1. Python
  2. Python | Pandas Series.lt ()

Python | Pandas Series.lt ()



Pandas Series.lt() is used to compare two rows and return a boolean value for each matching item.

Syntax: Series.lt (other, level = None, fill_value = None, axis = 0)

Parameters:
other: other series to be compared with
level: int or name of level in case of multi level
fill_value: Value to be replaced instead of NaN
axis: 0 or `index` to apply method by rows and 1 or `columns` to apply by columns.

Return type: Boolean series

Note: results are returned based on a comparison of the calling series & lt; other series.

To load the dataset used in the following example, click here.

In the following examples, the data frame being used contains data some NBA players. An image of the data frame before any operations is attached below. 

Example # 1:

This example compares the Age and Weight columns using the .lt () method. Since the values ​​in the weight columns are very large compared to the age column, therefore, the values ​​are divisible by 10 first. The .dropna () method removes null rows before comparison to avoid errors.

# pandas module import

import pandas as pd 

 
# create data frame

data = pd.read_csv ( " https://media.python.engineering/wp-content/uploads/nba. csv " ) 

  
# deleting null z to avoid errors

data.dropna (inplace = True ) 

 
# other series

other = data [ " Weight " ] / 10

 
# method call and return to new column

data [ "Age & lt; Weight " ] = data [ "Age" ]. lt (other)

 
# display
data

Output:
As shown in the output image, the new column is True, where the value in the Age column is less, than Weight / 10.

Example # 2: Handling values NaN

In this example, two series are created using pd.Series () . The series also contains a value of zero and therefore 10 is passed to the fill_value parameter to replace the zero values ​​with 10.

# import pandas module

import pandas as pd 

 
# numpy module import

import numpy as np

 
# create episode 1

series1 = pd.Series ([ 11 , 21 , 2 , 43 , 9 , 27 , np.nan, 110 , np.nan])

 
# creating series 2

series2 = pd.Series ([ 16 , np.nan, 2 , 23 , 5 , 40 , np.nan, 0 , 19 ])

 
# setting the replacement value to zero

na_replace = 10

 
# call and save the result

result = series1.lt (series2, fill_value < / code> = na_replace)

 
# display
result

Output:
As you can see from the results, the NaN values ​​have been replaced with 5 and the comparison is done after the replacement, and the new values ​​are used for comparison. 

 0 True 1 False 2 False 3 False 4 False 5 True 6 False 7 False 8 True dtype: bool