Question

我有一个很大的熊猫数据框，如下所示：

    System.out.println("Q1.HOW DO YOU RECHARGE?");
    System.out.println("A. SPENDING TIME ALONE");
    System.out.println("B. PUBLIC DISCUSSIONS OR PARTIES");
    System.out.println("C. BOTH OF THEM ARE FINE IF I'M IN THE MOOD");
    char choice1 = (char)br.read();//this line gets executed perfectly fine
    if(choice1 == 'A')
    ct+=20;
    else if(choice1 == 'B')
    ct+=10;
    else
    ct+=30;
    System.out.println("Q2.ARE YOU OPEN TO NEW PEOPLE?");
    System.out.println("A. YES OF COURSE");
    System.out.println("B. I DON'T WANT TO TALK TO NEW PEOPLE");
    System.out.println("C. I CAN'T OPEN UP TO NEW PEOPLE UNTIL THEY ARE CLOSE ENOUGH");
    char choice2 = (char)br.read();//this one doesn't
    if(choice2 == 'A')
    ct+=10;
    else if(choice2 == 'B')
    ct+=20;
    else
    ct+=30;
    System.out.println("Q3.WOULD YOU GO FIRST OR LAST FOR A PRESENTATION?");
    System.out.println("A. I GUESS I WOULD GO FIRST ONLY IF I'M VERY CONFIDENT");
    System.out.println("B. OF COURSE FIRST!");
    System.out.println("C. NO. SO THAT I CAN LEARN FROM OTHER'S MISTAKES");
    char choice3 = (char)br.read();//this does
    if(choice3 == 'A')
    ct+=30;
    else if(choice3 == 'B')
    ct+=10;
    else
    ct+=20;

由于数据帧非常大，因此我想将其存储在一个包含一部分数据帧的文件中（假设3个文件占每个数据帧的33％）。有什么办法吗？

Answer 1

当然，将其拆分：

split1 = len(df) // 3
split2 = split1 * 2

df.iloc[:split1].to_json(...)
df.iloc[split1:split2].to_json(...)
df.iloc[split2:].to_json(...)

每个文件写一列可能更有意义：

for col in df.columns:
    df[col].to_json(...)

熊猫：将数据帧存储在多个json文件中

1 个答案: