Loading [MathJax]/extensions/Safe.js

Pandas Plot

Load Data

import pandas as pd
import numpy as np

data = pd.read_csv('weatherAUS.csv')
data['Date']= pd.to_datetime(data['Date'])
data['Index'] = data['Date']
data = data.set_index('Index')

data.head()

	Date	Location	MinTemp	MaxTemp	Rainfall	Evaporation	Sunshine	WindGustDir	WindGustSpeed	WindDir9am	...	Humidity9am	Humidity3pm	Pressure9am	Pressure3pm	Cloud9am	Cloud3pm	Temp9am	Temp3pm	RainToday	RainTomorrow
Index
2008-12-01	2008-12-01	Albury	13.4	22.9	0.6	NaN	NaN	W	44.0	W	...	71.0	22.0	1007.7	1007.1	8.0	NaN	16.9	21.8	No	No
2008-12-02	2008-12-02	Albury	7.4	25.1	0.0	NaN	NaN	WNW	44.0	NNW	...	44.0	25.0	1010.6	1007.8	NaN	NaN	17.2	24.3	No	No
2008-12-03	2008-12-03	Albury	12.9	25.7	0.0	NaN	NaN	WSW	46.0	W	...	38.0	30.0	1007.6	1008.7	NaN	2.0	21.0	23.2	No	No
2008-12-04	2008-12-04	Albury	9.2	28.0	0.0	NaN	NaN	NE	24.0	SE	...	45.0	16.0	1017.6	1012.8	NaN	NaN	18.1	26.5	No	No
2008-12-05	2008-12-05	Albury	17.5	32.3	1.0	NaN	NaN	W	41.0	ENE	...	82.0	33.0	1010.8	1006.0	7.0	8.0	17.8	29.7	No	No

5 rows × 23 columns

Format

• Pandas plot returns a matplotlib AxesSubPlot
• Use matplotlib format functions to adjust plot format

ax = data.iloc[:1000, :].plot(x = 'Date', y = 'MinTemp') # matplotlib.axes._subplots.AxesSubplot
ax.set_xlabel('Date Label')
ax.set_ylabel('Temperature')

Text(0, 0.5, 'Temperature')

Line Plot

# Single plot for DataFrame
data.iloc[:1000, :].plot(x = 'Date', y = 'MinTemp')

<AxesSubplot: xlabel='Date'>

# Multiple plots for DataFrame
data.iloc[:1000, :].plot(x = 'Date', y = ['MinTemp', 'MaxTemp'])

<AxesSubplot: xlabel='Date'>

# Use index as x for Series
data.iloc[:1000, :]['MinTemp'].plot()

<AxesSubplot: xlabel='Index'>

# Multiple plots using index as x
data.iloc[:1000, :][['MinTemp', 'MaxTemp']].plot()

<AxesSubplot: xlabel='Index'>

Bar Chart

# Bar chart for Series
data_counts = data['WindDir9am'].value_counts() # Series
data_counts.plot.bar()

<AxesSubplot: >

# Bar chart for DataFrame
data_counts = data['WindDir9am'].value_counts().to_frame()
data_counts.reset_index(inplace=True)
data_counts.columns = ['Type', 'Count'] # DataFrame

data_counts.plot.bar(x = 'Type', y = 'Count', rot = 90)

<AxesSubplot: xlabel='Type'>

Histogram Plot

# Series hist
data['MinTemp'].plot.hist(bins=30)

<AxesSubplot: ylabel='Frequency'>

# DataFrame hist
data[['MinTemp', 'MaxTemp']].plot.hist(alpha=0.5, bins=30)

<AxesSubplot: ylabel='Frequency'>

# Stacked hist
data[['MinTemp', 'MaxTemp']].plot.hist(stacked=True, bins=30)

<AxesSubplot: ylabel='Frequency'>

# Group by
data.iloc[:6000, :].hist(column = 'MinTemp', by='Location')

array([<AxesSubplot: title={'center': 'Albury'}>,
       <AxesSubplot: title={'center': 'BadgerysCreek'}>], dtype=object)

Box Plot

# Box plot with outliers
# IQR = Q3 - Q1
# lower bound = (Q1 - 1.5 IQR)
# upper bound = (Q3 + 1.5 * IQR)
# points that fall outside the lower and upper bounds are labeled as outliers

color = dict(boxes='DarkGreen', whiskers='DarkOrange', medians='DarkBlue', caps='red')
data[['MinTemp', 'MaxTemp']].plot.box(color = color, sym='b+')

<AxesSubplot: >

Area Plot

# use index as x
data.iloc[:100, :][['Humidity9am', 'Humidity3pm']].plot.area(stacked=False)

<AxesSubplot: xlabel='Index'>

# use a specific column as x
data.iloc[:100, :].plot.area(x = 'Date', y = ['Humidity9am', 'Humidity3pm'], stacked=False)

<AxesSubplot: xlabel='Date'>

Scatter Plot

data.iloc[:100, :].plot.scatter(x = 'Date', y = 'MinTemp', rot=45)

<AxesSubplot: xlabel='Date', ylabel='MinTemp'>

ax = data.iloc[:100, :].plot.scatter(x = 'Date', y = 'MinTemp', label='Minimum Temperature', rot=45) data.iloc[:100, :].plot.scatter(x = 'Date', y = 'MaxTemp', color='DarkGreen', label='Maxminum Temperature', ax=ax, rot=45)

Hexagonal Bin Plot

data.iloc[:100, :].plot.hexbin(x = 'MinTemp', y = 'MaxTemp', gridsize=25)

<AxesSubplot: xlabel='MinTemp', ylabel='MaxTemp'>

Pie Plot

# Series
data['WindDir9am'].value_counts().plot.pie(figsize=(6, 6))

<AxesSubplot: ylabel='WindDir9am'>

Density Plot

data['MinTemp'].plot.kde()

<AxesSubplot: ylabel='Density'>

Andrews Curves

• Identify underlying structure in multidimensional data
• Map all features from a single observation or row of data onto a function

$$f_{x}(\theta) = x_{1}/\sqrt{2} + x_{2}sin(\theta) + x_{3}cos(\theta) + x_{4}sin(2\theta) + x_{5}cos(2\theta) + ...$$

from pandas.plotting import andrews_curves
data = pd.read_csv('Iris.csv')
data.head()

	Id	SepalLengthCm	SepalWidthCm	PetalLengthCm	PetalWidthCm	Species
0	1	5.1	3.5	1.4	0.2	Iris-setosa
1	2	4.9	3.0	1.4	0.2	Iris-setosa
2	3	4.7	3.2	1.3	0.2	Iris-setosa
3	4	4.6	3.1	1.5	0.2	Iris-setosa
4	5	5.0	3.6	1.4	0.2	Iris-setosa

andrews_curves(data[['MinTemp', 'Rainfall', 'WindGustSpeed', 'Humidity9am', 'RainTomorrow']], 'RainTomorrow')
#andrews_curves(data[['SepalLengthCm', 'SepalWidthCm', 'PetalLengthCm', 'PetalWidthCm', 'Species']], 'Species')

<AxesSubplot: >

Parallel Coordinates

from pandas.plotting import parallel_coordinates
parallel_coordinates(data[['MinTemp', 'Rainfall', 'WindGustSpeed', 'Humidity9am', 'RainTomorrow']], 'RainTomorrow')

<AxesSubplot: >

Lag Plot

• Used to check if a data set or time series is random
• Random data, no structure in plot
• Non-random data, exhibit a structure

from pandas.plotting import lag_plot
lag_plot(data['MinTemp'], lag=1)

<AxesSubplot: xlabel='y(t)', ylabel='y(t + 1)'>

Autocorrelation Plot

• Compute correlation coefficiencies for data values at varying time lags, lag vs Series.autocorr()
• Random, autocorrelations should be near zero
• Non-random, autocorrelations will be significantly non-zero

# Manually calculate correlation coefficiencies for each lag value
correlations = []
for i in range(6000):
    correlations.append(data['MinTemp'].autocorr(lag=i))
s = pd.Series(correlations)
ax = s.plot()
ax.set_xlabel('Lag')

Text(0.5, 0, 'Lag')

# Use autocorrelation plot
# If there is any missed value, autocorrelation plot does not plot
from pandas.plotting import autocorrelation_plot
ax = autocorrelation_plot(data.iloc[:100, :]['MinTemp'])

Bootstrap Plot

• Sample a random subset of a specified size from a data set, calculate mean, median, midrange, etc.
• Repeat the sampling many times to plot the distributions of mean, median, and midrange.

from pandas.plotting import bootstrap_plot

# size, number of data points in each sampling
# samples, number of times of sampling
bootstrap_plot(data['MinTemp'], size=50, samples=500)

RadViz

• A multivariate data visualization algorithm
• Plots each feature dimension uniformly around the circumference of a circle
• Each point normalizes its attributes on the axes from the center to each arc

from pandas.plotting import radviz
radviz(data[['MinTemp', 'Rainfall', 'WindGustSpeed', 'Humidity9am', 'RainTomorrow']], 'RainTomorrow')

<AxesSubplot: >

Reference

• Simple Introduction to Andrews Curves with Python
• Pandas Plot: Deep Dive Into Plotting Directly With Pandas
• Pandas API
• Tutorial