Question

我有一个数据框，

ip_df:
     name class    sec    details
0    tom  I        a      [{'class':'I','sec':'a','subjects':['numbers','ethics']},{'class':'I','sec':'b','subjects':['numbers','moral-science']},{'class':'I','sec':'c','subjects':['moral-science','ethics']},{'class':'I','subjects':['numbers','ethics1']}]
1    sam  I        d      [{'class':'I','sec':'a','subjects':['numbers','ethics']},{'class':'I','sec':'b','subjects':['numbers','moral-science']},{'class':'I','sec':'c','subjects':['moral-science','ethics']},{'class':'I','subjects':['numbers','ethics1']}]

，结果数据帧应该是

op_df:
      name  class  sec   subjects
0     tom   I      a     ['numbers','ethics']
1     sam   I      d     ['numbers','ethics1']

“ op_df”必须根据以下条件进行构图，

条件1：检查“详细信息”列中是否存在“类”和“秒”，如果存在，则添加一个新列，其名称分别为“主题”
条件2：如果“详细信息”列中不存在“类别”和“秒”，请检查“类别”是否匹配，如果是，则添加一个新列，其名称分别为“主题”
如果条件1和条件2都不存在，请在“主题”列中将默认值添加为[0,0]

Answer 1

如果两个条件都需要首先匹配值的解决方案，请使用next和iter技巧，如果不匹配，则添加默认值[0, 0]：

final = []
for a, b, c in zip(df['class'], df['sec'], df['details']):
    out = []
    for x in c:
        m1 = x['class'] == a 
        if m1 and x.get('sec') == b:
            out.append(x['subjects'])
        elif m1 and 'sec' not in list(x.keys()):
            out.append(x['subjects'])
    final.append(next(iter(out), [0,0]))

df['subjects'] =  final

熊猫：比较数据框的列并根据条件添加新的列和值

1 个答案: