def sankey_diagram():
sankey = Sankey()
sankey.add(flows=[2, -4.5, 0.5, 1, 1],
orientations=[0, 0, 1, -1, 1],
labels=['input1', 'output', 'input2', 'input3', 'input4'],
rotation=0)
sankey.finish()
def plotSankeyFlow(df, col_1="1", col_2="2"):
from matplotlib.sankey import Sankey
fig = plt.figure(figsize=(8, 6))
ax = fig.add_subplot(111, title="")
ax.axis('off')
df = pd.DataFrame({
"flow": [242, -121, -121],
"length": [2, 1, 1],
"label": ["1", "2", "3"],
"orientation": [0, 0, 1]
})
df1 = pd.DataFrame({
"flow": [121, -65, -17, -11, -8, -8, -8, -1, -1, -2],
"path": [2, 2, 2, 2, 2, 2, 2, 2, 2, 2],
"label": ["1", "2", "3", "4", "5", "6", "7", "8", "9", "0"],
"length": [2, 2, 2, 2, 2, 2, 2, 2, 2, 2],
"orientation": [0, 0, 1, -1, 1, -1, 1, -1, 1, -1]
})
sankey = Sankey(ax=ax, scale=0.02, offset=0.3)
sankey.add(flows=df['flow'],
pathlengths=df['length'],
labels=list(df['label']),
label='1',
orientations=df['orientation'])
sankey.add(flows=df1['flow'],
pathlengths=df1['length'],
labels=list(df1['label']),
label='2',
orientations=df1['orientation'],
fc='#FFFF00',
prior=0,
connect=(1, 0))
diagrams = sankey.finish()
plt.legend(loc='lower right')
plt.show()
def test_format_using_callable():
# test using callable by slightly incrementing above label example
def show_three_decimal_places(value):
return f'{value:.3f}'
s = Sankey(flows=[0.25],
labels=['First'],
orientations=[-1],
format=show_three_decimal_places)
assert s.diagrams[0].texts[0].get_text() == 'First\n0.250'
def createSankey(flows=None, labels=None, orientations=None):
'''create simple sankey diagram
default example:
flows=[0.25, 0.15, 0.60, -0.20, -0.15, -0.05, -0.50, -0.10],
labels=['', '', '', 'First', 'Second', 'Third', 'Fourth', 'Fifth'],
orientations=[-1, 1, 0, 1, 1, 1, 0,-1]'''
if flows == None:
flows = [0.25, 0.15, 0.60, -0.20, -0.15, -0.05, -0.50, -0.10]
if labels == None:
labels = ['', '', '', 'First', 'Second', 'Third', 'Fourth', 'Fifth']
if orientations == None: orientations = [-1, 1, 0, 1, 1, 1, 0, -1]
Sankey(flows=flows, labels=labels, orientations=orientations).finish()
plt.title("Sankey diagram with default settings")
plt.show()
def displayGraph ( applicationDict ):
REJECT_STATUS = -1
DEFAULT_STATUS = 0
INTERVIEW_STATUS = 1
numReject = 0
numGhost = 0
numInterviews = 0
totalNumApplications = 0
statisticsList = []
for company, applicationList in applicationDict.items ( ):
for position in applicationList:
if position.getJobStatus() == REJECT_STATUS:
numReject += 1
elif position.getJobStatus() == DEFAULT_STATUS:
numGhost += 1
elif position.getJobStatus ( ) == INTERVIEW_STATUS:
numInterviews += 1
totalNumApplications = numReject + numGhost + numInterviews
statisticsList.append(-numReject)
statisticsList.append(-numGhost)
statisticsList.append (-numInterviews )
statisticsList.append(totalNumApplications)
print ('\nOpening flowchart display in a new window.' )
fig = plt.figure()
subPlot = fig.add_subplot(1,1,1,xticks=[],yticks=[],
title='Sankey Diagram - Employment Search 2021')
myChart = Sankey(ax=subPlot, scale = 0.1, offset=0.25,head_angle=180,
format='%.0f', gap = 0.6, radius = 0.3, shoulder = .05,
margin = 0.5, unit=' Job applications')
myChart.add(flows= statisticsList,
labels=['Rejected', 'No Response', 'Interviewed', 'Total: '],
orientations=[0,0,0,0], pathlengths=[0.5,0.5,0.5,0.5],
trunklength=1.1, facecolor = 'r')
diagrams = myChart.finish()
diagrams[0].texts[-1].set_color('b')
diagrams[0].texts[-1].set_fontweight('bold')
diagrams[0].text.set_fontweight('bold')
plt.show()
def sankey_total_sales_by_region(df):
# Calculate sales
total_sales = sum(df['Sales'])
south_sales = df.loc[dataset['Region'] == 'South', 'Sales'].sum()
east_sales = df.loc[dataset['Region'] == 'East', 'Sales'].sum()
west_sales = df.loc[dataset['Region'] == 'West', 'Sales'].sum()
central_sales = df.loc[dataset['Region'] == 'Central', 'Sales'].sum()
# Plot Sankey
Sankey(
flows=[south_sales, east_sales, west_sales, central_sales, -total_sales],
labels=['South', 'East', 'West', 'Central', 'Total'],
orientations=[0, 0, 1, -1, 0],
# fill = False
).finish()
plt.title("Total Sales by Region, in $1000")
plt.show()
def sankey_total_sales_by_category_in_region_in_year(df,region,year,colour):
total_sales = df.loc[((df['Region'] == region) & (df['order_year'] == year)), 'Sales'].sum()
f_sales = df.loc[((df['Region'] == region) & (df['order_year'] == year)& (df['Category'] == 'Furniture')), 'Sales'].sum()
o_sales = df.loc[((df['Region'] == region) & (df['order_year'] == year)& (df['Category'] == 'Office Supplies')), 'Sales'].sum()
t_sales = df.loc[((df['Region'] == region) & (df['order_year'] == year)& (df['Category'] == 'Technology')), 'Sales'].sum()
# Sankey Diagram of sales by category in Region in Year
Sankey(
flows=[f_sales, o_sales, t_sales, -total_sales], # input values & output value with leading minus
labels=['Furniture', 'Office Supplies', 'Technology', 'Total'], # labels for all values
unit = '$', # units of I/O
orientations=[-1, 1, 0, 0],
facecolor = colour,
head_angle = 120,
scale=0.25,
rotation=-90, # diagram rotation
).finish()
plt.title(str(year)+' :Total Sales by Category in '+region)
plt.show()
def sankey_total_sales_by_region_by_year_in_K(df,year):
# Calculate sales
total_sales = df.loc[df['order_year'] == year, 'Sales'].sum()
south_sales = df.loc[((df['Region'] == 'South') & (df['order_year'] == year)), 'Sales'].sum()
east_sales = df.loc[((df['Region'] == 'East') & (df['order_year'] == year)), 'Sales'].sum()
west_sales = df.loc[((df['Region'] == 'West') & (df['order_year'] == year)), 'Sales'].sum()
central_sales = df.loc[((df['Region'] == 'Central') & (df['order_year'] == year)), 'Sales'].sum()
# reduce to sales in thousands for cleaner plotting
total_sales_K = int(total_sales/1000)
south_sales_K = int(south_sales/1000)
west_sales_K = int(west_sales/1000)
east_sales_K = int(east_sales/1000)
central_sales_K = int(central_sales/1000)
# Sankey Diagram of sales by Region in Year
Sankey(
flows=[south_sales_K, east_sales_K, west_sales_K, central_sales_K, -total_sales_K],
labels=['South', 'East', 'West', 'Central', 'Total'],
orientations=[0, 0, 1, -1, 0]
).finish()
plt.title(year+' :Total Sales by Region, in $1000')
plt.show()
import matplotlib.pyplot as plt
from matplotlib.sankey import Sankey
from matplotlib import pyplot
from datos import data
d=data('mtcars')
mpg=d.loc["Datsun 710",'mpg']
disp=d.loc["Datsun 710",'disp']
qsec=d.loc["Datsun 710",'qsec']
gear =d.loc["Datsun 710",'gear']
cyl=d.loc["Datsun 710",'cyl']
hp=d.loc["Datsun 710",'hp']
fig, ax = plt.subplots(figsize=(15, 15))
Sankey(ax,margin=10, flows=[disp, mpg,-qsec, -gear, -hp,-cyl,],
labels=['Displacement\n', 'Miles/(US) gallo', '1/4 Mile Time',
'Number of forward gears', 'Gross horsepower', 'Number of
cylinders'],
orientations=[-1, 0, 1, 0, 0, 0]).finish()
ax.axes.get_xaxis().set_visible(False)
ax.axes.get_yaxis().set_visible(False)
plt.title("Datsun 710 Sankey Diagram")
plt.show()
import numpy as np
import matplotlib.pyplot as plt
from matplotlib.sankey import Sankey
# basic sankey chart
Sankey(flows=[0.25, 0.15, 0.60, -0.20, -0.15, -0.05, -0.50, -0.10],
labels=['', '', '', 'First', 'Second', 'Third', 'Fourth', 'Fifth'],
orientations=[-1, 1, 0, 1, 1, 1, 0, -1]).finish()
plt.title("Sankey diagram with default settings")
source = [n + 1]
target = list(range(8))
input_vals = A.row(n)
w_min = np.min(input_vals)
w = np.log(input_vals) - np.log(w_min) + 0.5
w_T = np.sum(w)
values = [w_T]
for i in range(len(w)):
values.append(-w[i])
labels = ['BB']
for i in range(len(label)):
labels.append(label[i])
Sankey(flows=values,
labels=labels,
offset=3,
margin=8,
gap=4,
shoulder=1,
head_angle=130,
unit=None,
trunklength=30.0,
facecolor='lightgreen',
edgecolor='lightgreen',
alpha=0.8,
orientations=[0, 1, 1, 1, 1, 0, -1, -1, -1]).finish()
plt.title("Logarithmic Sankey Diagram of Credit Migration Rates")
plt.savefig("sankey.png")
plt.show()
# Matius Celcius Sinaga
# Ubuntu 14.0 | 32 bit
# Python 2.7
# matplotlib-1.4.0
import numpy as np
import matplotlib.pyplot as plt
from matplotlib.sankey import Sankey
Sankey(flows=[0.25, 0.15, 0.60, -0.20, -0.15, -0.05, -0.50, -0.10],
labels=['', '', '', 'Pertama', 'Kedua', 'Ketiga', 'Keempat', 'Kelima'],
orientations=[-1, 1, 0, 1, 1, 1, 0, -1]).finish()
plt.title("PEngaturan secara umum untuk menghasilkan diagram seperti ini.")
fig = plt.figure()
ax = fig.add_subplot(1,
1,
1,
xticks=[],
yticks=[],
title="Diagram alur pada layar")
sankey = Sankey(ax=ax,
scale=0.01,
offset=0.2,
head_angle=180,
format='%.0f',
unit='%')
sankey.add(flows=[25, 0, 60, -10, -20, -5, -15, -10, -40],
labels=[
'', '', '', 'Pertama', 'Kedua', 'Ketiga', 'Keempat', 'Kelima',
'Ohh.. yeah!'
],
fig = plt.figure(figsize=(8, 9))
ax = fig.add_subplot(1,
1,
1,
xticks=[],
yticks=[],
title="Rankine Power Cycle: Example 8.6 from Moran and "
"Shapiro\n\x22Fundamentals of Engineering Thermodynamics "
"\x22, 6th ed., 2008")
Hdot = [
260.431, 35.078, 180.794, 221.115, 22.700, 142.361, 10.193, 10.210, 43.670,
44.312, 68.631, 10.758, 10.758, 0.017, 0.642, 232.121, 44.559, 100.613,
132.168
] # MW
sankey = Sankey(ax=ax, format='%.3G', unit=' MW', gap=0.5, scale=1.0 / Hdot[0])
sankey.add(patchlabel='\n\nPump 1',
rotation=90,
facecolor='#37c959',
flows=[Hdot[13], Hdot[6], -Hdot[7]],
labels=['Shaft power', '', None],
pathlengths=[0.4, 0.883, 0.25],
orientations=[1, -1, 0])
sankey.add(patchlabel='\n\nOpen\nheater',
facecolor='#37c959',
flows=[Hdot[11], Hdot[7], Hdot[4], -Hdot[8]],
labels=[None, '', None, None],
pathlengths=[0.25, 0.25, 1.93, 0.25],
orientations=[1, 0, -1, 0],
prior=0,
connect=(2, 1))
def plot_sankey(self):
fig = plt.figure(figsize=(15, 10))
ax = fig.add_subplot(1,
1,
1,
xticks=[],
yticks=[],
title="Flow Diagram of N")
Nfluxes = self.dict_sankey
sankey = Sankey(ax=ax,
scale=1. / Nfluxes['total_DIN'],
offset=0.1,
head_angle=140,
format='%.0f',
unit='')
#prior 0
sankey.add(patchlabel='DIN',
facecolor='lightgreen',
rotation=180,
trunklength=0.25,
flows=[
Nfluxes['total_DIN'], -Nfluxes['Psm_DIN_uptake'],
-Nfluxes['Pmd_DIN_uptake'], -Nfluxes['Plg_DIN_uptake']
],
orientations=[0, -1, -1, -1],
pathlengths=[0.25, 0.5, 0.26, 0.52],
labels=['DIN', None, None, None])
#prior 1
sankey.add(patchlabel='Psm N',
facecolor='lightgreen',
rotation=180,
trunklength=0.4,
flows=[
Nfluxes['Psm_DIN_uptake'], -Nfluxes['Psm_exu_loss'],
-Nfluxes['Psm_agg_loss'], -Nfluxes['Zsm_Psm_graz']
],
orientations=[1, 1, 1, 0],
labels=['Psm\nDIN\nuptake', 'exu', 'agg', None],
pathlengths=[0.25, 0.1, 0.7, 0.5],
prior=0,
connect=(1, 0))
#prior 2
sankey.add(patchlabel='Zsm N',
facecolor='lightblue',
trunklength=0.4,
flows=[
Nfluxes['Zsm_Psm_graz'], Nfluxes['Zsm_bact_graz'],
-Nfluxes['Zsm_DOM_loss'], -Nfluxes['Zmd_Zsm_graz']
],
orientations=[0, 1, 1, -1],
labels=['Zsm_jprod', 'bact', 'DOM', ''],
pathlengths=[0.1, 0.5, 0.1, 0.12],
prior=1,
connect=(3, 0))
#prior 3
sankey.add(patchlabel='Pmd N',
facecolor='lightgreen',
rotation=180,
trunklength=1.0,
flows=[
Nfluxes['Pmd_DIN_uptake'], -Nfluxes['Pmd_exu_loss'],
-Nfluxes['Pmd_agg_loss'], -Nfluxes['Zmd_Pmd_graz']
],
orientations=[1, 1, 1, 0],
labels=['Pmd\nDIN\nuptake', 'exu', 'agg', None],
pathlengths=[0.55, 0.1, 1.35, 0.50],
prior=0,
connect=(2, 0))
#prior 4
sankey.add(patchlabel='Zmd N',
facecolor='lightblue',
trunklength=.6,
flows=[
Nfluxes['Zmd_Pmd_graz'], Nfluxes['Zmd_Zsm_graz'],
-Nfluxes['Zmd_DOM_loss'], -Nfluxes['Zmd_det_loss'],
-Nfluxes['Zlg_Zmd_graz'], -Nfluxes['jhploss_Zmd']
],
orientations=[0, 1, 1, 1, -1, -1],
labels=['Zmd_jprod', '', 'DOM', 'det', '', 'hp'],
pathlengths=[0.4, 0.1, 0.1, 1.0, 0.1, 1.0],
prior=2,
connect=(3, 1))
#prior 5
sankey.add(patchlabel='Plg N',
facecolor='lightgreen',
rotation=180,
trunklength=1.85,
flows=[
Nfluxes['Plg_DIN_uptake'], -Nfluxes['Plg_exu_loss'],
-Nfluxes['Plg_agg_loss'], -Nfluxes['Zlg_Plg_graz']
],
orientations=[1, 1, 1, 0],
labels=['Plg\nDIN\nuptake', 'exu', 'agg', None],
pathlengths=[0.55, 0.1, 2.2, 0.52],
prior=0,
connect=(3, 0))
#prior 6
sankey.add(patchlabel='Zlg N',
facecolor='lightblue',
trunklength=1.,
flows=[
Nfluxes['Zlg_Plg_graz'], Nfluxes['Zlg_Zmd_graz'],
-Nfluxes['Zlg_DOM_loss'], -Nfluxes['Zlg_det_loss'],
-Nfluxes['jhploss_Zlg']
],
orientations=[0, 1, 1, 1, -1],
labels=['Zmd_jprod', '', 'DOM', 'det', 'hp'],
pathlengths=[1.25, 0.1, 0.1, 1.8, 0.5],
prior=4,
connect=(4, 1))
sankey.finish()
# fig.text(0.60,0.75, 'NPP: {:d}'.format(Nfluxes['NPP']), fontsize=12, fontweight='bold')
# fig.text(0.22,0.55, 'mesozooP: {:d}'.format(Nfluxes['mesozooP']), fontsize=12, fontweight='bold')
# fig.text(0.25,0.75, 'HPP: {:d}'.format(Nfluxes['hpp']), fontsize=12, fontweight='bold')
# fig.text(0.30,0.25, 'Benthic N flux: {:d}'.format(Nfluxes['det_btf']), fontsize=12, fontweight='bold')
# fig.text(0.30,0.85, 'Pelagic N flux: {:d}'.format(Nfluxes['det_btf']), fontsize=12, fontweight='bold')
# fig.text(0.80,0.87, 'pe-ratio: {:4.2f}'.format(Nfluxes['pe-ratio']), fontsize=12)
# fig.text(0.80,0.84, 'z-ratio: {:6.2f}'.format(Nfluxes['z-ratio']), fontsize=12)
# fig.text(0.80,0.81, 'f-ratio: {:6.2f}'.format(Nfluxes['f-ratio']), fontsize=12)
plt.show()
return None
# -*- coding: utf-8 -*-
"""
Created on Thu Feb 25 20:47:03 2016
@author: Paul
"""
import numpy as np
import matplotlib.pyplot as plt
from matplotlib.sankey import Sankey
fig = plt.figure()
ax = fig.add_subplot(1, 1, 1, xticks=[], yticks=[],
title="Flow Diagram of Food Waste")
sankey = Sankey(ax=ax, scale=0.01, offset=0.2, head_angle=180,
format='%.0f', unit='mT')
sankey.add(flows=[22, 19, -3, -3.9, -0.92, -7, -26],
labels=['Production', 'Import', 'Farm Waste', 'Processing Waste', 'HaFS Waste', 'Consumer Waste',
'Consumer Use'],
orientations=[0, 1, -1, -1, -1, -1, 0],
pathlengths=[0.25, 0.25, 0.25, 0.5, 0.25, 0.7,
0.25],
patchlabel="Food Production\nand Waste",
alpha=0.2, lw=2.0) # Arguments to matplotlib.patches.PathPatch()
diagrams = sankey.finish()
diagrams[0].patch.set_facecolor('#37c959')
diagrams[0].texts[-1].set_color('b')
diagrams[0].text.set_fontweight('bold')
diagrams[0].texts[4].set_color('r')
diagrams[0].text.set_fontweight('bold')
# -*- coding:utf-8 -*-
import matplotlib.pyplot as plt
import matplotlib as mpl
import numpy as np
from matplotlib.sankey import Sankey
mpl.rcParams["font.sans-serif"]=["FangSong"]
mpl.rcParams["axes.unicode_minus"]=False
flows=[0.2,0.1,0.4,0.3,-0.6,-0.05,-0.15,-0.2]
labels=["","","","","family","trip","education","sport"]
orientations=[1,1,0,-1,1,-1,1,0]
sankey = Sankey()
sankey.add(flows=flows,
labels=labels,
orientations=orientations,
color="c",
fc="lightgreen",
patchlabel="Life Cost",
alpha=0.7)
diagrams = sankey.finish()
diagrams[0].texts[4].set_color("r")
diagrams[0].texts[4].set_weight("bold")
diagrams[0].text.set_fontsize(20)
diagrams[0].text.set_fontweight("bold")
def main(args):
print "Loading dataset..."
dsA = nc.Dataset(args.inputfile)
print '(A): ', args.inputfile
year = args.year
month = args.month
cht = args.cohort
numpfts = len(dsA.dimensions['PFTS'])
# needs to have one value per pft
litterfall = dsA.variables['LTRFALC'][cht,year*month,:]
# needs to be a single value (for whole soil pool)
rh = dsA.variables['RH'][0,year*month]
# one value per pft
gpp = dsA.variables['GPP'][cht, year*month,:]
# one value per pft
npp = dsA.variables['NPP'][cht, year*month, :]
ra = (npp-gpp)
pfts = []
for i in range(numpfts):
pfts.append({'gpp': gpp[i],
'ltfl': litterfall[i],
'Ra': ra[i],
})
for pft in pfts:
print pft
fig = plt.figure(figsize=(18,6))
ax = fig.add_subplot(1, 1, 1, xticks=[], yticks=[], title="Veg and Soil Fluxes")
sankey = Sankey(ax=ax,
tolerance=1e-23,
#scale=1.00,
offset=-1.75,
head_angle=130,
gap=3.0,
radius=0.5,
unit='gC/m2',
format='%.0f',
)
soil_flows = np.append(litterfall, rh)
# [gpp1, gpp2, ... rh, ]
print "litterfall: ", litterfall
print "rh: ", rh
print "gpp: ", gpp
print "ra: ", ra
print "soil_flows: ", soil_flows
# make each pft a bit longer
pathlens = [1+(numpfts-i) for i,x in enumerate(gpp)]
pathlens.append(0.5) # make the departing rh flow shorter
# make all the pft flows come in from top
orients = [1 for i in range(len(pathlens))]
orients[-1] = 0 # change the lasts one (rh) to a side flow
labels = ['pft%s'%i for i, v in enumerate(soil_flows)]
labels[-1] = 'RH'
sankey.add(patchlabel="Soil",
flows=soil_flows,
fc='brown', label='soil',
labels=labels,
orientations=orients,
#pathlengths=1.0,
pathlengths=pathlens,
rotation=0,
trunklength=7.0,
)
print "[PFT ]: gpp ltrfl ra"
for pft, val in enumerate(gpp):
print "[PFT%s]: %.05f %.05f %.05f"%(pft, gpp[pft], -1.0*litterfall[pft], ra[pft])
print "-------"
# pft = 1
# print gpp[pft]
# print -1.0*litterfall[pft]
# print ra[pft]
#
# sankey.add(flows=[gpp[pft], -1.0*litterfall[pft], ra[pft]],
# fc=(0.0, 0.8, 0.0), # rgb
# label='pft%s' % (pft),
# labels=['gpp%s'%pft, 'lf%s'%pft, 'ra%s'%pft,],
# orientations=[-1,1,0], # 1(top), 0(l/r), -1(bottom)
# # seems to be in relation to prev. diagram...
# pathlengths=[1,pft*10,1],
# trunklength=12.0,
# prior=0, connect=(pft,1)
# )
# VEGETATION FLOWS
for pft, val in enumerate(gpp):
if (-1.0*litterfall[pft] < 0) or (-1.0*litterfall[pft] > 0):
sankey.add(flows=[gpp[pft], -1.0*litterfall[pft], ra[pft]],
fc=(0.0, 0.8, 0.0), # rgb
label='pft%s' % (pft),
labels=['gpp%s'%pft, 'lf%s'%pft, 'ra%s'%pft,],
orientations=[-1,1,-1], # 1(top), 0(l/r), -1(bottom)
# seems to be in relation to prev. diagram...
#pathlengths=[1,pft+10,1],
trunklength=12.0,
prior=0, connect=(pft,1)
)
else:
pass
# sankey.add(flows=[gpp[pft], -1.0*litterfall[pft], ra[pft]],
# fc=(0.0, 0.8, 0.0), # rgb
# label='pft%s' % (pft),
# labels=['gpp%s'%pft, 'lf%s'%pft, 'ra%s'%pft,],
# orientations=[-1,1,-1], # 1(top), 0(l/r), -1(bottom)
# # seems to be in relation to prev. diagram...
# pathlengths=[1,6,1],
# #trunklength=2.0,
# #prior=0, connect=(pft,1)
# )
diagrams = sankey.finish()
#print diagrams
#diagrams[-1].patch.set_hatch('/')
plt.legend(loc='best')
plt.title("The default settings produce a diagram like this.")
wrapup(args)
def sankey_drawings():
""" Draw sanky diagrams for the figures """
from matplotlib import pyplot as plt
from matplotlib.sankey import Sankey
fig = plt.figure(figsize=(12, 8))
ax = fig.add_subplot(
1,
1,
1,
xticks=[],
yticks=[],
title=
"Statistics from the 2nd edition of\nfrom Audio Signal Processing for Music Applications by Stanford University\nand Universitat Pompeu Fabra of Barcelona on Coursera (Jan. 2016)"
)
learners = [14460, 9720, 7047, 3059, 2149, 351]
labels = [
"Total learners joined", "Learners that visited the course",
"Learners that watched a lecture", "Learners that browsed the forums",
"Learners that submitted an exercise",
"Learners that obtained a grade >70%\n(got a Statement of Accomplishment)"
]
colors = ["#FF0000", "#FF4000", "#FF8000", "#FFBF00", "#FFFF00"]
sankey = Sankey(ax=ax, scale=0.0015, offset=0.3)
for input_learner, output_learner, label, prior, color in zip(
learners[:-1], learners[1:], labels, [None, 0, 1, 2, 3], colors):
if prior != 3:
sankey.add(flows=[
input_learner, -output_learner, output_learner - input_learner
],
orientations=[0, 0, 1],
patchlabel=label,
labels=['', None, 'quit'],
prior=prior,
connect=(1, 0),
pathlengths=[0, 0, 2],
trunklength=10.,
rotation=0,
facecolor=color)
else:
sankey.add(flows=[
input_learner, -output_learner, output_learner - input_learner
],
orientations=[0, 0, 1],
patchlabel=label,
labels=['', labels[-1], 'quit'],
prior=prior,
connect=(1, 0),
pathlengths=[0, 0, 10],
trunklength=10.,
rotation=0,
facecolor=color)
diagrams = sankey.finish()
for diagram in diagrams:
diagram.text.set_fontweight('bold')
diagram.text.set_fontsize('10')
for text in diagram.texts:
text.set_fontsize('10')
ylim = plt.ylim()
plt.ylim(ylim[0] * 1.05, ylim[1])
import numpy as np
import matplotlib.pyplot as plt
from matplotlib.sankey import Sankey
fig = plt.figure()
ax = fig.add_subplot(1,
1,
1,
xticks=[],
yticks=[],
title="Veg and Soil Fluxes")
sankey = Sankey(
ax=ax,
unit=None,
tolerance=1e-6,
)
# SOIL INPUT FLOWS
# pft1 pft2 pft3
litterfall = [
0.15,
0.01,
0.10,
]
# SOIL OUTPUT FLOWS
rh = [-0.65]
soil_flows = np.append(litterfall, rh)
import numpy as np
import matplotlib.pyplot as plt
from matplotlib.sankey import Sankey
Sankey(flows=[0.50, -0.20, -0.15, -0.05, -0.50, -0.10],
labels=['Input', 'First', 'Second', 'Third', 'Fourth', 'Fifth'],
orientations=[0, -1, 1, -1, 1, 0]).finish()
plt.title("The default settings produce a diagram like this.")
plt.show()
#https://matplotlib.org/gallery/api/sankey_basics.html#sphx-glr-gallery-api-sankey-basics-py
'd': np.random.randn(50)}
data['b'] = data['a'] + 10 * np.random.randn(50)
data['d'] = np.abs(data['d']) * 100
plt.scatter('a', 'b', c='c', s='d', data=data) # c=color s = marker size
plt.xlabel('entry a')
plt.ylabel('entry b')
plt.show()
#sankey diagram (household budget)
from matplotlib.sankey import Sankey
fig = plt.figure(figsize = (15,8))
ax = fig.add_subplot(1, 1, 1, xticks=[], yticks=[],
title="Household Budget")
sankey = Sankey(ax=ax, scale=.1, offset=1, unit='%')
sankey.add(flows=[100, -50, -30, -20],
labels=['household budget', 'necessities', 'fun',
'saving'],
orientations=[0, 0, 1, -1],
trunklength = 10,
edgecolor = '#027368',
facecolor = '#027368')
sankey.add(flows=[50, -30, -10, -10],
labels=['','rent', 'groceries', 'other'],
trunklength = 2,
pathlengths = [3,3,3,3],
orientations=[0, 1, 0, -1],
prior=0, #which sankey are you connecting to (0-indexed)
connect=(1, 0), #flow number to connect: (prior, this)
edgecolor = '#58A4B0',
def draw_chart(self):
"""Plots chart from self.attributes"""
if not hasattr(self, "attributes"):
return
# The first element is always axes data
self._setup_axes(self.attributes[0])
for attribute in self.attributes[1:]:
series = copy(attribute)
# Extract chart type
chart_type_string = series.pop("type")
x_str, y_str = self.plot_type_xy_mapping[chart_type_string]
# Check xdata length
if x_str in series and \
len(series[x_str]) != len(series[y_str]):
# Wrong length --> ignore xdata
series[x_str] = range(len(series[y_str]))
else:
# Solve the problem that the series data may contain utf-8 data
series_list = list(series[x_str])
series_unicode_list = []
for ele in series_list:
if isinstance(ele, types.StringType):
try:
series_unicode_list.append(ele.decode('utf-8'))
except Exception:
series_unicode_list.append(ele)
else:
series_unicode_list.append(ele)
series[x_str] = tuple(series_unicode_list)
fixed_attrs = []
if chart_type_string in self.plot_type_fixed_attrs:
for attr in self.plot_type_fixed_attrs[chart_type_string]:
# Remove attr if it is a fixed (non-kwd) attr
# If a fixed attr is missing, insert a dummy
try:
fixed_attrs.append(tuple(series.pop(attr)))
except KeyError:
fixed_attrs.append(())
# Remove contour chart label info from series
cl_attrs = {}
for contour_label_attr in self.contour_label_attrs:
if contour_label_attr in series:
cl_attrs[self.contour_label_attrs[contour_label_attr]] = \
series.pop(contour_label_attr)
# Remove contourf attributes from series
cf_attrs = {}
for contourf_attr in self.contourf_attrs:
if contourf_attr in series:
cf_attrs[self.contourf_attrs[contourf_attr]] = \
series.pop(contourf_attr)
if not fixed_attrs or all(fixed_attrs):
# Draw series to axes
# Do we have a Sankey plot --> build it
if chart_type_string == "Sankey":
Sankey(self.__axes, **series).finish()
else:
chart_method = getattr(self.__axes, chart_type_string)
plot = chart_method(*fixed_attrs, **series)
# Do we have a filled contour?
try:
if cf_attrs.pop("contour_fill"):
cf_attrs.update(series)
if "linewidths" in cf_attrs:
cf_attrs.pop("linewidths")
if "linestyles" in cf_attrs:
cf_attrs.pop("linestyles")
if not cf_attrs["hatches"]:
cf_attrs.pop("hatches")
self.__axes.contourf(plot, **cf_attrs)
except KeyError:
pass
# Do we have a contour chart label?
try:
if cl_attrs.pop("contour_labels"):
self.__axes.clabel(plot, **cl_attrs)
except KeyError:
pass
# The legend has to be set up after all series are drawn
self._setup_legend(self.attributes[0])
Demonstrate the Sankey class by producing three basic diagrams.
"""
import matplotlib.pyplot as plt
from matplotlib.sankey import Sankey
###############################################################################
# Example 1 -- Mostly defaults
#
# This demonstrates how to create a simple diagram by implicitly calling the
# Sankey.add() method and by appending finish() to the call to the class.
Sankey(flows=[0.25, 0.15, 0.60, -0.20, -0.15, -0.05, -0.50, -0.10],
labels=['', '', '', 'First', 'Second', 'Third', 'Fourth', 'Fifth'],
orientations=[-1, 1, 0, 1, 1, 1, 0, -1]).finish()
plt.title("The default settings produce a diagram like this.")
###############################################################################
# Notice:
#
# 1. Axes weren't provided when Sankey() was instantiated, so they were
# created automatically.
# 2. The scale argument wasn't necessary since the data was already
# normalized.
# 3. By default, the lengths of the paths are justified.
###############################################################################
# Example 2
# Set the flow diagram Geometry
tketrun = 3
# Common Snake properties
if len(sys.argv) > 2:
tkelabel = sys.argv[3]
else:
tkelabel = "Stratified Turbulence"
#
rot = 0
fcol = "#DFDFDF"
#
snakey = Sankey(ax=ax,
format='%4.0f',
unit=r'$\mathrm{nW/kg}$',
margin=.5,
offset=-.5,
radius=.125,
gap=.125,
scale=10.0 / maxflow,
head_angle=120)
snakey.add(flows=tkeflows,
orientations=tkeori,
pathlengths=tkelen,
labels=tkelabs,
rotation=rot,
trunklength=tketrun,
facecolor=fcol,
label=tkelabel)
#
diagrams = snakey.finish()
for diagram in diagrams:
# Notice:
# 1. Since the sum of the flows is nonzero, the width of the trunk isn't
# uniform. If verbose.level is helpful (in matplotlibrc), a message is
# given in the terminal window.
# 2. The second flow doesn't appear because its value is zero. Again, if
# verbose.level is helpful, a message is given in the terminal window.
# Example 3
# This demonstrates:
# 1. Connecting two systems
# 2. Turning off the labels of the quantities
# 3. Adding a legend
fig = plt.figure()
ax = fig.add_subplot(1, 1, 1, xticks=[], yticks=[], title="Flow of Energy")
flows = [0.85, -0.40, -0.30, -0.15]
sankey = Sankey(ax=ax, unit=None)
sankey.add(
flows=flows,
labels=['', 'Metabolize', ' Activity', '\n\n\n\nGrowth \nand recovery'],
label='Metabolize',
orientations=[-1, 1, 1, 0])
sankey.add(flows=[-0.85, 1, -0.15],
label='Predation',
labels=['Net absorption', 'Energy provided \n by food', 'Unused'],
orientations=[1, 0, 0],
prior=0,
connect=(0, 0))
diagrams = sankey.finish()
#diagrams[-1].patch.set_hatch('/')
plt.legend(loc='best')
# Notice that only one connection is specified, but the systems form a
out_vals
# In[19]:
#create a sankey diagram that shows the 6 largest inflows and outflows for the given station
fig = plt.figure(figsize=(13, 9))
ax = fig.add_subplot(
1,
1,
1,
xticks=[],
yticks=[],
title=
"Largest Sources and Destinations For Divvy Bikes at Chicago Union Station (2018)"
)
sankey = Sankey(ax=ax, unit=None, shoulder=0.1, scale=1, offset=0.35, gap=0.9)
sankey.add(
flows=[
in1, in2, in3, in4, in5, in6, in_rest, out1, out2, out3, out4, out5,
out6, out_rest
],
labels=[
'Michigan & \n Washington \n (6.9%)',
'Columbus & \n Randolph \n (3.6%)', 'Kingsbury & \n Kinzie \n (3.5%)',
'State & \n Randolph \n (3.1%)', 'Michigan & \n Lake \n (3.0%)',
'Larrabee & \n Kingsbury \n (2.7%)', '',
'Michigan & \n Washington \n (6.4%)',
'Larrabee & \n Kingsbury \n (3.2%)',
'Columbus & \n Randolph \n (2.9%)', 'Kingsbury & \n Kinzie \n (2.9%)',
'State & \n Randolph \n (2.1%)', 'State & \n Kinzie \n (2.0%)', ''
],
def test_sankey():
# lets just create a sankey instance and check the code runs
sankey = Sankey()
sankey.add()
# In[42]:
labels = ['', 'Did Not Apply', 'Applied but\nDid Not\nBecome Member' , 'Became\nMember']
labels2 = ['', 'Applied but\nDid Not\nBecome Member', 'Did Not Apply', 'Became\nMember']
plt.close()
fig = plt.figure(figsize=(7,7.2), dpi=600, frameon=False)
ax = fig.add_subplot(2, 1, 1, xticks=[], yticks=[], frameon=False,)
ax.set_title("Flow Diagram of MuscleHub Visitors",y = 1.1, fontweight='bold')
ax2 = fig.add_subplot(2,1,2, xticks=[], yticks=[], frameon=False,)
sankey = Sankey(ax=ax, scale=0.0005, offset=0.65, head_angle=120,
format = '%d', unit=' visitors', gap=1,
radius=0.05, shoulder=0.05,)
sankey.add(flows = a_flow,
labels = labels,
orientations= [0, 1, 1, 0],
pathlengths=[0, 1, 2.5, 1],
trunklength=5.,
facecolor=tableau20[0],
label="Fitness Test"
).finish()
sankey2 = Sankey(ax=ax2, scale=0.0005, offset=0.65, head_angle=120,
format = '%d', unit=' visitors', gap=1,
radius=0.04, shoulder=0.05,)
sankey2.add(flows = b_flow,
labels = labels2,
def qc_sankey_diagram_speed(self,
stations='all',
wind_speed='vw10m(m/s)',
scale=10):
from matplotlib.sankey import Sankey
import matplotlib.pylab as pl
time_series = self.p.observation.time_series
if stations == 'all':
scale = 0.0000001
else:
time_series = time_series[time_series["name"].isin(stations)]
scale = 0.0000001 * scale
fig = plt.figure()
ax = fig.add_subplot(1, 1, 1, xticks=[], yticks=[], title=None)
sankey = Sankey(ax=ax,
scale=scale,
offset=0.2,
head_angle=135,
format="%i",
unit=' ')
all_speeds = time_series[wind_speed].count()
qc_1 = time_series[wind_speed][
time_series["qc_1_speed"] != 1].dropna().count()
qc_2 = time_series[wind_speed][(time_series["qc_1_speed"] == 1) & (
time_series["qc_2_speed"] != 1)].dropna().count()
qc_3 = time_series[wind_speed][
(time_series["qc_1_speed"] == 1) & (time_series["qc_2_speed"] == 1)
& (time_series["qc_3_speed"] != 1)].dropna().count()
qc_5 = time_series[wind_speed][
(time_series["qc_1_speed"] == 1) & (time_series["qc_2_speed"] == 1)
& (time_series["qc_3_speed"] == 1) &
(time_series["qc_5_speed"] != 1)].dropna().count()
qc_6 = time_series[wind_speed][
(time_series["qc_1_speed"] == 1) & (time_series["qc_2_speed"] == 1)
& (time_series["qc_3_speed"] == 1) &
(time_series["qc_5_speed"] == 1) &
(time_series["qc_6_speed"] != 1)].dropna().count()
colors = pl.cm.cividis(np.linspace(0, 1, 5))
result_1 = (all_speeds - qc_1)
sankey.add(flows=[all_speeds, -qc_1, -result_1],
labels=['All data', 'Unphysical values', None],
orientations=[0, -1, 0],
facecolor=colors[0])
# Arguments to matplotlib.patches.PathPatch
result_2 = result_1 - qc_2
sankey.add(flows=[result_1, -qc_2, -result_2],
labels=[None, "Excessive miss", None],
orientations=[0, 1, 0],
prior=0,
connect=(2, 0),
facecolor=colors[1])
result_3 = result_2 - qc_3
sankey.add(flows=[result_2, -qc_3, -result_3],
labels=[None, "Constant sequences", None],
orientations=[0, -1, 0],
prior=1,
connect=(2, 0),
facecolor=colors[2])
result_4 = result_3 - qc_5
sankey.add(flows=[result_3, -qc_5, -result_4],
labels=[None, "High variability", None],
orientations=[0, 1, 0],
prior=2,
connect=(2, 0),
facecolor=colors[3])
result_5 = result_4 - qc_6
sankey.add(flows=[result_4, -qc_6, -result_5],
labels=[None, "Bias", "Valid observations"],
orientations=[0, -1, 0],
prior=3,
connect=(2, 0),
facecolor=colors[4])
diagrams = sankey.finish()
diagrams[0].texts[-1].set_color('r')
diagrams[0].text.set_fontweight('bold')
def group_composition_sankey(names_and_networks,
project_name,
kind='years',
directory=plots_dir):
flows = compute_developer_flows(names_and_networks, project_name)
years = python_years if project_name == 'python' else debian_years
labels = years
color_map = cm.get_cmap('Dark2', len(years))
colors = [color_map(i) for i in range(len(years))]
# Plot
fig = plt.figure(figsize=(20, 14))
title = "Python developer mobility in the top connectivity level"
ax = fig.add_subplot(1, 1, 1, xticks=[], yticks=[], title=title)
sankey = Sankey(ax=ax, scale=0.35, offset=1)
last = None
previous_id = 0
#for i, flow in enumerate(flow for flow in flows if any(flow)):
for i, flow in enumerate(flows):
if not any(flow):
continue
if last is None: # First group
sankey.add(flows=flow,
fc=colors[i],
label=str(labels[i]),
orientations=[0, 0])
last = flow
previous_id += 1
else:
sankey.add(
flows=flow,
fc=colors[i],
label=str(labels[i]),
orientations=[0, 1, -1, 0],
labels=[None, '', '', ''],
#pathlengths=[0.5, 1.75, 1.75, 0.5],
pathlengths=[1, 2, 2, 1],
prior=previous_id - 1,
connect=(3, 0) if len(last) == 4 else (1, 0))
last = flow
previous_id += 1
diagrams = sankey.finish()
for diagram in diagrams:
diagram.text.set_fontweight('bold')
#diagram.text.set_fontsize('small')
#for text in diagram.texts:
# text.set_fontsize('small')
#leg = plt.legend(loc='best')
leg = ax.legend(bbox_to_anchor=(1, 0.1),
ncol=8,
fancybox=True,
shadow=True)
#for t in leg.get_texts():
# t.set_fontsize('small')
if directory is None:
plt.ion()
plt.show()
else:
fname = os.path.join(
directory, 'sankey_mobility_{}_{}'.format(project_name, kind))
plt.savefig("{}.eps".format(fname))
plt.savefig("{}.png".format(fname))
plt.close()
