how to remove NaN from the columns

Question

I have a dataframe with NaN. I have to remove nan at the starting rows only, and wants to keeps NaN after real number starts:

suppose: my data frame is something like:

a = pd.DataFrame({'data':[np.nan,np.nan,np.nan,np.nan,4,5,6,2,np.nan,1,3,4,5,np.nan,4,5,np.nan,np.nan]})

a= 
    data
    0   NaN
    1   NaN
    2   NaN
    3   NaN
    4   4.0
    5   5.0
    6   6.0
    7   2.0
    8   NaN
    9   1.0
    10  3.0
    11  4.0
    12  5.0
    13  NaN
    14  4.0
    15  5.0
    16  NaN
    17  NaN

and I tried to remove NaN at the beginning and wants data-frame like this:

        data
    4   4.0
    5   5.0
    6   6.0
    7   2.0
    8   NaN
    9   1.0
    10  3.0
    11  4.0
    12  5.0
    13  NaN
    14  4.0
    15  5.0
    16  NaN
    17  NaN

I tried to use this function but not working. Any help will be highly appreciated.

for w in np.arange(len(a)):
    if a.iloc[w] == np.nan:
        a.drop(a.index[w])

Answer 1

Get the first valid index and slice

idx = a.first_valid_index()
a.loc[idx:]

    data
4   4.0
5   5.0
6   6.0
7   2.0
8   NaN
9   1.0
10  3.0
11  4.0
12  5.0
13  NaN
14  4.0
15  5.0
16  NaN
17  NaN

Answer 2

try something like this:

start = a[a.data.notnull()].index[0]
new_df = a.loc[start:]

the first line finds the index of the first non-null value, the second cuts out all the entries before that from your dataframe.

Answer 3

Instead of removing the "bad" rows, you can locate and preserve the "good" rows:

b = a[a.data.fillna(method='ffill').notnull()]
#     data
#4    4.0
#5    5.0
#6    6.0
#7    2.0
#8    NaN
#9    1.0

Answer 4

Ummm , you should using first_valid_index() , but here is another way :-)

a.loc[a.data.notnull().nonzero()[0][0]:]
Out[1276]: 
    data
4    4.0
5    5.0
6    6.0
7    2.0
8    NaN
9    1.0
10   3.0
11   4.0
12   5.0
13   NaN
14   4.0
15   5.0
16   NaN
17   NaN