CODEWITHSUNDEEP
pythonpandasdataframe
Pajul
Pajul

Reputation: 135

Print the colname and rowname for values that meet certain condition

I am desperatly trying to figure out how to print out the row index and col name for specific values in my df.

I have the following df:

raw_data = {'first_name': [NaN, 'Molly', 'Tina', 'Jake', 'Amy'], 
        'last_name': ['Miller', 'Jacobson', 'Ali', 'Milner', 'Cooze'], 
        'age': [42, 52, NaN, 24, 73], 
        'preTestScore': [4, 24, 31, 33, 3],
        'postTestScore': [25, 94, 57, 62, 70]}

df = pd.DataFrame(raw_data, columns = ['first_name', 'last_name', 'age', 
'preTestScore','postTestScore'])

I now want to print out the index and column name for the NaN:

There is a missing value in row 0 for first_name.
There is a missing value in row 2 for age.

I have searched a lot and always found how to do something for one row. My idea is to first create a df with False and True

na = df.isnull()

Then I want to apply some function that prints the row number and col_name for every NaN value. I just cant figure out how to do this.

Thanks in advance for any help!

Upvotes: 7

Views: 251

Answers (5)

Cainã Max Couto da Silva
Cainã Max Couto da Silva

Reputation: 4929

As simple as:

np.where(df.isnull())

It returns a tuple with the row indexes, and column indexes with NAs, respectively.

Example:

na_idx = np.where(df.isnull())
for i,j in zip(*na_idx):
    print(f'Row {i} and column {j} ({df.columns[j]}) is NA.')

Upvotes: 3

Tom Wojcik
Tom Wojcik

Reputation: 6179

I think melting is the way to go.

I'd start by creating a dataframe with columns: index, column_name, value. Then filter column value by not null. And dump the result to dict.

df = pd.melt(df.reset_index(), id_vars=['index'], value_vars=df.columns)
selected = df[df['value'].isnull()].drop('value', axis=1).set_index('index')
resp = selected.T.to_dict(orient='records')[0]
s = "There is a missing value in row {idx} for {col_name}."

for record in resp.items():
    idx, col_name = record
    print(s.format(idx=idx, col_name=col_name))

Upvotes: 1

lww
lww

Reputation: 661

you can just create a variable NaN = "null" to indicate empty cell


import pandas as  pd

NaN = "null"
raw_data = {'first_name': [NaN, 'Molly', 'Tina', 'Jake', 'Amy'], 
        'last_name': ['Miller', 'Jacobson', 'Ali', 'Milner', 'Cooze'], 
        'age': [42, 52, NaN, 24, 73], 
        'preTestScore': [4, 24, 31, 33, 3],
        'postTestScore': [25, 94, 57, 62, 70]}

df = pd.DataFrame(raw_data, columns = ['first_name', 'last_name', 'age', 
'preTestScore','postTestScore'])
print(df)

output:

first_name last_name   age  preTestScore  postTestScore
0       null    Miller    42             4             25
1      Molly  Jacobson    52            24             94
2       Tina       Ali  null            31             57
3       Jake    Milner    24            33             62
4        Amy     Cooze    73             3             70

Upvotes: 0

oskros
oskros

Reputation: 3285

You could do something like the below:

for i, row in df.iterrows():
    nans = row[row.isna()].index
    for n in nans:
        print('row: %s, col: %s' % (i, n))

Upvotes: 2

piterbarg
piterbarg

Reputation: 8219

had to change the df a bit because of NaN. Replaced with np.nan

import numpy as np
import pandas as pd
raw_data = {'first_name': [np.nan, 'Molly', 'Tina', 'Jake', 'Amy'], 
        'last_name': ['Miller', 'Jacobson', 'Ali', 'Milner', 'Cooze'], 
        'age': [42, 52, np.nan, 24, 73], 
        'preTestScore': [4, 24, 31, 33, 3],
        'postTestScore': [25, 94, 57, 62, 70]}

you can do this

dfs = df.stack(dropna = False)
[f'There is a missing value in row {i[0]} for {i[1]}' for i in dfs[dfs.isna()].index]

prints a list

['There is a missing value in row 0 for first_name',
 'There is a missing value in row 2 for age']

Upvotes: 4

Related Questions