将 x 行合并为一行 CSV, Pandas

Question

I have a dataframe looking like that:

  ID  val1 val2 ... val10
1 1   45   2
2 1   34   90 
3 1   90   9
4 1   10   10
...
5000

I would like to merge the val-columns combining 10 rows (1-10, 11-20...) to 1 row so that it looks like this (every 10 rows the ID Changes, I do not wanna sort the data, just combine 10 rows to 1 row):

  ID  val1 val2 ... val100
1 1   45   2
2 2   10   60 
3 3   40   9
4 4   1   140
...
500

I am pretty new to pandas and I was only able to find an easy method to sort and then combine rows by having the same column value. The problem is, that I only want to combine the val-columns of 10 rows and sometimes in my date the same ID occurs again.

Answer 1

This should do:

import numpy as np
import pandas as pd

# create your input dataframe
df = pd.DataFrame({"ID": range(1, 5001)})
for n in range(1, 11):
    df[f"val{n}"] = np.random.randint(low=0, high=140, size=(5000, 1))

# initialize the output dataframe with the correct size and column names
out_df = pd.DataFrame(np.zeros((500, 100)))
out_df.columns = [f"val{n}" for n in range(1, 101)]
out_df.index += 1
out_df.index.name = "ID"

# copy the values from df to out_df
for n in range(10):
    out_df.iloc[0:500, n * 10: (n + 1) * 10 ] = df.copy().iloc[500 * n: 500 * (n+ 1), 1:]