def plot_distance_distribution(graph):
count = {}
for n in list(graph):
tmp_dict = nx.shortest_path_length(graph, source=n)
tmp_res = 0
for j in list(graph):
if n != j:
try:
tmp_res += tmp_dict[j]
except KeyError:
continue
count[n] = tmp_res / (len(list(graph)) - 1)
counts = {}
for n in count:
if n not in counts:
counts[n] = 0
counts[n] += 1
x, y = log_binning(count, 150)
trace = go.Scatter(x=np.array(y),
y=np.array(x),
mode='markers',
line=dict(color=('rgb(205, 12, 24)')))
layout = dict(title="Distance Distribution",
xaxis=dict(title='Average distance', type='log'),
yaxis=dict(title='Count', type='log'))
plot_data = [trace]
fig = go.Figure(data=plot_data, layout=layout)
image.save_as(fig, filename=IMG_PATH + 'distance_distribution.jpeg')
def sensitivity_analysis(si, names, year, num=0):
trace1 = go.Bar(x=names[num:],
y=si["S1"][num:],
name='S1',
error_y=dict(type='data',
array=si['S1_conf'][num:],
visible=True))
trace2 = go.Bar(x=names[num:],
y=si['ST'][num:],
name='Total',
error_y=dict(type='data',
array=si['ST_conf'][num:],
visible=True))
data = [trace1, trace2]
layout = go.Layout(barmode='group',
title=''.join(['Sensitivity analysis for ',
str(year)]))
if num:
title = '_groups_coeffs_'
else:
title = '_full_coeffs_'
fig = go.Figure(data=data, layout=layout)
image.save_as(fig,
filename=''.join([path_to_figure, title,
str(year), '.jpeg']))
def plot_kernel_density(sample, y, name):
x = np.array(sample)
y = np.array(y)
trace = go.Scatter(x=x, y=y)
data = [trace]
fig = go.Figure(data=data)
image.save_as(fig, filename=path_to_figure + name)
def plot_speedup(in_files):
data = _parse_in(in_files)
data['total_time'] = pd.to_datetime(data['total_time'])
seconds = 1000000 * data['total_time'].dt.second
minutes = 60000000 * data['total_time'].dt.minute
data['time'] = data['total_time'].dt.microsecond + seconds + minutes
ls_names = list(data['algorithm_name'].unique())
print(ls_names)
for i in ls_names:
seconds = []
arr_tmp = data.query('algorithm_name == @i').copy()
arr_tmp = arr_tmp.groupby('cpu_count')['time'].mean()
print(arr_tmp)
cpu = list(arr_tmp.index.values)
for j in cpu:
seconds += [arr_tmp[j]]
print(cpu, seconds)
seconds_sequential = seconds[0]
for k in range(len(seconds)):
seconds[k] = seconds_sequential / seconds[k]
print(cpu, seconds)
trace = go.Scatter(x=np.array(cpu),
y=np.array(seconds),
mode='lines+markers',
line=dict(color=('rgb(205, 12, 24)')))
layout = dict(title=i,
xaxis=dict(title='Cpu count'),
yaxis=dict(title='Speedup'))
plot_data = [trace]
fig = go.Figure(data=plot_data, layout=layout)
image.save_as(fig, filename=path_to_figure + i + '.jpeg')
def plot_avg_binned_degree_connectivity(connect):
items = sorted(connect.items())
x, y = log_binning(connect, 60)
trace = go.Scatter(x=np.array(x),
y=np.array(y),
mode='markers',
line=dict(color=('rgb(205, 12, 24)')))
layout = dict(title="Assortativity in",
xaxis=dict(title='k'),
yaxis=dict(title='$<k_{nn}>$'))
plot_data = [trace]
fig = go.Figure(data=plot_data, layout=layout)
image.save_as(fig, filename=IMG_PATH + 'assortativity_in_binned.jpeg')
def plot_res(temp, magnetization):
trace = go.Scatter(x=temp,
y=[abs(mag) for mag in magnetization],
mode='lines+markers')
layout = dict(
title='Magnetization dynamics with different fractions of conformists',
xaxis=dict(title='P_c'),
yaxis=dict(title='M'),
)
data = [trace]
fig = go.Figure(data=data, layout=layout)
plot(fig, image_filename="test.png")
image.save_as(fig, filename='conf_magnetization' + '.jpeg')
def plot_distributions(train,
test,
title='Rating distribution in Amazon dataset'):
x = ['1', '2', '3', '4', '5']
trace1 = go.Bar(x=x, y=train, name='Train')
trace2 = go.Bar(x=x, y=test, name='Test')
data = [trace1, trace2]
layout = go.Layout(title=title,
barmode='group',
yaxis=dict(title='# of reviews'),
xaxis=dict(title='Rating'))
fig = go.Figure(data=data, layout=layout)
image.save_as(fig,
filename=''.join([FIG_DIR, 'rating_distribution', '.jpeg']))
def coeff_visualization(data, years):
year_categories = [i for i in range(len(data[0]))]
trace = []
for i in range(len(data)):
trace += [
go.Scatter(x=year_categories,
y=data[i],
name=years[i],
mode='lines+markers')
]
layout = dict(
title='Checking validity of coeffs',
xaxis=dict(title='Transition to age group'),
yaxis=dict(title='Value'),
)
fig = go.Figure(data=trace, layout=layout)
image.save_as(fig, filename=path_to_figure + 'check_coeffs.jpeg')
def profile_compare_years(data, years, type, title):
traces = []
for year in years:
y = [
y * 1000
for y in data[data['date'] == year][group].values.tolist()[0]
]
traces += [go.Scatter(x=group, y=y, mode='lines+markers', name=year)]
years = [str(year) for year in years]
layout = dict(
title='Demographic profile prediction for Russia in ' +
', '.join(years) + ' (' + type + ')',
xaxis=dict(title='Age group'),
yaxis=dict(title='Amount of people'),
)
fig = go.Figure(data=traces, layout=layout)
image.save_as(fig, filename=path_to_figure + title + '.jpeg')
def show_profile(data, year, type, title):
y = [
y * 1000 for y in data[data['date'] == year][group].values.tolist()[0]
]
trace = go.Scatter(x=group,
y=y,
mode='lines+markers',
name='title',
line=dict(color=('rgb(205, 12, 24)')))
layout = dict(
title='Demographic profile prediction for Russia in ' + str(year) +
' (' + type + ')',
xaxis=dict(title='Age group'),
yaxis=dict(title='Amount of people'),
)
data = [trace]
fig = go.Figure(data=data, layout=layout)
image.save_as(fig, filename=path_to_figure + title + '.jpeg')
def profiles_m_f(fem_data, male_data, year, title):
y = [
y * 1000
for y in fem_data[fem_data['date'] == year][group].values.tolist()[0]
]
trace1 = go.Scatter(x=group, y=y, mode='lines+markers', name='female')
y = [
y * 1000
for y in male_data[male_data['date'] == year][group].values.tolist()[0]
]
trace2 = go.Scatter(x=group, y=y, mode='lines+markers', name='male')
layout = dict(
title='Demographic profile prediction for Russia in ' + str(year),
xaxis=dict(title='Age group'),
yaxis=dict(title='Amount of people'),
)
data = [trace1, trace2]
fig = go.Figure(data=data, layout=layout)
image.save_as(fig, filename=path_to_figure + title + '.jpeg')
def plot_binned_clustering_coeffs(data):
values = data['clustering']
values = [float(i) for i in values]
counts = {}
for n in values:
if n not in counts:
counts[n] = 0
counts[n] += 1
x, y = log_binning(counts, 60)
trace = go.Scatter(x=np.array(x),
y=np.array(y),
mode='markers',
line=dict(color=('rgb(205, 12, 24)')))
layout = dict(title="Clustering Distribution",
xaxis=dict(title='Clustering Coefficient', type='log'),
yaxis=dict(title='Count', type='log'))
plot_data = [trace]
fig = go.Figure(data=plot_data, layout=layout)
image.save_as(fig, filename=IMG_PATH + 'clustering_binned.jpeg')
def plot_binned_betweeness_centrality(data):
values = data['betweenesscentrality']
values = [float(i) for i in values]
counts = {}
for n in values:
if n not in counts:
counts[n] = 0
counts[n] += 1
x, y = log_binning(counts, 60)
trace = go.Scatter(x=np.array(x),
y=np.array(y),
mode='markers',
line=dict(color=('rgb(205, 12, 24)')))
layout = dict(title="Beetweeness Distribution",
xaxis=dict(title='Betweenness centrality', type='log'),
yaxis=dict(title='Count', type='log'))
plot_data = [trace]
fig = go.Figure(data=plot_data, layout=layout)
image.save_as(fig, filename=IMG_PATH + 'betweenesscentrality_binned2.jpeg')
def uncertainty_plot(min_population, max_population, average_population,
simulated_years):
trace1 = go.Scatter(y=min_population,
x=simulated_years,
mode='lines',
line=dict(color=('rgb(205, 12, 24)')))
trace2 = go.Scatter(y=max_population,
x=simulated_years,
mode='lines',
line=dict(color=('rgb(205, 12, 24)')))
trace3 = go.Scatter(y=average_population, x=simulated_years, mode='lines')
layout = go.Layout(
title='Uncertainty analysis',
xaxis=dict(title='Year'),
yaxis=dict(title='Population'),
)
data = [trace1, trace2, trace3]
fig = go.Figure(data=data, layout=layout)
image.save_as(fig,
filename=''.join([path_to_figure, 'uncertanty', '.jpeg']))
def compare_profiles(obtained, given, year, title):
if year not in list_of_years:
print('No given data for this year')
return
y = [
y * 1000
for y in obtained[obtained['date'] == year][group].values.tolist()[0]
]
trace1 = go.Scatter(x=group, y=y, mode='lines+markers', name='obtained')
y = [
y * 1000
for y in given[given['date'] == year][group].values.tolist()[0]
]
trace2 = go.Scatter(x=group, y=y, mode='lines+markers', name='given')
layout = dict(
title='Comparison of results for ' + str(year),
xaxis=dict(title='Age group'),
yaxis=dict(title='Amount of people'),
)
data = [trace1, trace2]
fig = go.Figure(data=data, layout=layout)
plot(fig, image_filename="test.png")
image.save_as(fig, filename=path_to_figure + title + '.jpeg')
def save_fig(figure, filename, out_directory='', dynamic=False, scale=None):
"""
Renders and saves plotly figures
Parameters
----------
figure : plotly figure object
plotly figure to render
filename : str
Name to save figure as
out_directory : str
Location to store the figure
dynamic : bool
Return dynamic (True) or static (False)
scale : int
Increase the resolution of saved image by 'scale'
Returns
-------
Dynamic or static image based on user request
"""
path = "{0}{1}".format(out_directory, filename)
if not exists("{}.png".format(path)):
name = "{}.png".format(path)
image.save_as(figure, name, scale=scale)
else:
name = "{}_{}.png".format(path, int(time()))
image.save_as(figure, name, scale=scale)
if dynamic == True:
return iplot(figure, filename=filename)
elif dynamic == False:
return Image("{}".format(name))
else:
print("Invalid entry")
),
hovermode='closest',
#plot_bgcolor='#EFECEA', #set background color
)
trace2= scatter_nodes([x,y],n,colors_graph)
data=Data([trace2])
fig = Figure(data=data, layout=layout)
#fig['layout'].update(annotations=make_annotations([x,y], n, colors_graph))
#offline.iplot(fig, filename='tst')
image.save_as(fig,args.emb_file+"-grafico.png",scale=3)
a = input('Pressione uma tecla para continuar: ')
if(a and int(a) == 0):
break
'''
g.vp.vertex_name[v]
g.vertex_index[v]
g.vertex(index)
'''
