如何使用 pandas 擦除某个列中的“nan”值？

Question

I have a column that has bunch of rows with a mixture of 'nan'. I only want to delete 'nan', not the entire row that includes 'nan'. Some cells in that column have multiple nans like: nan,nan,nan,nan and some cells has the name that I need with nan attached like: Jefferson,nan,nan,nan

How can I just erase nan?

Answer 1

You can try to replace nan with some value, like 0 or ""? pandas.DataFrame.fillna

Answer 2

From: geekforgeek

Methods to replace NaN values with zeros in Pandas DataFrame:

fillna()
The fillna() function is used to fill NA/NaN values using the specified method.
replace()
The dataframe.replace() function in Pandas can be defined as a simple method used to replace a string, regex, list, dictionary etc. in a DataFrame.

Steps to replace NaN values:

For one column using pandas:

df['DataFrame Column'] = df['DataFrame Column'].fillna(0)

For one column using numpy:

df['DataFrame Column'] = df['DataFrame Column'].replace(np.nan, 0)

For the whole DataFrame using pandas:

df.fillna(0)

For the whole DataFrame using numpy:

df.replace(np.nan, 0)

MY ADVICE: as you use strings, replace NaN by a space " " and when processing your dataframe skip the value when == " "

Answer 3

This should work. I'm using regular expressions to match 'nan' only or 'nan,<something>' and I'm replacing that by an empty string '' .

I decided to use regex because by your question I think you can't use a literal string since you don't know exactly is within the cell (can be any number of 'nan's .

import pandas as pd

data = {'names': ['Jefferson', 'nan', 'Olivia', 'nan', 'nan', 'nan,nan,nan', 'Rebekah'],
        'numbers': [1, 2, 3, 4, 5, 6, 7]}

df = pd.DataFrame(data=data)
df['names'] = df['names'].replace({r'^nan$': '', r'^nan,.*': ''}, regex=True)
df

If we are not talking about the string 'nan' then the df.fillna('') should do.