CODEWITHSUNDEEP
pythonpandaslistdrop
DevLeb2022
DevLeb2022

Reputation: 685

how to drop multiple columns in pandas?

How to drop multiple columns in pandas and python ?

import pandas as pd

df =pd.DataFrame({
            "source_number":                        [ 
             [11199,11328,11287,32345,12342,1232,13456,123244,13456],
             "location":          
             ["loc2","loc1-loc3","loc3","loc1","loc2-loc1","loc2","loc3-loc2","loc2","loc1"],
              "category": 
             ["cat1","cat2","cat1","cat3","cat3","cat3","cat2","cat3","cat2"],
             })    

def remove_columns(dataset,cols):
    for col in cols:
        del dataset[col]
    return dataset

for col in df.columns:
    df=remove_columns(df,col)
df.head()

in the code above the task is done and the columns are droped.

But when I tried this code on streamlit where the user select multiple columns that he want to remove from the dataframe.

But the problem is that the system just take the first element and not all the item in the list.

like if the user select location and source number the col variable will contains just location and display the below error:

KeyError: 'location'
Traceback:
File "f:\aienv\lib\site-packages\streamlit\script_runner.py", line 333, in _run_script
    exec(code, module.__dict__)
File "F:\AIenv\streamlit\app.py", line 373, in <module>
    sidebars[y]=st.sidebar.multiselect('Filter '+y, df[y].unique(),key="1")
File "f:\aienv\lib\site-packages\pandas\core\frame.py", line 2902, in __getitem__
    indexer = self.columns.get_loc(key)
File "f:\aienv\lib\site-packages\pandas\core\indexes\base.py", line 2893, in get_loc
    raise KeyError(key) from err

Streamlit code:

import numpy as np
import pandas as pd
import streamlit as st 

#function drop unwanted columns
def remove_columns(dataset,cols):
    for col in cols:
        del dataset[col]
    return dataset

df =pd.DataFrame({
          "source_number":                        [ 
           [11199,11328,11287,32345,12342,1232,13456,123244,13456],
           "location":          
           ["loc2","loc1-loc3","loc3","loc1","loc2-loc1","loc2","loc3-loc2","loc2","loc1"],
           "category": 
           ["cat1","cat2","cat1","cat3","cat3","cat3","cat2","cat3","cat2"],
           })    

drop_button = st.sidebar.button("Remove")
columns = st.sidebar.multiselect("Select  column/s", df.columns)
sidebars = {}
for y in columns:
    ucolumns=list(df[y].unique())   
    st.write(y)
    if (drop_button):
        df_drop=df.drop(y,axis=1,inplace=True)
        print(y)
    st.table(df)

Upvotes: 1

Views: 2972

Answers (2)

jezrael
jezrael

Reputation: 863166

Use DataFrame.drop:

def remove_columns(dataset,cols):
    return dataset.drop(cols, axis=1)

And for call pass function with no loop - is possible pass scalar or list:

df = remove_columns(df,'location')

df = remove_columns(df,['location','category'])

EDIT:

If need remove column seelcted in multiselect use:

drop_button = st.sidebar.button("Remove")

#in columns variable are selected values
columns = st.sidebar.multiselect("Select  column/s", df.columns)
print (columns)

#so if use button remove values by variable columns
if (drop_button):
    df.drop(columns,axis=1,inplace=True)

    st.table(df)

Upvotes: 1

Minh Qu&#226;n
Minh Qu&#226;n

Reputation: 36

Pandas have already implemented this inside of the function drop. You can use pandas.drop with parameter columns = [columns that you want to drop] like this instead:

df.drop(columns = ["source_number","location"])

I hope this is what you are looking for

Upvotes: 1

Related Questions