Data Analysis and Visualization in Python?

Python provides numerous libraries for data analysis and visualization mainly numpy, pandas, matplotlib, seaborn etc. In this section, we are going to discuss pandas library for data analysis and visualization which is an open source library built on top of numpy.

It allows us to do fast analysis and data cleaning and preparation.Pandas also provides numerous built-in visualization feautures which we are going to see below.

Installation

To install pandas, run the below command in your terminal −

pipinstall pandas

Orwe have anaconda, you can use

condainstall pandas

Pandas-DataFrames

Data framesa re the main tools when we are working with pandas.

code −

import numpy as np
import pandas as pd
from numpy.random import randn
np.random.seed(50)
df = pd.DataFrame(randn(6,4), ['a','b','c','d','e','f'],['w','x','y','z'])
df

Output

Pandas-Missing Data

Weare going to see some convenient ways to deal with missing data inpandas, which automatically gets filled with zero's or nan.

import numpy as np
import pandas as pd
from numpy.random import randn
d = {'A': [1,2,np.nan], 'B': [9, np.nan, np.nan], 'C': [1,4,9]}
df = pd.DataFrame(d)
df

Output

So,we are having 3 missing value in above.

df.dropna()

df.dropna(axis = 1)

df.dropna(thresh = 2)

df.fillna(value = df.mean())

Pandas − Import data

We are going to read the csv file which is either stored in our local machine(in my case) or we can directly fetch from the web.

#import pandas library
import pandas as pd

#Read csv file and assigned it to dataframe variable
df = pd.read_csv("SYB61_T03_Population Growth Rates in Urban areas and Capital cities.csv",encoding = "ISO-8859-1")

#Read first five element from the dataframe
df.head()

Output

Toread the number of rows and columns in our dataframe or csv file.

#Countthe number of rows and columns in our dataframe.
df.shape

Output

(4166,9)

Pandas − Dataframe Math

Operationson dataframes can be done using various tools of pandas forstatistics

#To computes various summary statistics, excluding NaN values
df.describe()

Output

# computes numerical data ranks
df.rank()

Output

.....

.....

Pandas − plot graph

import matplotlib.pyplot as plt
years = [1981, 1991, 2001, 2011, 2016]

Average_populations = [716493000, 891910000, 1071374000, 1197658000, 1273986000]

plt.plot(years, Average_populations)
plt.title("Census of India: sample registration system")
plt.xlabel("Year")
plt.ylabel("Average_populations")
plt.show()

Output

Scatter plot of above data:

plt.scatter(years,Average_populations)

Histogram:

import matplotlib.pyplot as plt

Average_populations = [716493000, 891910000, 1071374000, 1197658000, 1273986000]

plt.hist(Average_populations, bins = 10)
plt.xlabel("Average_populations")
plt.ylabel("Frequency")

plt.show()

Output