Python - Pandas - drop specific columns (axis)?-CodePudding

so i got a numeric list [0-12] that matches the length of my columns in my spreadsheet and also replaced the column headers with that list >> df.columns = list Now i want to drop specific columns out of that spreadsheet like this: enter image description here

To create the list of numbers to match the length of columns i got this:



listOfNumbers = []

column_name = []
for i in range(0, len(df.columns)):
    listOfNumbers.append(i)

df.columns = listOfNumbers

for i in range(1, len(df.columns)):
    for j in range(1, len(df.columns)):
        if i != colList[j]:
            df.drop(i, inplace=True)

and i got the list [1,2,3] as seen in the picture. But i always get this Error: KeyError: '[1] not found in axis'

i tried to replace df.drop(i, inplace=True) with df.drop(i, axis=1, inplace=True) but that didnt work either

Any suggestions? Thanks.

CodePudding user response：

the proper way will be:

columns_to_remove = [1, 2, 3] # columns to delete
df = df.drop(columns=df.columns[columns_to_remove])

So for your use case:

for i in range(1, len(df.columns)):
    for j in range(1, len(df.columns)):
        if i != colList[j]:
            df.drop(columns=df.columns[i], inplace=True)

CodePudding user response：

If you want to drop every column that does not appear in colList, this code does it, using set difference:

setOfNumbers = set(range(df.shape[1]))
setRemainColumns = set(colList)
for dropColumn in setOfNumbers.difference(setRemainColumns):
     df.drop(dropColumn, axis=1, inplace=True)