For example, I have some data like this:
column = pd.Series([1,2,3,np.nan,4,np.nan,7])
print column
Executing the command, the result looks like:
0 1.0
1 2.0
2 3.0
3 NaN
4 4.0
5 NaN
6 7.0
Now I want to know what is the first value before each NaN value, such as 3.0 which is before the first NaN. And 4.0 is the result before the second NaN value. Is there any built-in function in pandas which can complete this or shall I write a for loop to do this?
Same idea as @jezrael... numpyfied.
column[np.append(np.isnan(column.values)[1:], False)]
2 3.0
4 4.0
dtype: float64
Complete with pd.Series reconstruction
m = np.append(np.isnan(column.values)[1:], False)
pd.Series(column.values[m], column.index[m])
2 3.0
4 4.0
dtype: float64
Not nearly as quick but intuitive. Group by the the cumsum of isnull and take the last value. Of this result, get rid of last row.
column.groupby(column.isnull().cumsum()).last().iloc[:-1]
0 3.0
1 4.0
dtype: float64