遍历 dataframe 列并在 python pandas 中创建新列

Question

I have a dataframe like this DataFrame

I want to create a new column(seq) and do the following things

• df['seq'] = First item of every group(by id) is 0
• df['seq'] = Keep Increment until you see a Date in the date column by group(by id)
• if date is present reset the column seq to 0 again follow the same increment process

This is my expected answer expected output:

Thanks

Answer 1

The easiest solution it comes to me is to create an array and fill it with the value of a counter when you loop over your dataframe. You will add the new column after.

For example:

seq = np.zeros(len(df))
date = ''
id = 0
counter = 0
for i in range(len(df)):
    test_date = df['Dates'].iloc[i]
    test_id = df['id'].iloc[i]
    if (test_date and test_date != date) or (test_id != id):
        # new date or id detected
        date = test_date
        id = test_id
        counter = 0
    else:
        counter += 1
    seq[i] = counter

df['seq'] = seq

It might be a more efficient way to do that, but from my experience, performances are ok when using numpy array to add data in dataframe.