如何使用 pandas 將多行轉換為同一 ID 的單行

Question

我有以下格式的文本文件，它具有唯一 ID，每個唯一 ID 有四行，現在我需要將特定 ID 轉換為單行。 假設如果有 8 行，output 應該有 2 行。 而且它沒有 header ，我需要使用熊貓！

    xyz,name,,,12345
    2nd street,add,,,12345
    xyx@mail.com,email,,,12345
    575xxx5678,contact,,,12345

output

xyz,name,,,12345,2nd street,add,,,12345,xyx@mail.com,email,,,12345,575xxx5678,contact,,,12345

將唯一 ID 視為 12345，可以幫助我解決此問題。 那會很好。 提前致謝。

Answer 1

假設你有這個file.csv ：

www,contact,,,99999
xyz,name,,,12345
2nd street,add,,,12345
xyx@mail.com,email,,,12345
575xxx5678,contact,,,12345
qqq,contact,,,99999

要將其讀取到 pandas：

df = pd.read_csv("file.csv", names=["col1", "col2", "col3", "col4", "ID"])
print(df)

印刷：

           col1     col2  col3  col4     ID
0           www  contact   NaN   NaN  99999
1           xyz     name   NaN   NaN  12345
2    2nd street      add   NaN   NaN  12345
3  xyx@mail.com    email   NaN   NaN  12345
4    575xxx5678  contact   NaN   NaN  12345
5           qqq  contact   NaN   NaN  99999

然后將其轉換為您想要的 output：

x = (
    df.assign(ID2=df["ID"])
    .groupby("ID")
    .agg(list)
    .apply(lambda x: [v for l in zip(*x) for v in l], axis=1)
)

pd.DataFrame(x.tolist()).to_csv("output.txt", sep=",", header=None, index=None)

這將創建output.txt ：

xyz,name,,,12345,2nd street,add,,,12345,xyx@mail.com,email,,,12345.0,575xxx5678,contact,,,12345.0
www,contact,,,99999,qqq,contact,,,99999,,,,,,,,,,