Python Forum

Hi guys,

I am extracting tables using Pandas and want to get rid of empty rows in the dataframe. One of the tables looks like this:

Output:
 Instrument           Price           Order Date   Type Lots  Gap   Duration Note
0  nnNZD/USDnn  nnSelln0.73250  Sep 27, 2016 7:17pm  Limit        41  189 weeks     
1                                                                                   
2                                                                                   
3  nnGBP/CHFnn  nnSelln1.32000  Sep 1, 2016 11:33am  Limit       595  193 weeks     
4                                                                                   
5          

so to delete empty rows I use in code:

df.dropna(axis = 0, how = 'all', inplace=True)

Unfortunately does not work!

Any ideas please?

Your data frame probably filled with empty strings not nan values.

Probably your data frame is filled with empty strings. Here is a way to remove them.

df[df['Price'].astype(bool)]

So basically, what we are doing here is that

df['Price'].astype(bool)

returns True wherever there is a value in Price columns, and False wherever there is empty string.So when we use

df[df['Price'].astype(bool)]

,it returns just the rows that have True value for Price column.You can use any column in place of price

This does work thanks.

Is there a solution 'non column' specific?

So I really only want to remove rows where the whole row is empty (or with empty strings). This solution is based on if the rows of 'Column name' are empty as opposed to the complete row i.e across all columns.

---

I have used this before but is there a more elegant way?

for index, row in df.iterrows():                                                            #loop through rows of df for further df cleaning
            if row.any() == '':
                df.drop(index,inplace=True)