I have a very large data frame in python and I want to drop all rows that have a particular string inside a particular column.
For example, I want to drop all rows which have the string “XYZ” as a substring in the column C of the data frame.
Can this be implemented in an efficient way using .drop() method?
pandas has vectorized string operations, so you can just filter out the rows that contain the string you don’t want:
In : df = pd.DataFrame(dict(A=[5,3,5,6], C=["foo","bar","fooXYZbar", "bat"])) In : df Out: A C 0 5 foo 1 3 bar 2 5 fooXYZbar 3 6 bat In : df[~df.C.str.contains("XYZ")] Out: A C 0 5 foo 1 3 bar 3 6 bat
If your string constraint is not just one string you can drop those corresponding rows with:
df = df[~df['your column'].isin(['list of strings'])]
The above will drop all rows containing elements of your list
This will only work if you want to compare exact strings.
It will not work in case you want to check if the column string contains any of the strings in the list.
The right way to compare with a list would be :
searchfor = ['john', 'doe'] df = df[~df.col.str.contains('|'.join(searchfor))]
new_df = df[df.C != 'XYZ']
if you do not want to delete all NaN, use
df[~df.C.str.contains("XYZ") == True]
The below code will give you list of all the rows:-
df[df['C'] != 'XYZ']
To store the values from the above code into a dataframe :-
newdf = df[df['C'] != 'XYZ']
Slight modification to the code. Having na=False will skip empty values. Otherwise you can get an error TypeError: bad operand type for unary ~: float