JIYIK CN >

We will demonstrate methods to convert floating point numbers to integers in Pandas DataFrame - astype(int)and to_numeric()methods.

First, we NumPycreate a random array using the library and then convert it to DataFrame.

import pandas as pd
import numpy as np

df = pd.DataFrame(np.random.rand(5, 5) * 5)
print(df)

If you run this code, you will get the following output, with data types as follows float.

0 1 2 3 4
00.3024483.5519583.8786602.3803524.741592
14.0541870.9409520.4590584.3148010.524993
22.8917334.9268854.9557732.6263734.144166
31.1276393.1968234.1440201.3506320.401138
41.4235372.0194553.0389450.4366573.823888

astype(int)Convert floating point numbers floatto integers in Pandas usingint

To floatconvert to INT, we will use the method Pandasprovided by the package astype(int). The code is,

import pandas as pd
import numpy as np

df = pd.DataFrame(np.random.rand(5, 5) * 5)
print("*********** Random Float DataFrame ************")
print(df)
print("***********************************************")
print("***********************************************")
print("*********** Dataframe Converted into INT ************")
print(df.astype(int))
print("***********************************************")
print("***********************************************")

Output:

*********** Random Float DataFrame ************
0 1 2 3 4
01.5106181.0945614.1574194.4241954.872719
10.4576804.0029592.6609991.6869160.840159
21.7817783.8129240.5618270.5323280.752800
31.4565142.7749552.7001184.5033544.749377
42.2235204.8592380.4509773.2284442.541648
***********************************************
***********************************************
*********** Dataframe Converted into INT ************
 01234
011444
104210
213000
312244
424032
***********************************************
***********************************************

We can use df.round(0).astype(int)to floatround the value to int.

import pandas as pd
import numpy as np

df = pd.DataFrame(np.random.rand(5, 5) * 5)
print("*********** Random Float DataFrame ************")
print(df)
print("***********************************************")
print("***********************************************")
print("*********** Dataframe Converted into INT ************")
print(df.astype(int))
print("***********************************************")
print("***********************************************")
print("*********** Rounding Float value to INT ************")
print(df.round(0).astype(int))
print("***********************************************")
print("***********************************************")

After running the code, we will get the following output.

*********** Random Float DataFrame ************
0 1 2 3 4
03.3944580.0712023.3132054.7858780.705612
11.9603530.8420851.2123320.8183432.637940
24.4078114.2390781.4190271.5263101.043394
33.6542164.6729721.7964393.1684260.734009
40.8487961.0240521.8593190.8443781.747628
***********************************************
***********************************************
*********** Dataframe Converted into INT ************
 01234
030340
110102
244111
334130
401101
***********************************************
***********************************************
*********** Rounding Float value to INT ************
 01234
030351
121113
244121
345231
411212
***********************************************
***********************************************

to_numeric()floatMethod converts Pandas toint

This method provides the ability to safely convert non-numeric types (such as strings) to a suitable numeric type. Takes a single series and converts to numbers, coercing when told to

s = pd.Series(["1.0", "2", -3])
print(pd.to_numeric(s, downcast="integer"))

After running the code, we will get the following output.

01
12
2 -3
dtype: int8