Home

Introduntion to Matplotlib

Rajesh Majumder

April 7, 2023

Matplotlib is a plotting library for the Python programming language and its numerical mathematics extension NumPy. It provides an object-oriented API for embedding plots into applications using general-purpose GUI toolkits like Tkinter, wxPython, Qt, or GTK+.

Some of the major Pros of Matplotlib are:

• Generally easy to get started for simple plots.
• Support for custom labels and texts
• Great control of easy element in a figure
• High-quality output in many formats
• Very customizable in general

Installation of Matplotlib

Install it using this command:(write the command on the command python command prompt/ ANACONDA prompt)

pip install matplotlib

or

conda install matplotlib

Import Matplotlib

Once Matplotlib is installed, import it in your applications by adding the __import__ keyword:

Checking Matplotlib Version

The version string is stored under __version__ attribute.

import matplotlib as mat
print(mat.__version__)

3.5.2

Most of the Matplotlib utilities lies under the pyplot submodule, and are usually imported under the plt alias:

import matplotlib.pyplot as plt 

Example: Draw a line in a diagram from position (0,0) to position (6,250):

import numpy as np

xpoints = np.array([0, 6])      ## Data
ypoints = np.array([0, 250])

plt.plot(xpoints, ypoints)     ## Plotting Function

plt.show()                    ## To show the plot

How to display our plots ?

Note that, To show a plot, plot.show() function is used. But, there is an alternative :

• If we are using Matplotlib from within a script, the function plt.show() is our friend. plt.show() starts an event loop, looks for all currently active figure objects, and opens one or more interactive windows that display your figure or figures.

• If we are using Jupyter NoteBook, we can use the %matplotlib magic command after starting Notebook. At this point, any plt plot command will cause a figure window to open, and further commands can be run to update the plot. plt.show() in Matplotlib mode is not required.

Saing Figures to File

To save a figure in PNG, or JPEG we use the function savefig()

xpoints = np.array([0, 6])      ## Data
ypoints = np.array([0, 250])

plt.plot(xpoints, ypoints)     ## Plotting Function

plt.savefig("Test.jpeg")

Scatter Plot

x= np.arange(start=0,stop=10)    ##  np.arange() => Return evenly spaced values within a given interval
print("x:",x,"\n")
y= np.arange(start=11,stop=21)
print("y:",y,"\n")

x: [0 1 2 3 4 5 6 7 8 9] 

y: [11 12 13 14 15 16 17 18 19 20] 

(1) Plotting Scatter plot

plt.scatter(x,y)

<matplotlib.collections.PathCollection at 0x1fb5bbadc10>

(2) Changing the Shape of the points

plt.scatter(x,y,s=3)  # s=> shape parameter

<matplotlib.collections.PathCollection at 0x1fb5b82b430>

(3) Changing the color of the points

plt.scatter(x,y,c='r')  # c=> color parameter

<matplotlib.collections.PathCollection at 0x1fb5b8d57c0>

(4) Adding Axis Labels and Title

plt.scatter(x,y,c="black",s=6)
plt.xlabel("X axis")
plt.ylabel("Y axis")
plt.title("My Scatter Plot")

Text(0.5, 1.0, 'My Scatter Plot')

(5) Add labels to points

plt.annotate(<label>, <coords>) using the value itself as label:

plt.scatter(x,y,c="r",s=6)
plt.xlabel("X axis")
plt.ylabel("Y axis")
plt.title("My Scatter Plot")

## Creating Labels:
labs=[]
for i in range(len(x)):
    labs.append('('+str(x[i])+','+str(y[i])+')') ## Levels of the points
    plt.annotate(labs[i], # text
                 (x[i],y[i]), # coordinates to position the label
                textcoords="offset points", # how to position the text
                xytext=(0,10), # distance from text to points (x,y),
                ha='center') # horizontal alignment can be left, right or center

(6) Remove Axis Labels

plt.scatter(x,y,c="black",s=6)
plt.xlabel("X axis")
plt.ylabel("Y axis")
plt.xticks(ticks=[])   ## Removing X-axis ticks/labels
plt.title("My Scatter Plot")

Text(0.5, 1.0, 'My Scatter Plot')

plt.scatter(x,y,c="black",s=6)
plt.xlabel("X axis")
plt.ylabel("Y axis")
plt.yticks(ticks=[])   ## Removing Y-axis ticks/labels
plt.title("My Scatter Plot")

Text(0.5, 1.0, 'My Scatter Plot')

(7) Multiple Plots

## 1st Plot:
plt.subplot(1,2,1)  ## Creating ploting region. 1st plot of the  (1x2) ploting region 
plt.scatter(x,y)
plt.xlabel("X axis")
plt.ylabel("Y axis")
plt.title("Without point labels")
 
# 2nd Plot:
plt.subplot(1,2,2) ## Creating ploting region. 1st plot of the  (1x2) ploting region 
plt.scatter(x,y,c="r",s=6)
plt.xlabel("X axis")
plt.title("With point labels")
labs=[]
for i in range(len(x)):
    labs.append('('+str(x[i])+','+str(y[i])+')')
    plt.annotate(labs[i],(x[i],y[i]),textcoords="offset points",xytext=(0,6),ha='center',fontsize=6)

## Adjustment of subplot layout
plt.subplots_adjust(top=0.92, bottom=0.08, left=0.10, right=0.95, hspace=0.25,
                    wspace=0.35)

Line Diagram

(1) Plotting Line diagram

x= np.linspace(0,20,1000)
y1= np.sin(x)
y2= np.cos(x)

plt.subplot(2,1,1)
plt.plot(x,y1)
plt.xticks(ticks=[]) ## Remove the x-axis ticks and labels.
plt.title("sin(x)")

#Plotting 2nd line
plt.subplot(2,1,2)
plt.plot(x,y2)
plt.title("cos(x)")

Text(0.5, 1.0, 'cos(x)')

(2) Changing Color, line type, line width

plt.plot(x,y1,'g',linestyle="dashed",linewidth=2)  ## alternatively, linestyle="--"

## See the help for more details

[<matplotlib.lines.Line2D at 0x1fb5d00af70>]

(3) Multiple lines one same scale

plt.plot(x,y1,'-g',label='sin(x)')
plt.plot(x,y2,'--r',label='cos(x)')

[<matplotlib.lines.Line2D at 0x1fb636e4e20>]

(4) Add legend

plt.plot(x,y1,'-g',label='sin(x)')
plt.plot(x,y2,'--r',label='cos(x)')
plt.axis('equal')
plt.legend()

<matplotlib.legend.Legend at 0x1fb6240ba90>

(5) Changing axes limits

plt.plot(x,y1)
plt.title("sin(x)")
plt.xlabel("x")
plt.xlim(0,13)  ## Changing the x limit from [0,20]-> [0,13]

(0.0, 13.0)

Bar plot

## Creating Data
import pandas as pd
Country_data= pd.DataFrame(np.random.choice(["India","Australia","USA","UK","Germany","China","Russia"],size=100,replace=True))
Country_data= pd.DataFrame({'Country':list(set(Country_data[0])),'Freq':list(Country_data.value_counts())},index=None)
Country_data

	Country	Freq
0	USA	20
1	Russia	17
2	Australia	16
3	Germany	16
4	India	13
5	UK	10
6	China	8

(1) Plotting Bar diagram

plt.bar(Country_data['Country'],Country_data['Freq'])
plt.xlabel("Country Name")
plt.ylabel("Frequency")
plt.title("Country Data")

Text(0.5, 1.0, 'Country Data')

(2) Changing color and bin width

plt.bar(x=Country_data['Country'],height= Country_data['Freq'],width=0.4,color ='orange')
plt.xlabel("Country Name")
plt.ylabel("Frequency")
plt.title("Country Data")

Text(0.5, 1.0, 'Country Data')

(3) Horizontal Bar diagram

plt.barh(y=Country_data['Country'],width= Country_data['Freq'],height=0.4,color ='orange')
plt.xlabel("Frequency")
plt.ylabel("Country Name")
plt.title("Country Data")

Text(0.5, 1.0, 'Country Data')

(4) Add Text to the bars.

## 1st Plot:
plt.subplot(1,2,2)
plt.bar(x=Country_data['Country'],height= Country_data['Freq'],width=0.4,color ='orange')
plt.xlabel("Country Name")
plt.ylabel("Frequency")
plt.title("Country Data")
x_coordinate= np.arange(0,len(Country_data["Country"].values))
y_coordinate= Country_data["Freq"].values
labs=[]
for i in range(Country_data.shape[0]):
    labs.append(str(Country_data["Freq"].values[i]))
    plt.annotate(labs[i],(x_coordinate[i],y_coordinate[i]),textcoords="offset points",xytext=(0,3),ha='center',fontsize=6)


## 2nd Plot:
plt.subplot(1,2,1)
plt.barh(y=Country_data['Country'],width= Country_data['Freq'],height=0.4,color ='maroon')
plt.xlabel("Frequency")
plt.ylabel("Country Name")
plt.title("Country Data")
y_coordinate= np.arange(0,len(Country_data["Country"].values))
x_coordinate= Country_data["Freq"].values+0.5
labs=[]
for i in range(Country_data.shape[0]):
    labs.append(str(Country_data["Freq"].values[i]))
    plt.annotate(labs[i],(x_coordinate[i],y_coordinate[i]),textcoords="offset points",xytext=(0,3),ha='center',fontsize=6)


## Adjustment of subplot layout
plt.subplots_adjust(top=0.92, bottom=0.08, left=0.10, right=0.95, hspace=0.25,
                    wspace=0.35)

(5) Multiple Bar plots

Creating dataframe

df = pd.DataFrame({
    'Name': ['Arnab','Mainak','Kunal','Sourav'],
    'Age': [45,38,60,89],
    'Height(in cm)': [150, 180, 160, 170]
})

df

	Name	Age	Height(in cm)
0	Arnab	45	150
1	Mainak	38	180
2	Kunal	60	160
3	Sourav	89	170

Method 1: Providing multiple columns in y parameter

# plotting graph
df.plot(x="Name", y=["Age", "Height(in cm)"], kind="bar")

<AxesSubplot:xlabel='Name'>

Method 2: By plotting on the same axis

# plotting Height
ax = df.plot(x="Name", y="Height(in cm)", kind="bar")
# plotting age on the same axis
df.plot(x="Name", y="Age", kind="bar", ax=ax, color="maroon")

<AxesSubplot:xlabel='Name'>

Boxplot

plt.boxplot(x=np.random.normal(0,1,100))

{'whiskers': [<matplotlib.lines.Line2D at 0x1fb6e898df0>,
  <matplotlib.lines.Line2D at 0x1fb6e8a6100>],
 'caps': [<matplotlib.lines.Line2D at 0x1fb6e8a63d0>,
  <matplotlib.lines.Line2D at 0x1fb6e8a66a0>],
 'boxes': [<matplotlib.lines.Line2D at 0x1fb6e898c40>],
 'medians': [<matplotlib.lines.Line2D at 0x1fb6e8a6970>],
 'fliers': [<matplotlib.lines.Line2D at 0x1fb6e8a6c40>],
 'means': []}

Add color

plt.boxplot(x=np.random.normal(0,1,100),patch_artist=True)
plt.show()

Multiple Boxplot

np.random.seed(10)  
  
dataSet1 = np.random.normal(100, 10, 220)  
dataSet2 = np.random.normal(80, 20, 200)  
dataSet3 = np.random.normal(60, 35, 220)  
dataSet4 = np.random.normal(50, 40, 200)  
dataSet = [dataSet1, dataSet2, dataSet3, dataSet4]  
  

plt.boxplot(dataSet)  
plt.show()
  

Rotatig the axis

np.random.seed(10)  
  
dataSet1 = np.random.normal(100, 10, 220)  
dataSet2 = np.random.normal(80, 20, 200)  
dataSet3 = np.random.normal(60, 35, 220)  
dataSet4 = np.random.normal(50, 40, 200)  
dataSet = [dataSet1, dataSet2, dataSet3, dataSet4]  
  

plt.boxplot(dataSet,vert=False,patch_artist=True)  
plt.show()

Some Customuization in boxplot

np.random.seed(10)  
dataSet1 = np.random.normal(100, 10, 220)  
dataSet2 = np.random.normal(80, 20, 200)  
dataSet3 = np.random.normal(60, 35, 220)  
dataSet4 = np.random.normal(50, 40, 200)  
dataSet = [dataSet1, dataSet2, dataSet3, dataSet4]  
  
figure = plt.figure(figsize =(10, 7))  
ax = figure.add_subplot(111)  
bp = ax.boxplot(dataSet, patch_artist = True,notch ='True', vert = 0)  
colors = ['#00FF00','#0F00FF', '#F00FF0','#FFFF0F']  
for patch, color in zip(bp['boxes'], colors):  
    patch.set_facecolor(color)  
for whisker in bp['whiskers']:  
    whisker.set(color ='#8E008B',linewidth = 1.4,linestyle =":")  
for cap in bp['caps']:  
    cap.set(color ='#8E008B',linewidth = 2.1)  
for median in bp['medians']:  
    median.set(color ='blue',linewidth = 3)  
for flier in bp['fliers']:  
    flier.set(marker ='D',color ='#d7298c',alpha = 0.6)   
ax.set_yticklabels(['dataSet1', 'dataSet2','dataSet3', 'dataSet4'])  
plt.title("Customized box plot using attributes")  
ax.get_xaxis().tick_bottom()  
ax.get_yaxis().tick_left()    

Pie Chart

## Preparing Data
plt.pie(x=Country_data['Freq'],labels=Country_data['Country'])
plt.show()

Add Legend

plt.figure(figsize=(20,6))
plt.pie(x=Country_data['Freq'],labels=Country_data['Country'])
plt.legend()
plt.show()

Histogram

## Preparing Data
X= np.random.normal(loc=0,scale=1,size=1000)

## Histogram
plt.hist(X,color="maroon")
plt.show()

Overlapping two Histograms

## Data 1: X1 From N(0,1)
X1= np.random.normal(loc=0,scale=1,size=1000)

## Data 2: X2 From Log Normal(0,1)
X2= np.random.lognormal(mean=0,sigma=1,size=1000)

## Plotting Histograms:

plt.hist(X1,label="Normal Distribution")

plt.hist(X2,label="Log Normal Distribution")

plt.legend(loc="upper right")

plt.title(label=" My Histograms")
plt.show()

Transparency

## Plotting Histograms:

plt.hist(X1,label="Normal Distribution",alpha=0.5)

plt.hist(X2,label="Log Normal Distribution",alpha=0.5)

plt.legend(loc="upper right")

plt.title(label=" My Histograms")
plt.show()

Error Bars

## Preparing Data
Mean= [3.5,4.6,5.1]
se= [0.05,0.005,0.6]
y=[1,2,3]

plt.errorbar(Mean,y,xerr=se,fmt=".k",)

<ErrorbarContainer object of 3 artists>