def create_plot(feature):
# ===========================
# BOX-SCATTER PLOT
# ===========================
if feature == "All Countries Box-Scatter Plot":
fig = go.Figure()
for column in df.columns[1:-1].to_list():
fig.add_trace(
go.Box(y=df[column],
name=column,
boxpoints='all',
hovertext=df["country"]))
fig.update_layout(
title="All Countries Box-Scatter Plot",
updatemenus=[
go.layout.Updatemenu(
active=0,
buttons=list([
dict(
label='cureduexpen_pri',
method='update',
args=[
{
'visible': [
True, False, False, False, False,
False, False, False, False, False,
False, False, False, False, False,
False, False, False, False, False,
False, False, False
]
},
# the index of True aligns with the indices of plot traces
{
'title': 'Primary Education Expenditure',
'showlegend': True
}
]),
dict(label='cureduexpen_sec',
method='update',
args=[{
'visible': [
False, True, False, False, False, False,
False, False, False, False, False, False,
False, False, False, False, False, False,
False, False, False, False, False
]
}, {
'title': 'Secondary Education Expenditure',
'showlegend': True
}]),
dict(
label='cureduexpen_ter',
method='update',
args=[
{
'visible': [
False, False, True, False, False,
False, False, False, False, False,
False, False, False, False, False,
False, False, False, False, False,
False, False, False
]
},
# the index of True aligns with the indices of plot traces
{
'title': 'Tertiary Education Expenditure',
'showlegend': True
}
]),
dict(label='cureduexpen_total',
method='update',
args=[{
'visible': [
False, False, False, True, False, False,
False, False, False, False, False, False,
False, False, False, False, False, False,
False, False, False, False, False
]
}, {
'title': 'Total Education Expenditure',
'showlegend': True
}]),
dict(
label='eduattain_doctoral',
method='update',
args=[
{
'visible': [
False, False, False, False, True,
False, False, False, False, False,
False, False, False, False, False,
False, False, False, False, False,
False, False, False
]
},
# the index of True aligns with the indices of plot traces
{
'title': 'Doctoral Education Attainment',
'showlegend': True
}
]),
dict(label='eduattain_bachelor',
method='update',
args=[{
'visible': [
False, False, False, False, False, True,
False, False, False, False, False, False,
False, False, False, False, False, False,
False, False, False, False, False
]
}, {
'title': "Bachelor's Education Attainment",
'showlegend': True
}]),
dict(
label='eduattain_master',
method='update',
args=[
{
'visible': [
False, False, False, False, False,
False, True, False, False, False,
False, False, False, False, False,
False, False, False, False, False,
False, False, False
]
},
# the index of True aligns with the indices of plot traces
{
'title': "Master's Education Attainment",
'showlegend': True
}
]),
dict(label='eduattain_sec',
method='update',
args=[{
'visible': [
False, False, False, False, False, False,
False, True, False, False, False, False,
False, False, False, False, False, False,
False, False, False, False, False
]
}, {
'title': 'Secondary Education Attainment',
'showlegend': True
}]),
dict(
label='eduattain_postsec',
method='update',
args=[
{
'visible': [
False, False, False, False, False,
False, False, False, True, False,
False, False, False, False, False,
False, False, False, False, False,
False, False, False
]
},
# the index of True aligns with the indices of plot traces
{
'title':
'Post-Secondary Education Attainment',
'showlegend': True
}
]),
dict(label='eduattain_primary',
method='update',
args=[{
'visible': [
False, False, False, False, False, False,
False, False, False, True, False, False,
False, False, False, False, False, False,
False, False, False, False, False
]
}, {
'title': 'Primary Education Attainment',
'showlegend': True
}]),
dict(
label='eduattain_tertiary',
method='update',
args=[
{
'visible': [
False, False, False, False, False,
False, False, False, False, False,
True, False, False, False, False,
False, False, False, False, False,
False, False, False
]
},
# the index of True aligns with the indices of plot traces
{
'title': 'Tertiary Education Attainment',
'showlegend': True
}
]),
dict(label='eduattain_uppersec',
method='update',
args=[{
'visible': [
False, False, False, False, False, False,
False, False, False, False, False, True,
False, False, False, False, False, False,
False, False, False, False, False
]
}, {
'title':
'Upper Secondary Education Attainment',
'showlegend': True
}]),
dict(
label='expense',
method='update',
args=[
{
'visible': [
False, False, False, False, False,
False, False, False, False, False,
False, False, True, False, False,
False, False, False, False, False,
False, False, False
]
},
# the index of True aligns with the indices of plot traces
{
'title': 'Total Expenses (% GDP)',
'showlegend': True
}
]),
dict(label='life_expectatbirth_fem',
method='update',
args=[{
'visible': [
False, False, False, False, False, False,
False, False, False, False, False, False,
False, True, False, False, False, False,
False, False, False, False, False
]
}, {
'title': 'Life Expectancy, Female (Years)',
'showlegend': True
}]),
dict(
label='life_expectatbirth_male',
method='update',
args=[
{
'visible': [
False, False, False, False, False,
False, False, False, False, False,
False, False, False, False, True,
False, False, False, False, False,
False, False, False
]
},
# the index of True aligns with the indices of plot traces
{
'title': 'Life Expectancy Male (Years)',
'showlegend': True
}
]),
dict(label='life_expectatbirth_total',
method='update',
args=[{
'visible': [
False, False, False, False, False, False,
False, False, False, False, False, False,
False, False, False, True, False, False,
False, False, False, False, False
]
}, {
'title': 'Life Expectancy Total (Years)',
'showlegend': True
}]),
dict(
label='milexp_gdp',
method='update',
args=[
{
'visible': [
False, False, False, False, False,
False, False, False, False, False,
False, False, False, False, False,
False, True, False, False, False,
False, False, False
]
},
# the index of True aligns with the indices of plot traces
{
'title': 'Military Expenditure (% GDP)',
'showlegend': True
}
]),
dict(label='milexp_usd',
method='update',
args=[{
'visible': [
False, False, False, False, False, False,
False, False, False, False, False, False,
False, False, False, False, False, True,
False, False, False, False, False
]
}, {
'title': 'Military Expenditure (USD)',
'showlegend': True
}]),
dict(
label='totalreserves',
method='update',
args=[
{
'visible': [
False, False, False, False, False,
False, False, False, False, False,
False, False, False, False, False,
False, False, False, True, False,
False, False, False
]
},
# the index of True aligns with the indices of plot traces
{
'title':
'Total Reserves (% external debt)',
'showlegend': True
}
]),
dict(label='unemptotal_modeiloest',
method='update',
args=[{
'visible': [
False, False, False, False, False, False,
False, False, False, False, False, False,
False, False, False, False, False, False,
False, True, False, False, False
]
}, {
'title':
'Unemployment ILO (% Total Labor Force)',
'showlegend': True
}]),
dict(
label='unemptotal_nationalest',
method='update',
args=[
{
'visible': [
False, False, False, False, False,
False, False, False, False, False,
False, False, False, False, False,
False, False, False, False, False,
True, False, False
]
},
# the index of True aligns with the indices of plot traces
{
'title':
'Unemployment National (% Total Labor Force)',
'showlegend': True
}
]),
dict(label='happiness_score',
method='update',
args=[{
'visible': [
False, False, False, False, False, False,
False, False, False, False, False, False,
False, False, False, False, False, False,
False, False, False, True, False
]
}, {
'title': 'Happiness Score (1-10)',
'showlegend': True
}]),
dict(
label='social_support',
method='update',
args=[
{
'visible': [
False, False, False, False, False,
False, False, False, False, False,
False, False, False, False, False,
False, False, False, False, False,
False, False, True
]
},
# the index of True aligns with the indices of plot traces
{
'title': 'Social Support (1-10)',
'showlegend': True
}
]),
]))
])
# ===========================
# BAR PLOT
# ===========================
else:
# scale all columns from 0-1
scaler = MinMaxScaler()
df2 = df.copy()
df2.loc[:, df2.columns != 'country'] = scaler.fit_transform(
df2.loc[:, df2.columns != 'country'])
df2.drop('pay_usd', axis=1, inplace=True)
# PLOT DATA - BAR
fig = px.bar(df2,
x="country",
y=list(df2.columns[1:]),
title="All Countries Stacked Bar Plot by Attribute")
fig.update_layout(barmode='stack',
xaxis={
'categoryorder': 'array',
'categoryarray': by_continent
})
fig.update_xaxes(title={'text': None})
# ===========================
# RESIZE PLOT
# ===========================
fig.update_layout(autosize=False, width=1200, height=500)
return plotly.io.to_json(fig)
def get_box(cond):
return go.Box(y=[
cond['p05'], cond['p25'], cond['p50'], cond['p75'], cond['p95']
],
x=[cond['experiment']] * 5,
name=cond['config'])
7. Min and Max values are shown with "whiskers". 8. The main use of box plot is to perform a real analysis """ # Importing the libraries import plotly.offline as pyo import plotly.graph_objs as go # set up an array of 20 data points, with 20 as the median value y = [1,14,14,15,16,18,18,19,19,20,20,23,24,26,27,27,28,29,33,54] # BAsic box plot data = [go.Box(y=y)] pyo.plot(data,filename="basic_box_plot.html") # Changes applied """ In this exammple a additional parameter is 'boxpoints' which is use to show all points """ data_1 = [go.Box(y=y,boxpoints='all')] pyo.plot(data_1,filename="box_plot_1.html") """
def isf_compare_year():
#get data for each year
(all_2014, bucket_list_2014) = isf.get_isf_for_years('2014', '2014')
(all_2015, bucket_list_2015) = isf.get_isf_for_years('2015', '2015')
(all_2016, bucket_list_2016) = isf.get_isf_for_years('2016', '2016')
(all_2017, bucket_list_2017) = isf.get_isf_for_years('2017', '2017')
(all_2018, bucket_list_2018) = isf.get_isf_for_years('2018', '2018')
#get non-bucketed data for each year
allData_2014 = [data[0] for data in all_2014 if data[0]]
allData_2015 = [data[0] for data in all_2015 if data[0]]
allData_2016 = [data[0] for data in all_2016 if data[0]]
allData_2017 = [data[0] for data in all_2017 if data[0]]
allData_2018 = [data[0] for data in all_2018 if data[0]]
#layout dict
yaxis_dict = dict(title='mgdl/unit',
zeroline=True,
zerolinecolor='#800000',
showline=False,
rangemode='tozero')
#create plot for non-bucketed data
allWhisker_2014 = go.Box(y=allData_2014, name='2014')
allWhisker_2015 = go.Box(y=allData_2015, name='2015')
allWhisker_2016 = go.Box(y=allData_2016, name='2016')
allWhisker_2017 = go.Box(y=allData_2017, name='2017')
allWhisker_2018 = go.Box(y=allData_2018, name='2018')
layout_allWhisker = go.Layout(title=('All ISF values for each yaer'),
width=1000,
height=800,
yaxis=yaxis_dict)
graph_all = go.Figure(data=[
allWhisker_2014, allWhisker_2015, allWhisker_2016, allWhisker_2017,
allWhisker_2018
],
layout=layout_allWhisker)
graphJSON_all = json.dumps(graph_all, cls=plotly.utils.PlotlyJSONEncoder)
#create plot for 0-2am time bucket
plot14_0 = go.Box(y=bucket_list_2014[0], name='2014')
plot15_0 = go.Box(y=bucket_list_2015[0], name='2015')
plot16_0 = go.Box(y=bucket_list_2016[0], name='2016')
plot17_0 = go.Box(y=bucket_list_2017[0], name='2017')
plot18_0 = go.Box(y=bucket_list_2018[0], name='2018')
layout_0am = go.Layout(title=('ISF values for 0am-2am time bucket'),
width=1000,
height=800,
yaxis=yaxis_dict)
graph_0am = go.Figure(
data=[plot14_0, plot15_0, plot16_0, plot17_0, plot18_0],
layout=layout_0am)
graphJSON_0am = json.dumps(graph_0am, cls=plotly.utils.PlotlyJSONEncoder)
#create plot for 2-4am time bucket
plot14_1 = go.Box(y=bucket_list_2014[1], name='2014')
plot15_1 = go.Box(y=bucket_list_2015[1], name='2015')
plot16_1 = go.Box(y=bucket_list_2016[1], name='2016')
plot17_1 = go.Box(y=bucket_list_2017[1], name='2017')
plot18_1 = go.Box(y=bucket_list_2018[1], name='2018')
layout_2am = go.Layout(title=('ISF values for 2am-4am time bucket'),
width=1000,
height=800,
yaxis=yaxis_dict)
graph_2am = go.Figure(
data=[plot14_1, plot15_1, plot16_1, plot17_1, plot18_1],
layout=layout_2am)
graphJSON_2am = json.dumps(graph_2am, cls=plotly.utils.PlotlyJSONEncoder)
#create plot for 4-6am time bucket
plot14_2 = go.Box(y=bucket_list_2014[2], name='2014')
plot15_2 = go.Box(y=bucket_list_2015[2], name='2015')
plot16_2 = go.Box(y=bucket_list_2016[2], name='2016')
plot17_2 = go.Box(y=bucket_list_2017[2], name='2017')
plot18_2 = go.Box(y=bucket_list_2018[2], name='2018')
layout_4am = go.Layout(title=('ISF values for 4am-6am time bucket'),
width=1000,
height=800,
yaxis=yaxis_dict)
graph_4am = go.Figure(
data=[plot14_2, plot15_2, plot16_2, plot17_2, plot18_2],
layout=layout_4am)
graphJSON_4am = json.dumps(graph_4am, cls=plotly.utils.PlotlyJSONEncoder)
#create plot for 6am-8am time bucket
plot14_3 = go.Box(y=bucket_list_2014[3], name='2014')
plot15_3 = go.Box(y=bucket_list_2015[3], name='2015')
plot16_3 = go.Box(y=bucket_list_2016[3], name='2016')
plot17_3 = go.Box(y=bucket_list_2017[3], name='2017')
plot18_3 = go.Box(y=bucket_list_2018[3], name='2018')
layout_6am = go.Layout(title=('ISF values for 6am-8am time bucket'),
width=1000,
height=800,
yaxis=yaxis_dict)
graph_6am = go.Figure(
data=[plot14_3, plot15_3, plot16_3, plot17_3, plot18_3],
layout=layout_6am)
graphJSON_6am = json.dumps(graph_6am, cls=plotly.utils.PlotlyJSONEncoder)
#create plot for 8am-10am time bucket
plot14_4 = go.Box(y=bucket_list_2014[4], name='2014')
plot15_4 = go.Box(y=bucket_list_2015[4], name='2015')
plot16_4 = go.Box(y=bucket_list_2016[4], name='2016')
plot17_4 = go.Box(y=bucket_list_2017[4], name='2017')
plot18_4 = go.Box(y=bucket_list_2018[4], name='2018')
layout_8am = go.Layout(title=('ISF values for 8am-10am time bucket'),
width=1000,
height=800,
yaxis=yaxis_dict)
graph_8am = go.Figure(
data=[plot14_4, plot15_4, plot16_4, plot17_4, plot18_4],
layout=layout_8am)
graphJSON_8am = json.dumps(graph_8am, cls=plotly.utils.PlotlyJSONEncoder)
#create plot for 10am-12pm time bucket
plot14_5 = go.Box(y=bucket_list_2014[5], name='2014')
plot15_5 = go.Box(y=bucket_list_2015[5], name='2015')
plot16_5 = go.Box(y=bucket_list_2016[5], name='2016')
plot17_5 = go.Box(y=bucket_list_2017[5], name='2017')
plot18_5 = go.Box(y=bucket_list_2018[5], name='2018')
layout_10am = go.Layout(title=('ISF values for 10am-12pm time bucket'),
width=1000,
height=800,
yaxis=yaxis_dict)
graph_10am = go.Figure(
data=[plot14_5, plot15_5, plot16_5, plot17_5, plot18_5],
layout=layout_10am)
graphJSON_10am = json.dumps(graph_10am, cls=plotly.utils.PlotlyJSONEncoder)
#create plot for 12pm-2pm time bucket
plot14_6 = go.Box(y=bucket_list_2014[6], name='2014')
plot15_6 = go.Box(y=bucket_list_2015[6], name='2015')
plot16_6 = go.Box(y=bucket_list_2016[6], name='2016')
plot17_6 = go.Box(y=bucket_list_2017[6], name='2017')
plot18_6 = go.Box(y=bucket_list_2018[6], name='2018')
layout_12pm = go.Layout(title=('ISF values for 12pm-2pm time bucket'),
width=1000,
height=800,
yaxis=yaxis_dict)
graph_12pm = go.Figure(
data=[plot14_6, plot15_6, plot16_6, plot17_6, plot18_6],
layout=layout_12pm)
graphJSON_12pm = json.dumps(graph_12pm, cls=plotly.utils.PlotlyJSONEncoder)
#create plot for 2pm-4pm time bucket
plot14_7 = go.Box(y=bucket_list_2014[7], name='2014')
plot15_7 = go.Box(y=bucket_list_2015[7], name='2015')
plot16_7 = go.Box(y=bucket_list_2016[7], name='2016')
plot17_7 = go.Box(y=bucket_list_2017[7], name='2017')
plot18_7 = go.Box(y=bucket_list_2018[7], name='2018')
layout_2pm = go.Layout(title=('ISF values for 2pm-4pm time bucket'),
width=1000,
height=800,
yaxis=yaxis_dict)
graph_2pm = go.Figure(
data=[plot14_7, plot15_7, plot16_7, plot17_7, plot18_7],
layout=layout_2pm)
graphJSON_2pm = json.dumps(graph_2pm, cls=plotly.utils.PlotlyJSONEncoder)
#create plot for 4pm-6pm time bucket
plot14_8 = go.Box(y=bucket_list_2014[8], name='2014')
plot15_8 = go.Box(y=bucket_list_2015[8], name='2015')
plot16_8 = go.Box(y=bucket_list_2016[8], name='2016')
plot17_8 = go.Box(y=bucket_list_2017[8], name='2017')
plot18_8 = go.Box(y=bucket_list_2018[8], name='2018')
layout_4pm = go.Layout(title=('ISF values for 4pm-6pm time bucket'),
width=1000,
height=800,
yaxis=yaxis_dict)
graph_4pm = go.Figure(
data=[plot14_8, plot15_8, plot16_8, plot17_8, plot18_8],
layout=layout_4pm)
graphJSON_4pm = json.dumps(graph_4pm, cls=plotly.utils.PlotlyJSONEncoder)
#create plot for 6pm-8pm time bucket
plot14_9 = go.Box(y=bucket_list_2014[9], name='2014')
plot15_9 = go.Box(y=bucket_list_2015[9], name='2015')
plot16_9 = go.Box(y=bucket_list_2016[9], name='2016')
plot17_9 = go.Box(y=bucket_list_2017[9], name='2017')
plot18_9 = go.Box(y=bucket_list_2018[9], name='2018')
layout_6pm = go.Layout(title=('ISF values for 6pm-8pm time bucket'),
width=1000,
height=800,
yaxis=yaxis_dict)
graph_6pm = go.Figure(
data=[plot14_9, plot15_9, plot16_9, plot17_9, plot18_9],
layout=layout_6pm)
graphJSON_6pm = json.dumps(graph_6pm, cls=plotly.utils.PlotlyJSONEncoder)
#create plot for 8pm-10pm time bucket
plot14_10 = go.Box(y=bucket_list_2014[10], name='2014')
plot15_10 = go.Box(y=bucket_list_2015[10], name='2015')
plot16_10 = go.Box(y=bucket_list_2016[10], name='2016')
plot17_10 = go.Box(y=bucket_list_2017[10], name='2017')
plot18_10 = go.Box(y=bucket_list_2018[10], name='2018')
layout_8pm = go.Layout(title=('ISF values for 8pm-10pm time bucket'),
width=1000,
height=800,
yaxis=yaxis_dict)
graph_8pm = go.Figure(
data=[plot14_10, plot15_10, plot16_10, plot17_10, plot18_10],
layout=layout_8pm)
graphJSON_8pm = json.dumps(graph_8pm, cls=plotly.utils.PlotlyJSONEncoder)
#create plot for 10pm-12am time bucket
plot14_11 = go.Box(y=bucket_list_2014[11], name='2014')
plot15_11 = go.Box(y=bucket_list_2015[11], name='2015')
plot16_11 = go.Box(y=bucket_list_2016[11], name='2016')
plot17_11 = go.Box(y=bucket_list_2017[11], name='2017')
plot18_11 = go.Box(y=bucket_list_2018[11], name='2018')
layout_10pm = go.Layout(title=('ISF values for 10pm-12am time bucket'),
width=1000,
height=800,
yaxis=yaxis_dict)
graph_10pm = go.Figure(
data=[plot14_11, plot15_11, plot16_11, plot17_11, plot18_11],
layout=layout_10pm)
graphJSON_10pm = json.dumps(graph_10pm, cls=plotly.utils.PlotlyJSONEncoder)
return render_template('isfYear.html',
version=app.config['VERSION'],
page_title='Minerva Compare ISF Values',
graphJSON_all=graphJSON_all,
graphJSON_0am=graphJSON_0am,
graphJSON_2am=graphJSON_2am,
graphJSON_4am=graphJSON_4am,
graphJSON_6am=graphJSON_6am,
graphJSON_8am=graphJSON_8am,
graphJSON_10am=graphJSON_10am,
graphJSON_12pm=graphJSON_12pm,
graphJSON_2pm=graphJSON_2pm,
graphJSON_4pm=graphJSON_4pm,
graphJSON_6pm=graphJSON_6pm,
graphJSON_8pm=graphJSON_8pm,
graphJSON_10pm=graphJSON_10pm)
def box_plot(data):
data = [
go.Box(
y = data[0],
marker_color = '#76323F',
boxmean=True,
name = '1'
),
go.Box(
y = data[1],
marker_color = '#D4D1D3',
boxmean=True,
name = '2'
),
go.Box(
y = data[2],
marker_color = '#565656',
boxmean=True,
name = '3'
),
go.Box(
y = data[3],
marker_color = '#C09F80',
boxmean=True,
name = '4'
)
]
layout = go.Layout(
template = 'plotly_white',
autosize = False,
bargap = 0.35,
font = {
'family': 'Raleway',
'size': 10
},
height = 330,
legend = {
'x': -0.0228945952895,
'y': -0.189563896463,
'orientation': 'h',
'yanchor': 'top'
},
margin = {
'r': 0,
't': 20,
'b': 10,
'l': 10
},
showlegend = False,
title = '',
width = 330,
xaxis = {
'autorange': True,
'showline': True,
'title': '',
'type': 'category'
},
yaxis = {
'autorange': True,
'showgrid': True,
'showline': True,
'title': '',
'type': 'linear',
'zeroline': False
},
transition = {
'duration': 500,
'easing': 'cubic-in-out'
}
)
figure = go.Figure(data=data, layout=layout)
return figure
'#B1CEE6', '#7BD1C7', '#689FA8', '#1F3451', '#6DB29E', '#B1CEE6',
'#7BD1C7', '#689FA8'
]
fig = go.Figure()
for xd, yd, cls in zip(x_data, y_data, colors):
fig.add_trace(
go.Box(y=yd,
name=xd,
boxpoints='all',
jitter=0.5,
whiskerwidth=0.2,
fillcolor=cls,
marker={
'color': '#47535C',
'size': 2,
'line': {
'color': cls,
'width': 2
}
},
line_width=1))
fig.update_layout(yaxis=dict(
autorange=True,
showgrid=True,
zeroline=True,
dtick=250,
gridcolor='rgb(255, 255, 255)',
gridwidth=1,
data_frame['finishedSqFt'] = data_frame['finishedSqFt'].astype('int64')
data_frame['lotsizeSqFt'] = data_frame['lotsizeSqFt'].astype('int64')
print(data_frame['amount'].dtype)
print(data_frame['finishedSqFt'].dtype)
print(data_frame['lotsizeSqFt'].dtype)
data_frame = data_frame.loc[(data_frame['amount'] < 2000000)
& (data_frame['finishedSqFt'] < 50000) &
(data_frame['amount'] > 30000)]
###
# box plot
###
# Create a trace for a box plot
trace = go.Box(y=data_frame['amount'], name='amount', boxpoints='all')
# Assign it to an iterable object named myData
my_data = [trace]
# Add axes and title
my_layout = go.Layout(title='Box plot for house price of Zillow data')
# Setup figure
my_figure = go.Figure(data=my_data, layout=my_layout)
# Create the box plot
py.plot(my_figure, filename='box_zillow_amount')
###
# 2d density plot
age_bin5 = list()
i = 0
for d, a in itertools.izip(data[gene], data['Age']):
if int(a) <= 55:
age_bin1.append(d)
if 55 < int(a) <= 65:
age_bin2.append(d)
if 65 < int(a) <= 75:
age_bin3.append(d)
if 75 < int(a) <= 55:
age_bin4.append(d)
if int(a) > 85:
age_bin5.append(d)
trace1 = go.Box(y=age_bin1, name="Under 55")
trace2 = go.Box(y=age_bin2, name="55-65")
trace3 = go.Box(y=age_bin3, name="65-75")
trace4 = go.Box(y=age_bin4, name="75-85")
trace5 = go.Box(y=age_bin5, name="Over 85")
traces = [trace1, trace2, trace3, trace4, trace5]
layout = go.Layout(
title="CNA Value By Age for Gene: {}".format(gene),
xaxis=dict(title="CNA Value"),
yaxis=dict(title="Age"))
figure = go.Figure(data=traces, layout=layout)
plotly.offline.plot(figure,
filename="{}_correlations.html".format(gene))
def box_plot(df,
groupby=None,
val=None,
figsize=(1024, 512),
jitter=None,
marker_alpha=1,
marker_mode=None,
title='',
ylabel=''):
"""
Visualize box plots for all columns in the dataframe <df>.
Parameters
----------
df: pandas dataframe
each column represents a categorical variable; samples are
along columns.
groupby: str
The name of a column to define groups. If None provided,
we assume equal sample sizes and use columns as groups.
val: str
Used in conjuction with <groupby> to choose column values to plot.
If None, then we choose the first column (excluding <groupby>)
jitter: float (0-1)
The proportion of each box area to jitter datapoints
marker_alpha: float (0-1)
The opacity of the data points plotted
boxpoint_mode: 'all','suspectedoutliers', 'Outliers', Boolean, or None
Specifies the way that datapoints are plotted.
figsize: tuple
The (width, height) of the figure in pixels
title: str
The figure title
ylabel: str
The name of the y axis
outfile: filepath str
If provided, output to an HTML file at provided location
Example
-------
from sklearn.datasets import make_classification
N_FEATURES = 4
X, y = make_classification(n_samples=100, n_clusters_per_class=1, n_classes=4, n_features=N_FEATURES)
df = pd.DataFrame(X, columns=['feature_%d' % f for f in range(N_FEATURES)])
df['class'] = y
box_plot(df, title='Box Plot')
"""
layout = go.Layout(title=title,
height=figsize[1],
width=figsize[0],
yaxis=go.YAxis(title=ylabel))
data = []
if groupby is None:
for col in df.columns:
data.append(go.Box(y=df[col], name=col))
else:
groups = sorted(df[groupby].unique().tolist())
if val is None:
val = df.columns.drop(groupby)[
0] # choose first non-groupby column
for group in groups:
mask = df[groupby] == group
data.append(
go.Box(y=df.loc[mask, val],
name=group,
jitter=jitter,
boxpoints=marker_mode,
marker=dict(opacity=marker_alpha)))
ol.iplot(go.Figure(data=data, layout=layout), show_link=False)
def plot_box(data: Union["ps.DataFrame", "ps.Series"], **kwargs):
import plotly.graph_objs as go
import pyspark.pandas as ps
if isinstance(data, ps.DataFrame):
raise RuntimeError(
"plotly does not support a box plot with Koalas DataFrame. Use Series instead."
)
# 'whis' isn't actually an argument in plotly (but in matplotlib). But seems like
# plotly doesn't expose the reach of the whiskers to the beyond the first and
# third quartiles (?). Looks they use default 1.5.
whis = kwargs.pop("whis", 1.5)
# 'precision' is Koalas specific to control precision for approx_percentile
precision = kwargs.pop("precision", 0.01)
# Plotly options
boxpoints = kwargs.pop("boxpoints", "suspectedoutliers")
notched = kwargs.pop("notched", False)
if boxpoints not in ["suspectedoutliers", False]:
raise ValueError(
"plotly plotting backend does not support 'boxpoints' set to '%s'. "
"Set to 'suspectedoutliers' or False." % boxpoints)
if notched:
raise ValueError(
"plotly plotting backend does not support 'notched' set to '%s'. "
"Set to False." % notched)
colname = name_like_string(data.name)
spark_column_name = data._internal.spark_column_name_for(
data._column_label)
# Computes mean, median, Q1 and Q3 with approx_percentile and precision
col_stats, col_fences = BoxPlotBase.compute_stats(data, spark_column_name,
whis, precision)
# Creates a column to flag rows as outliers or not
outliers = BoxPlotBase.outliers(data, spark_column_name, *col_fences)
# Computes min and max values of non-outliers - the whiskers
whiskers = BoxPlotBase.calc_whiskers(spark_column_name, outliers)
fliers = None
if boxpoints:
fliers = BoxPlotBase.get_fliers(spark_column_name, outliers,
whiskers[0])
fliers = [fliers] if len(fliers) > 0 else None
fig = go.Figure()
fig.add_trace(
go.Box(
name=colname,
q1=[col_stats["q1"]],
median=[col_stats["med"]],
q3=[col_stats["q3"]],
mean=[col_stats["mean"]],
lowerfence=[whiskers[0]],
upperfence=[whiskers[1]],
y=fliers,
boxpoints=boxpoints,
notched=notched,
**
kwargs, # this is for workarounds. Box takes different options from express.box.
))
fig["layout"]["xaxis"]["title"] = colname
fig["layout"]["yaxis"]["title"] = "value"
return fig
plotly.tools.set_credentials_file(username='******', api_key='hPKWVIYRADr9YBrkCJKL')
import plotly.plotly as py
import plotly.graph_objs as go
x = ['$100', '$100', '$100', '$100', '$100', '$100',
'$120', '$120', '$120', '$120', '$120', '$120',
'$140', '$140', '$140', '$140', '$140', '$140',
'$160', '$160', '$160', '$160', '$160', '$160',
'$180', '$180', '$180', '$180', '$180', '$180',
'$200', '$200', '$200', '$200', '$200', '$200']
trace0 = go.Box(
y=[-100, -50, 48218, 48218, 126144, 190174, -100, 4595, 65295, 65295, 160411, 239776,
-100, 9938, 82335, 82335, 194883, 289253, 7995, 20713, 116389, 116389, 263419, 388706,
12200, 26172, 133416, 133416, 297892, 438704],
x=x,
name='zv50',
marker=dict(
color='#2388D0'
)
)
trace1 = go.Box(
y=[-100, -50, 16723, 16723, 32160, 50672, -100, 3348, 24494, 24494, 43291, 66795,
-100, 7844, 32343, 32343, 54540, 83021, 2621, 12440, 40077, 40077, 65819, 99318,
6458, 17052, 47905, 47905, 76965, 115451, 10151, 21691, 55856, 55856, 88191, 131697],
x=x,
name='zv60',
marker=dict(
color='#2F8FD3'
)
)
trace2 = go.Box(
max_txss.insert(position, max)
# fig = go.Figure()
# fig.add_scatter(x=tc,
# y=txs,
# mode='markers',
# line = dict(
# color = ('rgb(22, 96, 167)'),
# width = 1)
# )
# pio.write_image(fig, "../Backend/graphs/9/sim/"+log.split('.')[0]+".svg")
traces = go.Box(y=time_totals,
name='9 Sim Agents',
jitter=0.3,
pointpos=-1.8,
boxpoints='all',
marker=dict(color='rgb(10, 140, 208)', ),
boxmean=True,
showlegend=False)
tracer = go.Box(y=time_total,
name='9 Real Agents',
jitter=0.3,
pointpos=-1.8,
boxpoints='all',
marker=dict(color='rgb(8, 81, 156)', ),
boxmean=True,
showlegend=False)
data = [traces, tracer]
ys = [
responses_df['Q35_Part_1'].values, responses_df['Q35_Part_2'].values,
responses_df['Q35_Part_3'].values, responses_df['Q35_Part_4'].values,
responses_df['Q35_Part_5'].values, responses_df['Q35_Part_6'].values
]
names = [
"Self-taught", 'Online courses (Coursera, Udemy, edX, etc.)', 'Work',
'University', 'Kaggle competitions', 'Other'
]
#colors = ['rgba(93, 164, 214, 0.5)', 'rgba(255, 144, 14, 0.5)', 'rgba(44, 160, 101, 0.5)', 'rgba(255, 65, 54, 0.5)', 'rgba(207, 114, 255, 0.5)', 'rgba(127, 96, 0, 0.5)']
colors = ["#3498db", "#95a5a6", "#e74c3c", "#34495e", "#2ecc71", "#df6a84"]
trace = []
for i in range(6):
trace.append(go.Box(y=ys[i], name=names[i],
marker=dict(color=colors[i], )))
layout = go.Layout(
title='Box plots on % contribution of each ML / DS training category')
fig = go.Figure(data=trace, layout=layout)
iplot(fig, filename="TimeSpent")
# **Observations:**
#
# * Looking at the median of each of the learning categories, it seems there is no one category that completely dominated the learning process of ML / DS
# * Self-taught seems to have higher percentage of share in the learning process compared to others.
# * Only less than half of the respondents have the percentage share of 'University' as greater than 0
#
#
# ### Distribution of DS / ML Learning Category at different Countries:
#
#######
# Objective: Make a DataFrame using the Abalone dataset (../data/abalone.csv).
# Take two independent random samples of different sizes from the 'rings' field.
# HINT: np.random.choice(df['rings'],10,replace=False) takes 10 random values
# Use box plots to show that the samples do derive from the same population.
######
# Perform imports here:
import plotly.offline as pyo
import plotly.graph_objs as go
import pandas as pd
import numpy as np
# create a DataFrame from the .csv file:
df = pd.read_csv('data/abalone.csv')
# take two random samples of different sizes:
sample1 = np.random.choice(df['rings'],15,replace=False)
sample2 = np.random.choice(df['rings'],45,replace=False)
# create a data variable with two Box plots:
data = [go.Box(y=sample1,name='Sample1'),go.Box(y=sample2,name='Sample2')]
# add a layout
layout = go.Layout(title='Two Samples')
# create a fig from data & layout, and plot the fig
fig = go.Figure(data=data,layout=layout)
pyo.plot(fig)
net_data[tr_exploit][test_exploit].append(d)
summaries[tr_exploit] = get_summary(data)
with open(loc_str.format(tr_exploit), 'w') as f:
json.dump(data, f, indent=2)
summaries['net'] = get_summary(raw_data)
with open(loc_str.format('summary'), 'w') as f:
json.dump(summaries, f, indent=2)
# plot data
for tr_exploit in exploits:
accs = [[entry['accuracy'] for entry in net_data[tr_exploit][ex]]
for ex in exploits]
boxes = [go.Box(
y=accs[i],
name=exploits[i],
boxmean='sd'
) for i in range(len(exploits))]
layout = go.Layout(
title='ADD-GAN: Accuracy per Exploit Trained on {}'.format(tr_exploit),
yaxis=dict(title='Accuracy (%)')
)
fig = go.Figure(data=boxes, layout=layout)
py.plot(fig, filename='add-gan-cross-results-{}'.format(tr_exploit))
item_2 = 0
item_3 = 0
item_4 = 0
item_5 = 0
for item_i in heartDisease[item]:
for i in range(0,6):
if (item == i):
item_i +=1
heartDisease_i = 0
for i in range (0,6):
heartDisease_i = (item_i/len(heartDisease)) * 100
print("The percentage of level", i, "in the response variable is: {0:.2f}".format(heartDisease_i))
classImbalance('heartdisease')
trace0 = go.Box(
y=heartDisease['age'],
name='age'
)
trace1 = go.Box(
y=heartDisease['sex'],
name='sex'
)
trace2 = go.Box(
y=heartDisease['cp'],
name='cp'
)
trace3 = go.Box(
y=heartDisease['trestbps'],
name='trestbps'
)
trace4 = go.Box(
y=heartDisease['chol'],
df3 = df3.groupby(['TownID', 'Latitude', 'Longitude'])['Total'].sum()
df3 = df3.values.tolist()
df4 = pd.read_csv("F_area.csv", encoding='utf-8')
df4 = df4[['Total', 'Longitude', 'Latitude', 'TownID']]
df4 = df4.groupby(['TownID', 'Latitude', 'Longitude'])['Total'].sum()
df4 = df4.values.tolist()
df5 = pd.read_csv("C_area.csv", encoding='utf-8')
df5 = df5[['Total', 'Longitude', 'Latitude', 'TownID']]
df5 = df5.groupby(['TownID', 'Latitude', 'Longitude'])['Total'].sum()
df5 = df5.values.tolist()
trace1 = go.Box(
y=df1,
name="台北市",
boxpoints='outliers',
marker=dict(color='rgba(93, 164, 214, 0.5)', ),
line=dict(
width=1,
color='rgba(93, 164, 214, 0.5)',
),
)
trace2 = go.Box(
y=df2,
name="桃園市",
boxpoints='outliers',
marker=dict(color='rgba(255, 144, 14, 0.5)', ),
line=dict(
width=1,
color='rgba(255, 144, 14, 0.5)',
),
)
trace3 = go.Box(
def data_visualization(data_set):
fig_check_null = px.imshow(data_set.isnull(),
labels=dict(x='Features', y="Samples"),
x=list(data_set.columns.values),
contrast_rescaling='minmax')
fig_check_null.update_layout(title={
'text': "NaN values detection",
'y': 0.95,
'x': 0.5,
'xanchor': 'center',
'yanchor': 'top'
},
showlegend=False)
fig_check_null.show()
# removed features with multiple NaN values
data_set = data_set.drop([
'elevation_gain', 'elevation_loss', 'max_run_cadence', 'steps',
'avg_stride_length', 'min_elevation', 'max_elevation',
'avg_double_cadence', 'max_double_cadence', 'max_vertical_speed',
'water_estimated', 'min_respiration_rate', 'max_respiration_rate',
'avg_respiration_rate', 'activity_training_load'
],
axis=1)
corr = data_set.corr()
trace = go.Heatmap(z=corr.values,
x=corr.index.values,
y=corr.columns.values)
data = [trace]
fig_corr = go.Figure(data=data,
layout={
'title': {
'text': "Correlation Matrix",
'y': 0.9,
'x': 0.5,
'xanchor': 'center',
'yanchor': 'top'
},
})
fig_corr.show()
# removed features with low correlation with activity type
data_set = data_set.drop(['event_type_id', 'moving_duration'], axis=1)
# removed features highly correlated with others
data_set = data_set.drop([
'aerobic_training_effect', 'avg_hr', 'start_time_local',
'start_time_gmt', 'elapsed_duration', 'max_temperature'
],
axis=1)
fig_box = make_subplots(rows=1, cols=len(data_set.columns.values))
col_count = 1
for col in data_set:
fig_box.add_trace(go.Box(y=data_set[col].values,
name=data_set[col].name),
row=1,
col=col_count)
col_count += 1
fig_box.update_layout(title={
'text': "Outliers Detection",
'y': 0.9,
'x': 0.5,
'xanchor': 'center',
'yanchor': 'top'
},
showlegend=False)
fig_box.show()
return data_set
def make_pb1(color):
trace22 = go.Box(x=qdata.Open.pct_change(), marker=dict(color=color))
data22 = [trace22]
f22 = dict(data=data22)
return f22
def sessionsbytime_figure(df, selected_groupby):
fig = plotly.subplots.make_subplots(rows=1, cols=1)
fig.update_layout(margin=dict(l=40, r=40, t=40, b=40))
# TODO: if weekly is chosen, show the actual session name instead of a dot
# TODO: use different shapes for PET vs MR
# TODO: try to connect baseline with followup with arc line or something
# or could have "by subject" choice that has a subject per y value
# Customize figure
#fig['layout'].update(xaxis={'automargin': True}, yaxis={'automargin': True})
from itertools import cycle
import plotly.express as px
palette = cycle(px.colors.qualitative.Plotly)
#palette = cycle(px.colors.qualitative.Vivid)
#palette = cycle(px.colors.qualitative.Bold)
for mod, sesstype in itertools.product(df.MODALITY.unique(), df.SESSTYPE.unique()):
#print(sesstype, mod)
# Get subset for this session type
dfs = df[(df.SESSTYPE == sesstype) & (df.MODALITY == mod)]
# Nothing to plot so go to next session type
if dfs.empty:
continue
# Plot base on view
view = 'default'
if view == "month":
# TBD
pass
elif view == 'all':
# Let's do this for the all time view to see histograms by year
# or quarter or whatever fits well
# Plot this session type
fig.append_trace(
go.Histogram(
hovertext=dfs['SESSION'],
name='{} ({})'.format(sesstype, len(dfs)),
x=dfs['DATE'],
y=dfs['PROJECT'],
),
_row,
_col)
elif view == 'weekly':
# Let's do this only for the weekly view and customize it specifically
# for Mon thru Fri and allow you to choose this week and last week
dfs['ONE'] = 1
# Plot this session type
fig.append_trace(
go.Bar(
hovertext=dfs['SESSION'],
name='{} ({})'.format(sesstype, len(dfs)),
x=dfs['DATE'],
y=dfs['ONE'],
),
_row,
_col)
# width function of number of days being plotted
#@width =
#print(fig.layout.xaxis.width)
fig.update_layout(
barmode='stack',
width=900,
#bargroupgap=0,
#wbidth=100,
bargap=0.1)
else:
# Create boxplot for this var and add to figure
# Default to the jittered boxplot with no boxes
# markers symbols, see https://plotly.com/python/marker-style/
if mod == 'MR':
symb = 'circle-dot'
elif mod == 'PET':
symb = 'diamond-wide-dot'
else:
symb = 'diamond-tall-dot'
_color = next(palette)
# Convert hex to rgba with alpha of 0.5
if _color.startswith('#'):
_rgba = 'rgba({},{},{},{})'.format(
int(_color[1:3], 16),
int(_color[3:5], 16),
int(_color[5:7], 16),
0.7)
else:
_r,_g,_b = _color[4:-1].split(',')
_a = 0.7
_rgba = 'rgba({},{},{},{})'.format(_r, _g, _b, _a)
# Plot this session type
_row = 1
_col = 1
fig.append_trace(
go.Box(
name='{} {} ({})'.format(sesstype, mod, len(dfs)),
x=dfs['DATE'],
y=dfs[selected_groupby],
boxpoints='all',
jitter=0.7,
text=dfs['SESSION'],
pointpos=0.5,
orientation='h',
marker={
'symbol': symb,
'color': _rgba,
'size': 12,
'line': dict(width=2, color=_color)
},
line={'color': 'rgba(0,0,0,0)'},
fillcolor='rgba(0,0,0,0)',
hoveron='points',
),
_row,
_col)
# show lines so we can better distinguish categories
fig.update_yaxes(showgrid=True)
#fig.update_xaxes(range=[])
#full_fig = fig.full_figure_for_development()
#print(full_fig.layout.xaxis.range)
x_mins = []
x_maxs = []
for trace_data in fig.data:
x_mins.append(min(trace_data.x))
x_maxs.append(max(trace_data.x))
x_min = min(x_mins)
x_max = max(x_maxs)
#print('x_min=', x_min, 'x_max=', x_max)
if x_min == '2021-11-01' or x_min == '2021-11-10':
fig.update_xaxes(
range=('2021-10-31', '2021-12-01'),
tickvals=[
'2021-11-01',
'2021-11-08',
'2021-11-15',
'2021-11-22',
'2021-11-29'])
fig.update_layout(width=900)
return fig
def isfplots():
'''A box-and-whisker plot with isf data sorted in 2-hr time buckets '''
(all, bucket_list) = isf.get_all_isf_plus_buckets()
allData = [data[0] for data in all if data[0]]
all_whisker = go.Box(y=allData, name='all isf')
bucket0 = go.Box(y=bucket_list[0], name='0am-2am')
bucket1 = go.Box(y=bucket_list[1], name='2am-4am')
bucket2 = go.Box(y=bucket_list[2], name='4am-6am')
bucket3 = go.Box(y=bucket_list[3], name='6am-8am')
bucket4 = go.Box(y=bucket_list[4], name='8am-10am')
bucket5 = go.Box(y=bucket_list[5], name='10am-12pm')
bucket6 = go.Box(y=bucket_list[6], name='12pm-14pm')
bucket7 = go.Box(y=bucket_list[7], name='14pm-16pm')
bucket8 = go.Box(y=bucket_list[8], name='16pm-18pm')
bucket9 = go.Box(y=bucket_list[9], name='18pm-20pm')
bucket10 = go.Box(y=bucket_list[10], name='20pm-22pm')
bucket11 = go.Box(y=bucket_list[11], name='22pm-24pm')
layout = go.Layout(
title=('isf values'),
width=1500,
height=1000,
yaxis=dict(
title='mgdl/unit',
# the y zeroline is the line where y=0
zeroline=True,
zerolinecolor='#800000',
zerolinewidth=2,
# this is the vertical line at the left edge
showline=False,
rangemode='tozero'))
graph = go.Figure(data=[
all_whisker, bucket0, bucket1, bucket2, bucket3, bucket4, bucket5,
bucket6, bucket7, bucket8, bucket9, bucket10, bucket11
],
layout=layout)
graphJSON = json.dumps(graph, cls=plotly.utils.PlotlyJSONEncoder)
return render_template('isfplots.html',
version=app.config['VERSION'],
page_title='Minerva ISF values',
graphJSON=graphJSON)
import plotly.graph_objs as go
fig = go.Figure()
fig.add_trace(go.Box(
x=[2,3,1,5],
y=[
['First','First','First','First'],
["A","A","A","A"]
],
name="A",
orientation="h"
))
fig.add_trace(go.Box(
x=[8,3,6,5],
y=[
['First','First','First','First'],
["B","B","B","B"]
],
name="B",
orientation="h"
))
fig.add_trace(go.Box(
x=[2,3,2,5],
y=[
['Second','Second','Second','Second'],
["C","C","C","C"]
],
name="C",
orientation="h"
))
fig.add_trace(go.Box(
x=[7.5,3,6,4],
for i in sorted(df.year.unique())],
name='sum')
]
})
],
className="six columns"),
],
className="row"),
html.Div([
html.H3('Nombre de jours d\'accomplissement par année'),
dcc.Graph(
id='g3',
figure={
'data': [
go.Box(y=df[df.year == i]['days_to_complete'].tolist(),
name=str(i),
boxpoints=False) for i in sorted(df.year.unique())
]
})
]),
html.U(html.H2('Détails')),
html.Div([
html.Div([
dcc.Slider(
id='year-slider',
min=df['year'].min(),
max=df['year'].max(),
value=df['year'].min(),
marks={str(year): str(year)
for year in df['year'].unique()})
],
# Nearly 80% of the borrowers are female.
#
# **Countrywise Loan Amount Distribution:**
#
# Now let us look at the loan amount distribution at country level.
# In[ ]:
trace = []
for name, group in kiva_loans_df.groupby("country"):
trace.append (
go.Box(
x=group["loan_amount_trunc"].values,
name=name
)
)
layout = go.Layout(
title='Loan Amount Distribution by country',
width = 800,
height = 2000
)
#data = [trace0, trace1]
fig = go.Figure(data=trace, layout=layout)
py.iplot(fig, filename="LoanAmountCountry")
# **Sectorwise Loan Amount distribution:**
# In[ ]:
y_data = [y0, y1, y2, y3, y4]
colors = [
'rgba(93, 164, 214, 0.5)', 'rgba(255, 144, 14, 0.5)',
'rgba(44, 160, 101, 0.5)', 'rgba(255, 65, 54, 0.5)',
'rgba(207, 114, 255, 0.5)', 'rgba(127, 96, 0, 0.5)'
]
traces = []
for xd, yd, cls in zip(x_data, y_data, colors):
traces.append(
go.Box(
y=yd,
name=xd,
boxpoints=False,
jitter=0.5,
whiskerwidth=0.2,
fillcolor=cls,
marker=dict(size=2, ),
line=dict(width=1),
))
layout = go.Layout(
title='Difference in sales {} from cluster to cluster'.format(
field_to_plot),
yaxis=dict(
autorange=True,
showgrid=True,
zeroline=True,
dtick=50,
gridcolor='black',
gridwidth=0.1,
def update_graph(festival_name, genre_name):
if festival_name == 'Sundance':
s, s_winner, s_loser = newDataset(sundance, genre_name)
elif festival_name == 'Tribeca':
s, s_winner, s_loser = newDataset(tribeca, genre_name)
elif festival_name == 'Chicago':
s, s_winner, s_loser = newDataset(chicago, genre_name)
elif festival_name == 'Berlin':
s, s_winner, s_loser = newDataset(berlin, genre_name)
elif festival_name == 'Rotterdam':
s, s_winner, s_loser = newDataset(rotterdam, genre_name)
elif festival_name == 'Cannes':
s, s_winner, s_loser = newDataset(cannes, genre_name)
elif festival_name == 'Venice':
s, s_winner, s_loser = newDataset(venice, genre_name)
elif festival_name == 'SXSW':
s, s_winner, s_loser = newDataset(sxsw, genre_name)
elif festival_name == 'Seattle':
s, s_winner, s_loser = newDataset(seattle, genre_name)
elif festival_name == 'SanFrancisco':
s, s_winner, s_loser = newDataset(san, genre_name)
elif festival_name == 'Slamdance':
s, s_winner, s_loser = newDataset(slam, genre_name)
elif festival_name == 'Locarno':
s, s_winner, s_loser = newDataset(locarno, genre_name)
elif festival_name == 'Sitges':
s, s_winner, s_loser = newDataset(sitges, genre_name)
elif festival_name == 'Toronto':
s, s_winner, s_loser = newDataset(toronto, genre_name)
elif festival_name == 'KarlovyVary':
s, s_winner, s_loser = newDataset(kv, genre_name)
elif festival_name == 'HongKong':
s, s_winner, s_loser = newDataset(hongkong, genre_name)
elif festival_name == 'Austin':
s, s_winner, s_loser = newDataset(austin, genre_name)
elif festival_name == 'Torino':
s, s_winner, s_loser = newDataset(torino, genre_name)
elif festival_name == 'Marrakech':
s, s_winner, s_loser = newDataset(marrakech, genre_name)
elif festival_name == 'Tokyo':
s, s_winner, s_loser = newDataset(tokyo, genre_name)
elif festival_name == 'GoldenHorse':
s, s_winner, s_loser = newDataset(goldenhorse, genre_name)
elif festival_name == 'BuenosAires':
s, s_winner, s_loser = newDataset(buenosaires, genre_name)
elif festival_name == 'Gramado':
s, s_winner, s_loser = newDataset(gramado, genre_name)
elif festival_name == 'Cairo':
s, s_winner, s_loser = newDataset(cairo, genre_name)
elif festival_name == 'Havana':
s, s_winner, s_loser = newDataset(havana, genre_name)
elif festival_name == 'Rio':
s, s_winner, s_loser = newDataset(rio, genre_name)
elif festival_name == 'SaoPaulo':
s, s_winner, s_loser = newDataset(saopaulo, genre_name)
elif festival_name == 'AsiaPacific':
s, s_winner, s_loser = newDataset(asiapacific, genre_name)
elif festival_name == 'India':
s, s_winner, s_loser = newDataset(india, genre_name)
elif festival_name == 'Sydney':
s, s_winner, s_loser = newDataset(sydney, genre_name)
elif festival_name == 'Beijing':
s, s_winner, s_loser = newDataset(beijing, genre_name)
elif festival_name == 'TokyoF':
s, s_winner, s_loser = newDataset(tokyof, genre_name)
elif festival_name == 'AAFCA':
s, s_winner, s_loser = newDataset(aafca, genre_name)
s, s_winner, s_loser = newDataset(brisbane, genre_name)
elif festival_name == 'Jerusalem':
s, s_winner, s_loser = newDataset(jerusalem, genre_name)
elif festival_name == 'Haifa':
s, s_winner, s_loser = newDataset(haifa, genre_name)
elif festival_name == 'GrandBell':
s, s_winner, s_loser = newDataset(grandbell, genre_name)
elif festival_name == 'Fajr':
s, s_winner, s_loser = newDataset(fajr, genre_name)
elif festival_name == 'Singapore':
s, s_winner, s_loser = newDataset(singapore, genre_name)
elif festival_name == 'Yamagata':
s, s_winner, s_loser = newDataset(yamagata, genre_name)
elif festival_name == 'Shanghai':
s, s_winner, s_loser = newDataset(shanghai, genre_name)
elif festival_name == 'Kerala':
s, s_winner, s_loser = newDataset(kerala, genre_name)
elif festival_name == 'Taipei':
s, s_winner, s_loser = newDataset(taipei, genre_name)
elif festival_name == 'Jeonju':
s, s_winner, s_loser = newDataset(jeonju, genre_name)
elif festival_name == 'Moscow':
s, s_winner, s_loser = newDataset(moscow, genre_name)
elif festival_name == 'Edinburgh':
s, s_winner, s_loser = newDataset(edinburgh, genre_name)
elif festival_name == 'Mannheim-Heidelberg':
s, s_winner, s_loser = newDataset(mannheimheidelberg, genre_name)
elif festival_name == 'San Sebastián':
s, s_winner, s_loser = newDataset(sansebastián , genre_name)
elif festival_name == 'Taormina':
s, s_winner, s_loser = newDataset(taormina, genre_name)
elif festival_name == 'London':
s, s_winner, s_loser = newDataset(london, genre_name)
elif festival_name == 'Thessaloniki':
s, s_winner, s_loser = newDataset(thessaloniki, genre_name)
allfilm = newbo[newbo['Genre_' + festival_name] == genre_name]
allfilm_outlier, allfilm_new = outliers(allfilm, feature)
s_outlier, s_new = outliers(s, feature)
s_winner_new = drop_contenders_new(s_winner, s_new, False)
s_loser_new = drop_contenders_new(s_loser, s_new, False)
allmean = cal_mean(allfilm_new, feature)
smean = cal_mean(s_new, feature)
slmean = cal_mean(s_loser_new, feature)
swmean = cal_mean(s_winner_new, feature)
title1 = festival_name + ' ' + genre_name + ' Box Office Boxplot'
title2 = genre_name + ' Films Average Box Office'
figure = go.Figure(data=[go.Box(y=s_winner_new[feature], name='Winners', marker_color='#C4DFE6'),
go.Box(y=s_loser_new[feature], name='Nominees', marker_color='#66A5AD'),
go.Box(y=s_new[feature], name='Contenders', marker_color='#07575B'),
go.Box(y=allfilm_new[feature], name='Non-Contenders', marker_color='#003B46')],
layout=go.Layout(title={'text': title1, 'x': 0.5}))
figure2 = go.Figure(data=[go.Bar(x=[allmean, smean, slmean, swmean],
y=['others', 'contenders', 'nominee', 'winner'],
text=combine[festival_name]['boxoffice'][genre_name],
textposition='auto',
marker_color=['#003B46', '#07575B', '#66A5AD', '#C4DFE6'],
orientation='h')],
layout=go.Layout(title={'text': title2, 'x': 0.5}))
figure.update_layout(
paper_bgcolor='#1e2130',
plot_bgcolor='#1e2130',
legend={'font': {'color': 'darkgray'}},
font={'color': 'darkgray'},
showlegend=True
)
figure2.update_layout(
paper_bgcolor='#1e2130',
plot_bgcolor='#1e2130',
legend={'font': {'color': 'darkgray'}},
font={'color': 'darkgray'},
showlegend=False
)
return figure, figure2
def main(_type='alpha',
_low_connectivity=False,
_plots=None,
_plot_types=None):
if _plots is None:
_plots = ['same', 'different']
if _plot_types is None:
_plot_types = ['stacked_bar', 'box', 'bar']
_same_arrays = []
_different_arrays = []
_same_highest = []
_different_highest = []
if _type == 'alpha':
_alphas = ALPHAS
_betas = [BETA] * len(ALPHAS)
_names = _alphas
elif _type == 'beta':
_alphas = [ALPHA] * len(BETAS)
_betas = BETAS
_names = _betas
else:
raise Exception(
'No such type. Only \'alpha\' or \'beta\' are acceptable.')
for _alpha, _beta in zip(_alphas, _betas):
print('Alpha:', _alpha, 'beta:', _beta)
_, _same_time_lags_arrays, _different_time_lags_arrays, _same_time_lags_highest, \
_different_time_lags_highest = same_inner_correlation_vs_different_inner_correlation_cross_correlation.compute_fiber_densities(
_alpha=_alpha, _beta=_beta, _low_connectivity=_low_connectivity)
_same_arrays.append(_same_time_lags_arrays[TIME_LAG_INDEX])
_different_arrays.append(_different_time_lags_arrays[TIME_LAG_INDEX])
_same_highest.append(_same_time_lags_highest)
_different_highest.append(_different_time_lags_highest)
if _plots is not None:
# stacked bar plot
if 'stacked_bar' in _plot_types:
for _name, _sums in zip(['same', 'different'],
[_same_highest, _different_highest]):
if _name in _plots:
_y_arrays = [[], [], []]
for _type_sums in _sums:
_left_wins, _none_wins, _right_wins = 0, 0, 0
for _time_lag, _type_sum in zip(
same_inner_correlation_vs_different_inner_correlation_cross_correlation
.TIME_LAGS, _type_sums):
if _time_lag > 0:
_left_wins += _type_sum
elif _time_lag < 0:
_right_wins += _type_sum
else:
_none_wins += _type_sum
_total = sum(_type_sums)
_y_arrays[0].append(_left_wins / _total)
_y_arrays[1].append(_none_wins / _total)
_y_arrays[2].append(_right_wins / _total)
_colors_array = config.colors(3)
_fig = go.Figure(data=[
go.Bar(x=_names,
y=_y_array,
name=_name,
marker={'color': _color}) for _name, _y_array,
_color in zip(['Leader', 'None', 'Follower'],
_y_arrays, _colors_array)
],
layout={
'xaxis': {
'title': _type.capitalize(),
'zeroline': False,
'tickmode': 'array',
'tickvals': _names,
'type': 'category'
},
'yaxis': {
'title':
'Highest correlation fraction',
'range': [0, 1.1],
'zeroline': False,
'tickmode': 'array',
'tickvals': [0, 0.5, 1]
},
'barmode': 'stack',
'legend': {
'xanchor': 'right',
'yanchor': 'top',
'bordercolor': 'black',
'borderwidth': 2
},
})
save.to_html(_fig=_fig,
_path=os.path.join(paths.PLOTS,
save.get_module_name()),
_filename='plot_stacked_bar_' + _type +
'_low_con_' + str(_low_connectivity) + '_' +
_name)
# box plot
if 'box' in _plot_types:
for _name, _arrays in zip(['same', 'different'],
[_same_arrays, _different_arrays]):
if _name in _plots:
_fig = go.Figure(
data=[
go.Box(y=_y,
name=_name,
boxpoints=False,
line={'width': 1},
marker={
'size': 10,
'color': '#2e82bf'
},
showlegend=False)
for _y, _name in zip(_arrays, _names)
],
layout={
'xaxis': {
'title': _type.capitalize(),
'zeroline': False,
'tickmode': 'array',
'tickvals': _names,
'type': 'category'
},
'yaxis': {
'title':
'Inner correlation' if _name == 'same' else
'Different network correlation',
'range': [-1, 1.1],
'zeroline':
False,
'tickmode':
'array',
'tickvals': [-1, -0.5, 0, 0.5, 1]
}
})
save.to_html(_fig=_fig,
_path=os.path.join(paths.PLOTS,
save.get_module_name()),
_filename='plot_box_' + _type + '_low_con_' +
str(_low_connectivity) + '_' + _name)
# bar plot
if 'bar' in _plot_types:
for _name, _sums in zip(['same', 'different'],
[_same_highest, _different_highest]):
if _name in _plots:
_fig = go.Figure(data=go.Bar(
x=_names,
y=[
_type_sums[TIME_LAG_INDEX] / sum(_type_sums)
for _type_sums in _sums
],
marker={'color': '#2e82bf'}),
layout={
'xaxis': {
'title': _type.capitalize(),
'zeroline': False,
'tickmode': 'array',
'tickvals': _names,
'type': 'category'
},
'yaxis': {
'title': 'Lag ' + str(TIME_LAG) +
' highest correlation fraction',
'range': [0, 1.1],
'zeroline': False,
'tickmode': 'array',
'tickvals': [0, 0.5, 1]
}
})
save.to_html(_fig=_fig,
_path=os.path.join(paths.PLOTS,
save.get_module_name()),
_filename='plot_bar_' + _type + '_low_con_' +
str(_low_connectivity) + '_' + _name)
def main():
_simulations = load.structured()
_simulations = filtering.by_time_points_amount(_simulations, TIME_POINTS)
_simulations = filtering.by_categories(
_simulations,
_is_single_cell=False,
_is_heterogeneity=True,
_is_low_connectivity=False,
_is_causality=False,
_is_dominant_passive=False,
_is_fibrin=False
)
_simulations = filtering.by_pair_distance(_simulations, _distance=PAIR_DISTANCE)
_simulations = filtering.by_heterogeneity(_simulations, _std=STD)
print('Total simulations:', len(_simulations))
_fiber_densities = compute_fiber_densities(_simulations)
_window_distances_communicating = [[] for _i in OFFSETS_X]
_window_distances_non_communicating = [[] for _i in OFFSETS_X]
# window distances loop
for _window_distance_index, _window_distance in enumerate(OFFSETS_X):
print('Window distance:', _window_distance)
# communicating loop
for _simulation in tqdm(_simulations, desc='Communicating loop'):
_left_cell_fiber_densities = _fiber_densities[(_simulation, _window_distance, 'left_cell')]
_right_cell_fiber_densities = _fiber_densities[(_simulation, _window_distance, 'right_cell')]
_correlation = compute_lib.correlation(
compute_lib.derivative(_left_cell_fiber_densities, _n=DERIVATIVE),
compute_lib.derivative(_right_cell_fiber_densities, _n=DERIVATIVE)
)
_window_distances_communicating[_window_distance_index].append(_correlation)
# non-communicating loop
_simulations_indices = range(len(_simulations))
for _simulation_1_index in tqdm(_simulations_indices, desc='Non-communicating pairs loop'):
_simulation_1 = _simulations[_simulation_1_index]
for _simulation_2_index in _simulations_indices[_simulation_1_index + 1:]:
_simulation_2 = _simulations[_simulation_2_index]
for _simulation_1_cell_id, _simulation_2_cell_id in product(['left_cell', 'right_cell'],
['left_cell', 'right_cell']):
_simulation_1_fiber_densities = \
_fiber_densities[(_simulation_1, _window_distance, _simulation_1_cell_id)]
_simulation_2_fiber_densities = \
_fiber_densities[(_simulation_2, _window_distance, _simulation_2_cell_id)]
_correlation = compute_lib.correlation(
compute_lib.derivative(_simulation_1_fiber_densities, _n=DERIVATIVE),
compute_lib.derivative(_simulation_2_fiber_densities, _n=DERIVATIVE)
)
_window_distances_non_communicating[_window_distance_index].append(_correlation)
# rank sums
print('Wilcoxon rank-sum tests between communicating and non-communicating:',
ranksums(_window_distances_communicating[_window_distance_index],
_window_distances_non_communicating[_window_distance_index]))
# plot
_data = []
_colors_array = config.colors(2)
for _communicating, _communicating_text, _pair_distances, _color in \
zip([True, False], ['Communicating', 'Non-communicating'],
[_window_distances_communicating, _window_distances_non_communicating],
_colors_array):
_y = []
_x = []
for _window_distance_index, _window_distance in enumerate(OFFSETS_X):
_y += _pair_distances[_window_distance_index]
_x += [_window_distance for _i in _pair_distances[_window_distance_index]]
_data.append(
go.Box(
y=_y,
x=_x,
name=_communicating_text,
boxpoints='all' if _communicating else False,
jitter=1,
pointpos=0,
line={
'width': 1
},
fillcolor='white',
marker={
'size': 10,
'color': _color
},
opacity=0.7
)
)
_fig = go.Figure(
data=_data,
layout={
'xaxis': {
'title': 'Window distance (cell diameter)',
'zeroline': False,
'tickmode': 'array',
'tickvals': OFFSETS_X,
'type': 'category'
},
'yaxis': {
'title': 'Correlation',
'range': [-1, 1],
'zeroline': False,
'tickmode': 'array',
'tickvals': [-1, -0.5, 0, 0.5, 1]
},
'boxmode': 'group',
'legend': {
'xanchor': 'right',
'yanchor': 'top',
'bordercolor': 'black',
'borderwidth': 2
}
}
)
save.to_html(
_fig=_fig,
_path=os.path.join(paths.PLOTS, save.get_module_name()),
_filename='plot'
)
def listToBox(values, n):
#TODO set colour based on Firefox/Tor Original/Ublock?
return go.Box(y=values, name=n,
boxpoints=False) #,jitter=0.3,pointpos=-1.8)
def update_main_graph(parameter, dimension, trace):
if parameter == 'age':
lake = lake_age
sea = sea_age
elif parameter == 'education':
lake = lake_edu
sea = sea_edu
else:
lake = lake_gen
sea = sea_gen
if dimension == 'agreeableness':
column = "A"
elif dimension == 'conscientiosness':
column = "C"
elif dimension == 'extraversion':
column = "E"
elif dimension == 'neuroticism':
column = "N"
else:
column = "O"
trace_ocean = go.Box(
name = "Respondents",
x = sea[parameter],
y = sea[column],
)
trace_lake = go.Scatter(
name = "Mean",
x = lake[parameter],
y = lake[column],
)
trace_progress = go.Scatter(
name = "Progress",
x = [lake.loc[0, parameter], lake.loc[len(lake.index) - 1, parameter]],
y = [lake.loc[0, column], lake.loc[len(lake.index) - 1, column]],
)
data = []
if 'mean' in trace:
data.append(trace_lake)
if 'res' in trace:
data.append(trace_ocean)
if 'prog' in trace:
data.append(trace_progress)
layout = go.Layout(
xaxis=dict(
title=parameter.title(),
showgrid=True,
zeroline=True,
zerolinecolor='#969696',
zerolinewidth=4
),
yaxis=dict(
title=dimension.title(),
range=[1, 6],
showgrid=True,
gridcolor='#bdbdbd',
gridwidth=1
)
)
figure = go.Figure(data = data, layout = layout)
return figure
