K-fold "split_num" issue

[ohleevah]

Hello,

I am currently trying to run through your "code for publication" on my ECG data and have successfully gotten as far as creating the train and test sets. Could you please clarify this "split_num" column in the long_term_outcome data frame (as far as I understand it, it is referencing an existing column and assigning it as a value also called "split_num")?

`

get train and validation splits

    for index, row in long_term_outcome.iterrows():
        if index >= 0:     
            split_num = row['split_num']
            if split_num == 0:
                Train_dict[str(row['id'])] = 1
                
            elif split_num == 1:
                Val_dict[str(row['id'])] = 1

`

Is this something you set up in the data frame prior to starting the loop? If so, could you provide more details on the contents of this variable?