In this article, we will discuss NOT IN filter in pandas, NOT IN is a membership operator used to check whether the data is present in dataframe or not. It will return true if the value is not present, otherwise false
Let’s create a sample dataframe
Python3
# import pandas moduleimport pandas as pd # create dataframedata1 = pd.DataFrame({'name': ['sravan', 'harsha', 'jyothika'], 'subject1': ['python', 'R', 'php'], 'marks': [96, 89, 90]}, index=[0, 1, 2]) # displaydata1 |
Output:
sample dataframe
Method 1: Use NOT IN Filter with One Column
We are using isin() operator to get the given values in the dataframe and those values are taken from the list, so we are filtering the dataframe one column values which are present in that list.
Syntax: dataframe[~dataframe[column_name].isin(list)]
where
- dataframe is the input dataframe
- column_name is the column that is filtered
- list is the list of values to be removed in that column
Python3
# import pandas moduleimport pandas as pd # create dataframedata1 = pd.DataFrame({'name': ['sravan', 'harsha', 'jyothika'], 'subject1': ['python', 'R', 'php'], 'marks': [96, 89, 90]}, index=[0, 1, 2]) # consider a listlist1 = ['harsha', 'jyothika'] # filter in name columnprint(data1[~data1['name'].isin(list1)])print("============") # consider a listlist2 = ['R'] # filter in name columnprint(data1[~data1['subject1'].isin(list2)])print("============") # consider a listlist3 = [96, 89] # filter in name columnprint(data1[~data1['marks'].isin(list3)]) |
Output:
NOT IN Filter with One Column
Method 2: Use NOT IN Filter with Multiple Column
Now we can filter in more than one column by using any() function. This function will check the value that exists in any given column and columns are given in [[]] separated by a comma.
Syntax: dataframe[~dataframe[[columns]].isin(list).any(axis=1)]
Python3
# import pandas moduleimport pandas as pd # create dataframedata1 = pd.DataFrame({'name': ['sravan', 'harsha', 'jyothika'], 'subject1': ['python', 'R', 'php'], 'marks': [96, 89, 90]}, index=[0, 1, 2]) # consider a listlist1 = ['harsha', 'jyothika', 96] # filter in name and marks columnprint(data1[~data1[['name', 'marks']].isin(list1).any(axis=1)])print("============") # consider a listlist2 = ['R', 'sravan'] # filter in name and subject1 columnprint(data1[~data1[['subject1', 'name']].isin(list2).any(axis=1)]) |
Output:
NOT IN Filter with Multiple Column
Method 3: Use numpy with NOT IN filter
This is similar to the above functionality.
Syntax: dataframe[~numpy.isin(dataframe[‘column’], list)]
Python3
# import pandas moduleimport numpy as npimport pandas as pd # create dataframedata1 = pd.DataFrame({'name': ['sravan', 'harsha', 'jyothika'], 'subject1': ['python', 'R', 'php'], 'marks': [96, 89, 90]}, index=[0, 1, 2]) # consider a listlist1 = ['harsha', 'jyothika', 96] # filter in name columndata1[~np.isin(data1['name'], list1)] |
Output:
numpy with NOT IN filter
