How do I build multi skipper on pandas automatically

Question

Sorry for asking to simple, on mongoDB usually the function that work is Skip, but I get lost when looking pandas equivalent. My pandas query 1st-1000th for is

list1 = a['id'].head(1000)

But, I want to query the 1001st-2000th entry , 2001st-3000th entry, etc.

I'm expected the answer to be saved as df1, df2, ... , dfn

How suppose I do this?

`.head(2000).tail(1000)` works, but probably isn't optimal – Mad Physicist Aug 21 '18 at 13:06 — Mad Physicist, Aug 21 '18 at 13:06
Thanks, at least it works – Nabih Bawazir Aug 21 '18 at 13:07 — Nabih Bawazir, Aug 21 '18 at 13:07

score 4 · Answer 1 · edited Aug 21 '18 at 13:23

4

You can use iloc function, and please remember that indices are zero based

a.iloc[0:1000,]
a.iloc[1000:2000,]

iloc will allow you to filter like this

edited Aug 21 '18 at 13:23

Nabih Bawazir

6,381
7
37
70

answered Aug 21 '18 at 13:06

loegare

142
1
11

1

Oh my god, it just `iloc`, I forget about this function – Nabih Bawazir Aug 21 '18 at 13:06
1

Indices are zero based. `[:1000]` or `[0:1000]`, and `[1000:2000]` – Mad Physicist Aug 21 '18 at 13:10

BENY · Accepted Answer · 2018-08-21T13:24:03.443

1

Try with // and groupby, and save your data frame into a list

l = [x for _,x in df.groupby(np.arange(len(df))//1000)]

Update

variables = locals()
for i,j in df.groupby(np.arange(len(df))//1000):
    variables["df{0}".format(i+1)] = j

edited Aug 21 '18 at 13:24

answered Aug 21 '18 at 13:15

BENY

317,841
20
164
234

Great, that is what I want, my data is 25600 rows, can you save them to `df1`, ...., `df26` ? – Nabih Bawazir Aug 21 '18 at 13:20
@NabihBawazir you can create a dict for that like `d={y:x for y,x in df.groupby(np.arange(len(df))//1000)}` and get the data frame like d[0],d[1].... – BENY Aug 21 '18 at 13:20
That is not split the output to multiple dataframe – Nabih Bawazir Aug 21 '18 at 13:23
1

best answer selected – Nabih Bawazir Aug 21 '18 at 13:28
@Wen - No, dont do this, dictionary is way [for this](https://stackoverflow.com/a/5037218) – jezrael Aug 21 '18 at 13:37
@jezrael I list dict method for him in the comment ...since he is asking for split it to dataframe .. – BENY Aug 21 '18 at 13:38

How do I build multi skipper on pandas automatically

2 Answers2