4.1. Checking datatypes

4.1.1. Data Types in Python,Pandas and Numpy

	Python	Pandas	Numpy
integer	int	int64	int_,int8,int16,int32,int64
float	float	float64	float_,float16,float32,float64
string	str	object	string_
boolean	bool	bool	bool_

4.1.2. Note

• No date datatype is present in python

• Extra Datatypes in Pandas:datetime64,timedelta[ns],category

• Extra Datatypes in Python:list,tupple,set,dictinary

# Dataframe for testing purposes
import pandas as pd
import numpy as np

df=pd.DataFrame({'Name':['Sahil', 'Sonia', 'Sourav', 'Vishal'],
        'Age':[20, 21, 19, 18]})

# Series for testing purposes
series1=pd.Series([1,2,3])

# Numpy Array for testing purposes
arr = np.array([1,2,3])

df

	Name	Age
0	Sahil	20
1	Sonia	21
2	Sourav	19
3	Vishal	18

series1

0    1
1    2
2    3
dtype: int64

arr

array([1, 2, 3])

4.1.3. Checking datatype

# For basic Python data types -> only type can be used
type('a') # <class 'str'>
type(1) # <class 'int'>
type(1.0) # <class 'float'>
type(1.9) # <class 'float'>
type([1,2,3]) # <class 'list'>
type([1,2,3,'sahil']) # <class 'list'>
type(True) # <class 'bool'>
type(('a','b')) # <class 'tuple'>
type({'a':'b'}) # <class 'dict'>


# For other things like Numpy,Pandas(Series,Dataframe)
    # We can use both type() and .dtype ( and .dtypes() also )
    # type(numpy arr | series | df) will tell the type of container
    # .dtype -> will tell us the datatype of elements inside it,which is useful
    # .dtypes -> will tell datatype of entire dataframe columns

# For numpy array
type(arr) # numpy.ndarray
arr.dtype # int32

# For Series
type(df.Name) # pandas.core.series.Series
type(df.Age) # pandas.core.series.Series
df.Name.dtype # object
df.Age.dtype # 'int64'

# For Dataframe
type(df) # pandas.core.frame.DataFrame
df.dtypes

Name    object
Age      int64
dtype: object

4.1.4. Summary

• For Python’s inbuilt basic datatypes

  • type()

• For numpy and Pandas columns/series/dataframes

  • .dtype and .dtypes() will tell type of elements (int,object,float)

  • type() will tell type of container (series,nparray,dataframe)