How to count NaN values for each row in pandas dataframe?-CodePudding

I want to count NaN values for each row in a DataFrame and then get the one with the minimum numbers of such values. My solution is too slow, also it is not a pandas-way to do it using for loop. Is there a better and faster way to do it?

max_not_nan = 13 # a maximum possible value of NaN's (number of columns   1) 
row_number = 0
for i in range(df.shape[0]):
  if df.iloc[i].isna().sum() < max_not_nan:
    max_not_nan = df.iloc[i].isna().sum()
    row_number = i

It works fine expect the time complexity

CodePudding user response：

Can you try this:

df['nan_count'] = df.isnull().sum(axis=1) #get nan counts for each row as a new column

max_nan=df[df['nan_count']==df['nan_count'].max()] #get the row with the max nan count
min_nan=df[df['nan_count']==df['nan_count'].min()] #get the row with the min nan count

CodePudding user response：

transactions.isnull().sum(axis=1).sort_values()