I have the following function to change the value of column 'A' to np.NaN if the current value is '...'. However, for some reason, nothing is changing.
def findEmpty(row):
row['A'] = np.NaN
energy[energy['A'] == '...'].apply(lambda x:findEmpty(x))
Can someone help?
Pandas apply returns a DataFrame with applied values.
Try this:
energy = energy[energy['A'] == '...'].apply(lambda x:findEmpty(x))
here is a much more efficient solution
energy.ix[(energy['A'] == '...'), 'A'] = np.NaN
be aware that lambda functions are in pure Python. In other words, you basically throw away all the nice C speed-ups in Pandas when you do so. If you use large datasets, always try to avoid loops and lambda functions (when possible)
Pandas has a 'replace' method which would improve the readability
energy.replace('...', np.NaN, inplace=True)
