Quick start to Pandas

Pre-requisite : Numpy

What is Pandas ?

• Pandas is a third party package which has to be installed explicitly
• Built on top of Numpy package
• Used in data analysis
• Feature loaded compared to numpy
• Open source and free
• Can be downloaded from either Python Package Index or from Conda environment

Pandas has basically two types of data structures. They are Series and data frames

Series :

• series are one dimensional array
• Build’s an index array to store the indices and values array to store the values
• Internally uses a numpy array
• Value mutable but not size mutable

eg:

import pandas as pd
s1 = pd.Series([10, 20, 30, 40, 50])
print(s1)
print(f"shape = {s1.shape}")
print(f"size = {s1.size}")
print(f"dimensions = {s1.ndim}")
print(f"data type = {s1.dtype}")
print(id(s1))
print(id(s1[0]))
print(type(s1[0]))

output:

0    10
1    20
2    30
3    40
4    50
dtype: int64
shape = (5,)
size = 5
dimensions = 1
data type = int64
1751034379184
1751154510320
<class 'numpy.int64'>

series internal structure

Data Frame :

• Multi dimensional array
• Creates a kind of ordered list to store the data in table like structure
• Stored in the form of rows and columns
• every column represents a series

eg:

import pandas as pd
patients = [
    {"name": "p1", "bp": 80, "temperature": 37, "infected": 1},
    {"name": "p2", "bp": 50, "temperature": 33, "infected": 0},
    {"name": "p3", "bp": 100, "temperature": 34, "infected": 1},
    {"name": "p4", "bp": 75, "temperature": 35, "infected": 0}
]

df = pd.DataFrame(patients)
print(type(df['name']))
print(df.describe())
print('-' * 50)
print(df.info())

output:

<class 'pandas.core.series.Series'>
               bp  temperature  infected
count    4.000000     4.000000   4.00000
mean    76.250000    34.750000   0.50000
std     20.564938     1.707825   0.57735
min     50.000000    33.000000   0.00000
25%     68.750000    33.750000   0.00000
50%     77.500000    34.500000   0.50000
75%     85.000000    35.500000   1.00000
max    100.000000    37.000000   1.00000
--------------------------------------------------
<class 'pandas.core.frame.DataFrame'>
RangeIndex: 4 entries, 0 to 3
Data columns (total 4 columns):
 #   Column       Non-Null Count  Dtype 
---  ------       --------------  ----- 
 0   name         4 non-null      object
 1   bp           4 non-null      int64 
 2   temperature  4 non-null      int64 
 3   infected     4 non-null      int64 
dtypes: int64(3), object(1)
memory usage: 256.0+ bytes
None

Data frames internal structure

Data frame comes with many attributes like:

• size
• ndmin
• index
• values
• shape

Data frame also comes with various functions like:

• head()
• tail()
• describe()
• info()
• reshape()

I hope this blog really helped you in understanding the basics of Pandas

Thank you for reading this blog. If you find any corrections please post them in the comments section

Credits goes to: Amit Kulkarini