Given a list of numbers including some missing values, turn it into a pandas dataframe, impute the missing values with the mean, and finally return the dataframe. code example
Example 1: missing values in a dataset python
df.isnull().sum()
Example 2: filling the missing data in pandas
note:to fill a specific value
varable = 1
def fill_mod_acc(most_related_coloum_name,missing_data_coloum):
if np.isnan(missing_data_coloum):
return varable[most_related_coloum_name]
else:
return missing_data_coloum
df['missing_data_coloum'] = df.apply(lambda x:fill_mod_acc(x['most_related_coloum_name'],x['missing_data_coloum']),axis=1)
Note:to fill mean from existing closley related coloum
varable = df.groupby('most_related_coloum_name').mean()['missing_data_coloum']
def fill_mod_acc(most_related_coloum_name,missing_data_coloum):
if np.isnan(missing_data_coloum):
return varable[most_related_coloum_name]
else:
return missing_data_coloum
df['missing_data_coloum'] = df.apply(lambda x:fill_mod_acc(x['most_related_coloum_name'],x['missing_data_coloum']),axis=1)