How to create n data frames with same structure according to value?-CodePudding

My initial dataframe looks as follows:

name	id
test	1
test	2
test	2
test	3
test	4
test	5
test	6
test	7
test	8
test	9
test	9
test	10
test	11
test	12
test	13

Now I want to create three subdataframes (or a list of dataframes) which contain the rows with the following ids:

df[0]: 1, 4, 7, 10, 13

df[1]: 2, 5, 8, 11

df[2]: 3, 6, 9, 12

So in fact df[1] results in the following:

name	id
test	2
test	2
test	5
test	8
test	11

I tried it with a loop and "append" but I read it is quite slow if you have many rows. I am really not sure how to do it in a nice and effective way with pandas.

CodePudding user response：

df is your initial dataframe:

dfs = []
a = [[1, 4, 7, 10, 13],[2, 5, 8, 11],[3, 6, 9, 12]]

for i in a:
    dfs.append(df.loc[df['id'].isin(i)])


print(dfs[0])
print(dfs[1])
print(dfs[2])

output:

    name  id
0   test   1
4   test   4
7   test   7
11  test  10
14  test  13
    name  id
1   test   2
2   test   2
5   test   5
8   test   8
12  test  11
    name  id
3   test   3
6   test   6
9   test   9
10  test   9
13  test  12

CodePudding user response：

Given the pattern, it seems that you want to group based on the result of df.id % 3. In that case, here is a vectorized approach using DataFrame.groupby

groups = [g for k, g in df.groupby(df.id % 3, sort=False)]

Input

>>> df

    name  id
0   test   1
1   test   2
2   test   2
3   test   3
4   test   4
5   test   5
6   test   6
7   test   7
8   test   8
9   test   9
10  test   9
11  test  10
12  test  11
13  test  12
14  test  13
    name  id
0   test   1
4   test   4
7   test   7
11  test  10
14  test  13

Output

>>> groups[0]

    name  id
0   test   1
4   test   4
7   test   7
11  test  10
14  test  13 

>>> groups[1]

    name  id
1   test   2
2   test   2
5   test   5
8   test   8
12  test  11 

>>> groups[2]

    name  id
3   test   3
6   test   6
9   test   9
10  test   9
13  test  12