Question

假设我有一个带有ID，月份和状态的pandas DataFrame。每个ID有几行，每行代表一个月以及给定月份的状态。我想找到具有所有可能状态的ID。我该怎么做？

async function createselbox(arr) {
    console.log('In createselbox');
    var startstr = `Something`;
    mytemp = '';
    for (var i = 0; i < arr.length; i++) {
        mytemp = mytemp + '<option>' + arr[i] + '</option>';
    }
    var ret = startstr + mytemp + `Something else    `;
    let IsStaff = await CheckiStaff();
    console.log('Checkifstaff is ' + IsStaff);
    if (IsStaff) {
        console.log('Got true createselbox');
        console.log('In sel loop');
        ret = startstr + mytemp + `Something more`;
    } else {
        console.log('Got false createselbox');
    }
    return ret;
}

function CheckifStaff() {
    var data = {
        "isstaff": 'check'
    };
    data = $(this).serialize() + "&" + $.param(data);
    return new Promise((resolve, reject) => {
      $.ajax({
        type: "POST",
        dataType: "html",
        url: "/functions/checkifstaff",
        data: data,
        success: function (data) {
            var recdata = data;
            if (recdata.indexOf('Has add permissions') != -1) {
                console.log('Returning true');
                resolve(1);
            } else {
                console.log('Returning false');
                resolve(0);
            }
        },
        error: function (xhr, textStatus, errorThrown) {
            console.log('An error occured in CheckifStaff');
            resolve(0);
        }
    });
    });
    
}

我希望找到输出12的代码，因为该ID包含所有可能的状态值（在这种情况下为'a'和'b'）

Answer 1

尝试：

>>> df2 = df.groupby('ID').aggregate(set)
>>> df2

     Month  Status
ID      
11  {1, 2}  {a}
12  {1, 2}  {b, a}

>>> max_states = 2   # you should know that
>>> list(df2[df2.apply(lambda x: len(x['Status'])==max_states, axis=1)].index)

[12]

Answer 2

您可以通过使用crosstab然后仅考虑具有所有非零元素的行来实现此目的：

#Creates a crosstab the variable 'Status'
df1 = pd.crosstab(df['ID'], df['Status'])
#Considers only rows where all values are non-zero
df1 = df1[(df1.T != 0).all()]

print(df1)

Status  a  b
ID          
12      1  1

如何查找包含熊猫中另一列的所有可能值的行值

2 个答案: