuracy: {}'.format(res['summary']['accuracy']))
return res
def run(self):
# Craete folder to save images in
imagesFolder = os.path.join(allImagesFolder, (str(int(time.time())))+util.rand_str())
os.makedirs(imagesFolder)
# Save images
imageIndex = 0
for i in self.images:
imageName = str(imageIndex).zfill(6) + '.jpg'
imageIndex += 1
cv2.imwrite(os.path.join(imagesFolder, imageName), np.uint8(i))
# Render into avi
inputF = os.path.join(imagesFolder, "%06d.jpg")
outputF = os.path.join(imagesFolder, "file.avi")
command = "/usr/local/bin/ffmpeg -r 5 -i {i} {o}".format(
i = inputF, o = outputF)
print(command)
os.system(command)
util.save_data(self.data, outputF)
shutil.rmtree(imagesFolder)
def accepted(self):
login = self.ui.loginEdit.text()
password = self.ui.passwordEdit.text()
dbname = self.ui.dbEdit.text()
dbase = QtSql.QSqlDatabase.addDatabase("QMYSQL")
dbase.setHostName("localhost")
dbase.setDatabaseName(dbname)
dbase.setUserName(login)
dbase.setPassword(password)
if not dbase.open():
dbase.close()
for name in QtSql.QSqlDatabase.connectionNames():
QtSql.QSqlDatabase.removeDatabase(name)
QtWidgets.QMessageBox.critical(self, "Ошибка авторизации", "Не удалось установить соединение с СУБД с заданными параметрами. Убедитесь, что СУБД запущена, а данные для входа введены верно.")
else:
LoginDialog.wnd = MainWindow(dbase)
LoginDialog.wnd.show()
password = password if self.ui.rememberPassword.isChecked() else ""
save_data(login, password, dbname)
self.accept()
def router(paramstring):
"""
Router function that calls other functions depending on the provided
paramstrings
"""
# Parse a URL-encoded paramstring to the dictionary of
# {<parameter>: <value>} elements
params = dict(parse_qsl(paramstring[1:]))
# Check the parameters passed to the plugin
if params:
s = util.load_data(addon, "lynda_session")
if s == False:
xbmcgui.Dialog().ok(addonname, "Could not load session data")
return
if params['action'] != 'play':
ga_track(params['action'])
if params['action'] == 'search':
keyboard = xbmc.Keyboard("", "Search", False)
keyboard.doModal()
if keyboard.isConfirmed() and keyboard.getText() != "":
query = keyboard.getText()
# xbmcgui.Dialog().ok(addonname, query)
courses = scrape.course_search(s, query)
list_courses(s, courses)
elif params['action'] == 'list_my_courses':
courses = scrape.get_my_courses(s)
list_courses(s, courses)
elif params['action'] == 'list_categories_letters':
list_categories_letters(s)
elif params['action'] == 'list_category_letter_contents':
list_category_letter_contents(s, params['letter'])
elif params['action'] == 'list_category_courses':
link = params['link']
courses = scrape.courses_for_category(s, link)
list_courses(s, courses)
elif params['action'] == 'list_course_chapters':
list_course_chapters(s, params['courseId'])
elif params['action'] == 'list_course_chapter_videos':
list_course_chapter_videos(s, params['courseId'], params['chapterIndex'])
elif params['action'] == 'play':
# Play a video from a provided URL.
play_video(s, params['videoId'])
else:
ga_track('list_root_options')
s = auth.initSession()
name = None
DEBUG = xbmcplugin.getSetting(__handle__, "debug")
DEBUG = DEBUG == 'true'
print("DEBUG VAR: ", DEBUG)
auth_type = xbmcplugin.getSetting(__handle__, "auth_type")
if auth_type == "Organisation":
username = xbmcplugin.getSetting(__handle__, "username")
password = xbmcplugin.getSetting(__handle__, "password")
org_url = xbmcplugin.getSetting(__handle__, "org_url")
ret = auth.org_login(s,
username=username,
password=password,
orgURL=org_url,
LDEBUG=DEBUG)
if ret != False:
name = ret
else:
xbmcgui.Dialog().ok(addonname,
"Could not login.",
"Please check your credentials are correct.")
elif auth_type == "Library":
libraryCardNum = xbmcplugin.getSetting(__handle__, "libraryCardNum")
libraryCardPin = xbmcplugin.getSetting(__handle__, "libraryCardPin")
org_url = xbmcplugin.getSetting(__handle__, "org_url")
ret = auth.library_login(s,
libCardNum=libraryCardNum,
libCardPin=libraryCardPin,
orgDomain=org_url,
LDEBUG=DEBUG)
if ret != False:
name = ret
else:
xbmcgui.Dialog().ok(addonname,
"Could not login.",
"Please check your credentials are correct.")
if not util.save_data(addon, "lynda_session", s):
xbmcgui.Dialog().ok(addonname, "Could not save requests session")
list_root_options(name)
def kNN_static(A, s, K, max_rounds, eps=1e-6, conv_stop=True, save=False):
'''Simulates the static K-Nearest Neighbors Model.
In this model, each node chooses his K-Nearest Neighbors during the
averaging of his opinion.
Args:
A (NxN numpy array): Adjacency matrix (its diagonal is the stubborness)
s (1xN numpy array): Initial opinions (intrinsic beliefs) vector
K (int): The number of the nearest neighbors to listen to
max_rounds (int): Maximum number of rounds to simulate
eps (double): Maximum difference between rounds before we assume that
the model has converged (default: 1e-6)
plot (bool): Plot preference (default: False)
conv_stop (bool): Stop the simulation if the model has converged
(default: True)
save (bool): Save the simulation data into text files
Returns:
A txN vector of the opinions of the nodes over time
'''
N, z, max_rounds = preprocessArgs(s, max_rounds)
# All nodes must listen to themselves for the averaging to work
A_model = A + np.eye(N)
# The matrix contains 0/1 values
A_model = A_model.astype(np.int8)
z_prev = z.copy()
opinions = np.zeros((max_rounds, N))
opinions[0, :] = s
for t in trange(1, max_rounds):
for i in range(N):
# Find neighbors in the underlying social network
neighbor_i = A_model[i, :] > 0
# Sort the nodes by opinion distance
sorted_dist = np.argsort(abs(z_prev - z_prev[i]))
# Change the order of the logican neighbor_i array
neighbor_i = neighbor_i[sorted_dist]
# Keep only sorted neighbors
friends_i = sorted_dist[neighbor_i]
# In case that we have less than K friends numpy
# will return the whole array (< K elements)
k_nearest = friends_i[0:K]
z[i] = np.mean(z_prev[k_nearest])
opinions[t, :] = z
z_prev = z.copy()
if conv_stop and \
norm(opinions[t - 1, :] - opinions[t, :], np.inf) < eps:
print('K-Nearest Neighbors (static) converged after {t} '
'rounds'.format(t=t))
break
if save:
timeStr = datetime.now().strftime("%m%d%H%M")
simid = 'kNNs' + timeStr
save_data(simid, N=N, max_rounds=max_rounds, eps=eps,
rounds_run=t+1, A=A, s=s, K=K,
opinions=opinions[0:t+1, :])
return opinions[0:t+1, :]
def hk_local_nomem(A, s, op_eps, max_rounds, eps=1e-6, conv_stop=True,
save=False):
'''Simulates the model of Hegselmann-Krause with an Adjacency Matrix
Contrary to the standard Hegselmann-Krause Model, here we make use of
an adjacency matrix that represents an underlying social structure
independent of the opinions held by the members of the society. This
variant does not store the intermediate opinions and as a result uses
much less memory.
Args:
A (NxN numpy array): Adjacency matrix (its diagonal is the stubborness)
s (1xN numpy array): Initial opinions (intrinsic beliefs) vector
op_eps: ε parameter of the model
max_rounds (int): Maximum number of rounds to simulate
eps (double): Maximum difference between rounds before we assume that
the model has converged (default: 1e-6)
conv_stop (bool): Stop the simulation if the model has converged
(default: True)
save (bool): Save the simulation data into text files
Returns:
t, z where t is the convergence time and z the vector of the
final opinions.
'''
N, z, max_rounds = preprocessArgs(s, max_rounds)
# All nodes must listen to themselves for the averaging to work
A_model = A + np.eye(N)
z_prev = z.copy()
for t in trange(1, max_rounds):
for i in range(N):
# Neighbors in the underlying social network
neighbor_i = A_model[i, :] > 0
opinion_close = np.abs(z_prev - z_prev[i]) <= op_eps
# The node listens to those who share a connection with him
# in the underlying network and also have an opinion
# which is close to his own
friends_i = np.logical_and(neighbor_i, opinion_close)
z[i] = np.mean(z_prev[friends_i])
if conv_stop and \
norm(z - z_prev, np.inf) < eps:
print('Hegselmann-Krause (Local Knowledge) converged after {t} '
'rounds'.format(t=t))
break
z_prev = z.copy()
if save:
timeStr = datetime.now().strftime("%m%d%H%M")
simid = 'hkloc' + timeStr
save_data(simid, N=N, max_rounds=max_rounds, eps=eps,
rounds_run=t+1, A=A, s=s, op_eps=op_eps, opinions=z)
return t, z
def OnCheckBox(self, event, attr):
shared.options[attr] = event.IsChecked()
save_data(shared.options)
def kNN_dynamic_nomem(A, s, K, max_rounds, eps=1e-6, conv_stop=True,
save=False):
'''Simulates the dynamic K-Nearest Neighbors Model. Reduced Memory.
In this model, each nodes chooses his K-Nearest Neighbors during the
averaging of his opinion. Opinions over time are not saved.
Args:
A (NxN numpy array): Adjacency matrix (its diagonal is the stubborness)
s (1xN numpy array): Initial opinions (intrinsic beliefs) vector
K (int): The number of the nearest neighbors to listen to
max_rounds (int): Maximum number of rounds to simulate
eps (double): Maximum difference between rounds before we assume that
the model has converged (default: 1e-6)
conv_stop (bool): Stop the simulation if the model has converged
(default: True)
save (bool): Save the simulation data into text files
Returns:
t, z, Q where t is the convergence time, z the vector of the
final opinions and Q the final adjacency matrix of the network
'''
N, z, max_rounds = preprocessArgs(s, max_rounds)
# All nodes must listen to themselves for the averaging to work
A_model = A + np.eye(N)
# The matrix contains 0/1 values
A_model = A_model.astype(np.int8)
z_prev = z.copy()
for t in trange(1, max_rounds):
Q = np.zeros((N, N))
# TODO: Verify that this contains the original paths of A
A_squared = A_model.dot(A_model)
for i in range(N):
# Find 2-neighbors in the underlying social network
neighbor2_i = A_squared[i, :] > 0
# Sort the nodes by opinion distance
sorted_dist = np.argsort(abs(z_prev - z_prev[i]))
# Change the order of the logican neighbor2_i array
neighbor2_i = neighbor2_i[sorted_dist]
# Keep only sorted neighbors
friends_i = sorted_dist[neighbor2_i]
# In case that we have less than K friends numpy
# will return the whole array (< K elements)
k_nearest = friends_i[0:K]
Q[i, k_nearest] = 1/k_nearest.size
z[i] = np.mean(z_prev[k_nearest])
A_model = Q.copy()
if conv_stop and \
norm(z - z_prev, np.inf) < eps:
print('K-Nearest Neighbors (dynamic) converged after {t} '
'rounds'.format(t=t))
break
z_prev = z.copy()
if save:
timeStr = datetime.now().strftime("%m%d%H%M")
simid = 'kNNd' + timeStr
save_data(simid, N=N, max_rounds=max_rounds, eps=eps,
rounds_run=t+1, A=A, s=s, K=K, opinions=z, Q=Q)
return t, z, Q
def OnTextCtrl(self, e, attr, ctrl):
self.GetParent().SetFocus()
shared.options[attr] = ctrl.GetValue()
save_data(shared.options)
def OnEvent(self, event, attr, func):
if not self.format in shared.format_dict:
shared.format_dict[self.format] = {}
shared.format_dict[self.format][attr] = func()
save_data(shared.format_dict, file=self.savefile)
def OnCheckBox(self, event, attr, ctrl, related=[]):
checked = ctrl.IsChecked()
shared.options[attr] = checked
for item in related:
item.Enable(checked)
save_data(shared.options)
def OnChoice(self, event, attr, edict=None):
shared.options[attr] = edict[event.GetString()]
save_data(shared.options)
def meetFriend_nomem(A, s, max_rounds, eps=1e-6, conv_stop=True, save=False):
'''Simulates the random meeting model.
Runs a maximum of max_rounds rounds of the "Meeting a Friend" model. If the
model converges sooner, the function returns. The stubborness matrix of
the model is extracted from the diagonal of matrix A. This function does
not save the opinions overtime and cannot generate a plot. However it uses
very little memory and is useful for determining the final opinions and
the convergence time of the model.
Args:
A (NxN numpy array): Adjacency matrix (its diagonal is the stubborness)
s (1xN numpy array): Initial opinions (intrinsic beliefs) vector
max_rounds (int): Maximum number of rounds to simulate
eps (double): Maximum difference between rounds before we assume that
the model has converged (default: 1e-6)
conv_stop (bool): Stop the simulation if the model has converged
(default: True)
save (bool): Save the simulation data into text files
Returns:
t, z where t is the convergence time and z the vector of the
final opinions.
'''
N, z, max_rounds = preprocessArgs(s, max_rounds)
nonzero_ids = [np.nonzero(A[i, :])[0] for i in xrange(A.shape[0])]
z_prev = z.copy()
if np.size(np.nonzero(A.sum(axis=1))) != N:
raise ValueError("Matrix A has one or more zero rows")
for t in trange(1, max_rounds):
# Update the opinion for each node
for i in range(N):
r_i = rchoice(A[i, :], nonzero_ids[i])
if r_i == i:
op = s[i]
else:
op = z_prev[r_i]
z[i] = (op + t*z_prev[i]) / (t+1)
if conv_stop and \
norm(z - z_prev, np.inf) < eps:
print('Meet a Friend converged after {t} rounds'.format(t=t))
break
z_prev = z.copy()
if save:
timeStr = datetime.now().strftime("%m%d%H%M")
simid = 'mf' + timeStr
save_data(simid, N=N, max_rounds=max_rounds, eps=eps,
rounds_run=t+1, A=A, s=s, opinions=z)
return t, z
def meetFriend(A, s, max_rounds, eps=1e-6, conv_stop=True, save=False):
'''Simulates the random meeting model.
Runs a maximum of max_rounds rounds of the "Meeting a Friend" model. If the
model converges sooner, the function returns. The stubborness matrix of
the model is extracted from the diagonal of matrix A.
Args:
A (NxN numpy array): Adjacency matrix (its diagonal is the stubborness)
s (1xN numpy array): Initial opinions (intrinsic beliefs) vector
max_rounds (int): Maximum number of rounds to simulate
eps (double): Maximum difference between rounds before we assume that
the model has converged (default: 1e-6)
conv_stop (bool): Stop the simulation if the model has converged
(default: True)
save (bool): Save the simulation data into text files
Returns:
A txN vector of the opinions of the nodes over time
'''
N, z, max_rounds = preprocessArgs(s, max_rounds)
nonzero_ids = [np.nonzero(A[i, :])[0] for i in xrange(A.shape[0])]
z_prev = z.copy()
opinions = np.zeros((max_rounds, N))
opinions[0, :] = s
# Cannot allow zero rows because rchoice() will fail
if np.size(np.nonzero(A.sum(axis=1))) != N:
raise ValueError("Matrix A has one or more zero rows")
for t in trange(1, max_rounds):
# Update the opinion for each node
for i in range(N):
r_i = rchoice(A[i, :], nonzero_ids[i])
if r_i == i:
op = s[i]
else:
op = z_prev[r_i]
z[i] = (op + t*z_prev[i]) / (t+1)
z_prev = z.copy()
opinions[t, :] = z
if conv_stop and \
norm(opinions[t - 1, :] - opinions[t, :], np.inf) < eps:
print('Meet a Friend converged after {t} rounds'.format(t=t))
break
if save:
timeStr = datetime.now().strftime("%m%d%H%M")
simid = 'mf' + timeStr
save_data(simid, N=N, max_rounds=max_rounds, eps=eps,
rounds_run=t+1, A=A, s=s,
opinions=opinions[0:t+1, :])
return opinions[0:t+1, :]
def reducer(a, b):
highlighted_image, detected = detect_motion(a, b)
if detected:
save_data(highlighted_image)
return b
def ga(A, B, s, max_rounds, eps=1e-6, conv_stop=True, save=False, **kwargs):
'''Simulates the Generalized Asymmetric Coevolutionary Game.
This model does nto require an adjacency matrix. Connections between
nodes are calculated depending on the proximity of their opinions.
Args:
A (NxN numpy array): Adjacency matrix (its diagonal is the stubborness)
B (NxN numpy array): The stubborness of each node
s (1xN numpy array): Initial opinions (intrinsic beliefs) vector
op_eps: ε parameter of the model
max_rounds (int): Maximum number of rounds to simulate
eps (double): Maximum difference between rounds before we assume that
the model has converged (default: 1e-6)
conv_stop (bool): Stop the simulation if the model has converged
(default: True)
save (bool): Save the simulation data into text files
**kargs: Arguments c, eps, and p for dynamic_weights function (eps and
p need to be specified only if c='pow') (default: c='linear')
Returns:
A txN vector of the opinions of the nodes over time
'''
# Check if c function was specified
if kwargs:
c = kwargs['c']
# Extra parameters for pow function
eps_c = kwargs.get('eps', 0.1)
p_c = kwargs.get('eps', 2)
else:
# Otherwise use linear as default
c = 'linear'
eps_c = None
p_c = None
N, z, max_rounds = preprocessArgs(s, max_rounds)
# The matrix contains 0/1 values
A_model = A.astype(np.int8)
opinions = np.zeros((max_rounds, N))
opinions[0, :] = s
for t in trange(1, max_rounds):
Q = dynamic_weights(A_model, s, z, c, eps_c, p_c) + B
Q = row_stochastic(Q)
B_temp = np.diag(np.diag(Q))
Q = Q - B_temp
z = Q.dot(z) + B_temp.dot(s)
opinions[t, :] = z
if conv_stop and \
norm(opinions[t - 1, :] - opinions[t, :], np.inf) < eps:
print('G-A converged after {t} rounds'.format(t=t))
break
if save:
timeStr = datetime.now().strftime("%m%d%H%M")
simid = 'ga' + timeStr
save_data(simid, N=N, max_rounds=max_rounds, eps=eps,
rounds_run=t+1, A=A, s=s, B=B, c=c, eps_c=eps_c,
p_c=p_c, opinions=opinions[0:t+1, :])
return opinions[0:t+1, :]
def hk_rand(s, K, op_eps, max_rounds, eps=1e-6, conv_stop=True, save=False):
'''Simulate the model of Hegselmann-Krause with random sampling.
In each round every node chooses K other nodes uniformly at random and
updates his opinion to be the average of the opinions of those K nodes
that have a opinion distance at most equal to op_eps.
Args:
s (1xN numpy array): Initial opinions (intrinsic beliefs) vector
K (int): The number of nodes which will be randomly chosen in each
round.
op_eps: ε parameter of the model
max_rounds (int): Maximum number of rounds to simulate
eps (double): Maximum difference between rounds before we assume that
the model has converged (default: 1e-6)
conv_stop (bool): Stop the simulation if the model has converged
(default: True)
save (bool): Save the simulation data into text files
Returns:
A txN vector of the opinions of the nodes over time
'''
N, z, max_rounds = preprocessArgs(s, max_rounds)
z_prev = z.copy()
opinions = np.zeros((max_rounds, N))
opinions[0, :] = s
for t in trange(1, max_rounds):
for i in range(N):
# Choose K random nodes as temporary "neighbors"
rand_sample = np.array(stdrand.sample(xrange(N), K))
neighbors_i = np.zeros(N, dtype=bool)
neighbors_i[rand_sample] = 1
# Always choose yourself
neighbors_i[i] = 1
# The node chooses only those with a close enough opinion
friends_i = np.abs(z_prev - z_prev[i]) <= op_eps
friends_i = np.logical_and(neighbors_i, friends_i)
z[i] = np.mean(z_prev[friends_i])
opinions[t, :] = z
z_prev = z.copy()
if conv_stop and \
norm(opinions[t - 1, :] - opinions[t, :], np.inf) < eps:
print('Hegselmann-Krause (random) converged after {t}'
' rounds'.format(t=t))
break
if save:
timeStr = datetime.now().strftime("%m%d%H%M")
simid = 'hk' + timeStr
save_data(simid, N=N, max_rounds=max_rounds, eps=eps,
rounds_run=t+1, s=s, op_eps=op_eps,
opinions=opinions[0:t+1, :])
return opinions[0:t+1, :]