Data Viz Advanced
2021, May 05
Advanced Data viz operation, including heatmap, diverging bars, dumbbel plot, funnel, waffle charts, time series and many others.
In [1]:
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
import numpy as np
In [2]:
large = 22; med = 16; small = 12
params = {'axes.titlesize': large,
'legend.fontsize': med,
'figure.figsize': (14, 5),
'axes.labelsize': med,
'axes.titlesize': med,
'xtick.labelsize': med,
'ytick.labelsize': med,
'figure.titlesize': large}
plt.rcParams.update(params)
Correlogram and Heatmap¶
In [3]:
cars = pd.read_csv("https://github.com/selva86/datasets/raw/master/mtcars.csv")
fig = plt.figure(figsize=(10,10))
mask = np.zeros_like(cars.corr())
mask[np.triu_indices_from(mask)] = True
sns.heatmap(data=cars.corr(),
xticklabels=cars.corr().columns,
yticklabels=cars.corr().columns,
cmap='gist_stern',
center=0,
annot=True,
cbar_kws={"orientation": "vertical"}, # vertical or horizontal
vmin=-1, vmax= 1,
linewidths=.5, # space between each square
mask = mask) # diagonal matrix!
plt.show()
Diverging bars, remove legend sns¶
In [4]:
def show_values_on_bars(axs, h_v="v", space_x=0.12,space_y=0.1, fontdict=None):
def _show_on_single_plot(ax):
if h_v == "v":
for p in ax.patches:
_x = p.get_x() + p.get_width() / 2 + space_x
_y = p.get_y() + p.get_height() + (space_y)
if not np.isnan(p.get_width()):
value = round(p.get_height(),2)
if fontdict: ax.text(_x, _y, value, ha="left", fontdict=fontdict)
else: ax.text(_x, _y, value, ha="left")
elif h_v == "h":
for p in ax.patches:
try:
_x = p.get_x() + p.get_width() + space_x
_y = p.get_y() + p.get_height() + space_y
if not np.isnan(p.get_width()):
value = round(p.get_width(),2)
if value < 0: _x-=0.27
if fontdict: ax.text(_x, _y, value, ha="left", fontdict=fontdict)
else: ax.text(_x, _y, value, ha="left")
except: print(f'Error while preparing {str(p)}')
if isinstance(axs, np.ndarray):
for idx, ax in np.ndenumerate(axs):
_show_on_single_plot(ax)
else:
_show_on_single_plot(axs)
In [5]:
cars = pd.read_csv("https://github.com/selva86/datasets/raw/master/mtcars.csv")
plt.figure(figsize=(12,12))
cars = cars.fillna(0)
# create column with distance in standard deviation
xs = cars['mpg']
cars['dist'] = (xs - xs.mean())/ xs.std() # (values - mean) / sd
cars.sort_values(by='dist', inplace=True, ascending=False)
# create colours
def cat(x):
if x < 0: return 'green'
else: return 'blue'
cars['col'] = cars['dist'].apply(cat)
g1 = sns.barplot(data = cars, y = "cars", x = "dist",
palette= cars.col)
g1.set_xlabel('$Mileage$', fontsize=25)
g1.set_ylabel('$Model$', fontsize=25)
show_values_on_bars(g1, 'h')
plt.title('Diverging Bars\n', fontsize=25)
plt.show()
Diverging dot-plot¶
In [6]:
def text_scatter(data, variable, orient='v', space_orient=0.02, space_support=.0, space_orient_n=-0.05, fontdict=None):
copy = data.copy()
copy.sort_values(by= variable, ascending=True)
copy.reset_index(inplace=True)
for i in range(len(copy)):
if orient == 'v': plt.text( x= i + space_support ,
y= copy[variable][i] + space_orient,
s = round(copy[variable][i],1),
fontdict = fontdict)
else: plt.text(y= i + space_support ,
x= copy[variable][i] + space_orient if copy[variable][i]> 0 else copy[variable][i] + space_orient_n,
s = round(copy[variable][i],1),
fontdict = fontdict)
In [7]:
plt.figure(figsize=(12,14))
cars.reset_index(drop=True, inplace=True) # for the iteration!
g1 = sns.scatterplot(data = cars, y = "cars", x = "dist",
hue= cars.col,
palette = {'blue':'red', 'green':'green'},
s=700).legend_.remove()
text_scatter(data=cars, variable='dist', orient='h', space_support= 0.2, space_orient= -0.06,
space_orient_n=-0.08, fontdict=dict(color='white',size=11, weight='bold'))
plt.hlines(y = cars.cars, xmin=cars.dist.min(), xmax=cars.dist.max(), linestyles='dashdot')
plt.title("Divergint Dot Plot")
plt.show()
Lollipop with HLINES¶
In [8]:
plt.figure(figsize=(12,12))
cars.sort_values(by='dist', inplace=True, ascending=True)
higligted_point = [300 if i != 'Fiat 128' else 500 for i in cars.cars]
colors = ['red' if i != 'Fiat 128' else 'gold' for i in cars.cars]
plt.scatter(data = cars, y = "cars", x = "dist",
color=colors,
s=higligted_point)
plt.hlines(y = cars.cars,
xmin=0,
xmax = cars.dist) #horizontal lines, not available in sns.
plt.show()
Dumbbel Plot¶
In [9]:
df = pd.read_csv("https://raw.githubusercontent.com/selva86/datasets/master/health.csv")
df.sort_values(by= 'pct_2014', inplace=True)
fig = plt.figure(figsize=(10,10))
g1= sns.scatterplot(data=df, y = df.Area, x = df.pct_2013, s = 100, color='lightblue', label='2013')
g1 = sns.scatterplot(data=df, y = df.Area, x = df.pct_2014, s = 100, color='royalblue', label='2014')
for row in range(len(df)):
plt.hlines(y = df.Area, xmin = df.pct_2013, xmax=df.pct_2014, color='lightblue')
g1.set_xticks([0, 0.1,0.2,0.3,0.5])
g1.set_xticklabels(['0%', '10%', "20%", "30%", '50%'])
g1.set_xlabel("Poverty")
plt.show()
Funnel¶
In [10]:
funnel = pd.read_csv("https://raw.githubusercontent.com/selva86/datasets/master/email_campaign_funnel.csv")
plt.figure(figsize=(10,10), dpi= 80)
g1 = sns.barplot(data= funnel[funnel.Gender == 'Male'], y= 'Stage', x='Users', color='lightblue')
g2 = sns.barplot(data= funnel[funnel.Gender == 'Female'], y= 'Stage', x='Users', color='pink')
g1.invert_yaxis()
plt.show()
Waffle Chart¶
In [11]:
from pywaffle import Waffle
data = {'Democratic': 48, 'Republican': 46, 'Libertarian': 3}
fig = plt.figure(
FigureClass=Waffle,
rows=10,
values=data,
colors=("#232066", "#983D3D", "#DCB732"),
legend={'loc': 'upper left', 'bbox_to_anchor': (1, 1)},
icons='beer', icon_size=30, # https://fontawesome.com/v5.15/icons?d=gallery&p=2&m=free
icon_legend=True
)
In [12]:
df = pd.read_csv("https://github.com/selva86/datasets/raw/master/mpg_ggplot2.csv")
df = df.groupby('class').size().reset_index(name='counts')
fig = plt.figure(
FigureClass=Waffle,
plots={
'111': {
'values': df['counts'],
'labels': ["{0} ({1})".format(n[0], n[1]) for n in df[['class', 'counts']].itertuples()],
'legend': {'loc': 'upper left', 'bbox_to_anchor': (1.05, 1), 'fontsize': 12},
'title': {'label': '# Vehicles by Class', 'loc': 'center', 'fontsize':18}
},
},
rows=7,
figsize=(10, 4)
)
c:\users\dell\venv\jupyter\lib\site-packages\pywaffle\waffle.py:394: MatplotlibDeprecationWarning: Passing non-integers as three-element position specification is deprecated since 3.3 and will be removed two minor releases later. self.ax = self.add_subplot(loc, aspect="equal")
Tree Map¶
In [13]:
import squarify
df_raw = pd.read_csv("https://github.com/selva86/datasets/raw/master/mpg_ggplot2.csv")
df = df_raw.groupby('class').size() # now class is the index of the new dataframe / series
df = df.reset_index(name='counts') # must assign again
labels = []
for row in range(len(df)): labels.append(str(df.loc[row, ['class']].values[0]) + " #" + str(df.loc[row, ['counts']].values[0]))
sizes = df.counts.tolist()
[plt.cm.Spectral(i/float(len(labels))) for i in range(len(labels))] # fancy color prof
np.random.seed(10)
colors = np.random.rand(7,3) # random tensor rgb 7x3
squarify.plot(sizes=sizes,
label=labels,
color=colors,
alpha=0.8)
plt.axis('off')
plt.show()
Calendar HeatMap¶
In [14]:
import calmap
df = pd.read_csv("https://raw.githubusercontent.com/selva86/datasets/master/yahoo.csv", parse_dates=['date'])
df.set_index('date', inplace=True)
# Plot
calmap.calendarplot(df.loc['2014','VIX.Close'],
fig_kws={'figsize': (15,12)},
yearlabel_kws={'color':'black', 'fontsize':10},
subplot_kws={'title':'Yahoo Stock Prices'},
cmap='inferno')
plt.yticks(fontsize=14)
plt.show()
Time Series¶
In [15]:
def peak_detection(data):
data = data.values
doublediff = np.diff(np.sign(np.diff(data)))
peak_locations = np.where(doublediff == -2)[0] + 1
doublediff2 = np.diff(np.sign(np.diff(-1*data)))
trough_locations = np.where(doublediff2 == -2)[0] + 1
return (peak_locations, trough_locations )
def peak_annotation(x, y, peak, low):
for t, p in zip(low[1::2], peak[::2]):
plt.text(x[p], y[p]+15, x[p].strftime("%m-%d"), horizontalalignment='center', color='darkgreen')
plt.text(x[t], y[t]-35,x[t].strftime("%m-%d"), horizontalalignment='center', color='darkred')
In [16]:
df = pd.read_csv('https://github.com/selva86/datasets/raw/master/AirPassengers.csv', parse_dates=['date'])
g1 = sns.lineplot(data=df, x='date', y='value', color='firebrick')
# if in the data frame are casted as time series seaborn will take care of that!
# custom ticks!
mths = pd.date_range('1948-01-01', periods=12, freq='Y')
g1.set_xticks(mths.strftime('%Y-%m-%d'))
g1.set(xticklabels=mths.strftime('%Y'))
peak, low = peak_detection(data=df.value)
peak_annotation (x= df.date, y = df.value, peak=peak, low=low )
plt.title('TimeSeries')
plt.show()
Multiple Time Series with two y Axes¶
In [17]:
df = pd.read_csv("https://github.com/selva86/datasets/raw/master/economics.csv", parse_dates=['date'])
plt.figure(figsize=(18,10), dpi=50)
plt.style.use('ggplot')
g1 = sns.lineplot(data=df, x='date', y = 'psavert', color='blue', linewidth = 1.5)
g1.set_xlabel('date', fontdict=dict(size=25))
g1.set_ylabel('psavert', fontdict=dict(size=25, color='blue'))
# plt.twinx() <-
g2 = plt.twinx()
sns.lineplot(ax=g2, data=df, x='date', y='unemploy', color='firebrick',linewidth =2)
g2.set_ylabel('unemploy', fontdict=dict(size=25, color='red'))
# color ticks and axes:
[t.set_color('blue') for t in g1.yaxis.get_ticklabels()]
[t.set_color('firebrick') for t in g2.yaxis.get_ticklabels()]
plt.title('Multiple time series', fontdict=dict(size=20))
plt.show()
Stacked Area chart¶
In [18]:
from matplotlib.dates import DateFormatter
df = pd.read_csv('https://raw.githubusercontent.com/selva86/datasets/master/nightvisitors.csv')
plt.figure(figsize=(12,7))
g1 = plt.stackplot(df.yearmon,
df.iloc[:,1], df.iloc[:,2], df.iloc[:,3], df.iloc[:,4],
df.iloc[:,5], df.iloc[:,6], df.iloc[:,7], df.iloc[:,8], labels= df.columns.values[1:])
plt.xticks(df.yearmon[::2], df.yearmon[::2],rotation=60)
plt.ylim(0,100000)
plt.legend( loc='best', mode = "expand",ncol=int(len(df.columns)/2))
plt.show()
In [19]:
df = pd.read_csv("https://github.com/selva86/datasets/raw/master/economics.csv", parse_dates=['date'])
plt.figure(figsize=(15,10))
plt.fill_between(x = df.date, y1 = df.psavert, y2 = 0,
label = 'psavert',
alpha=0.5) #must set alpha lower than 1 or they will overlap
plt.fill_between(x = df.date, y1 = df.uempmed, y2 = 0,
label = 'uempmed',
alpha=0.5)
years= df.date.dt.strftime("%Y").unique()[::2]
plt.xticks(years,years, rotation = 60, style='italic')
plt.legend()
plt.show()
