根据条件从现有数据框列中创建新列

Question

GroupID列具有两个值，一个为字母数字，另一个为数字，我想使用该列的字母数字部分来创建新列，并具有某种条件，例如如果o_dict中存在字母数字，则它应该返回值，否则应该返回“ NOT IN DIC “

GroupID
0   ad32s;#1214;#rf343;#4343
1   wd435;#6464;#ed532;#5454
2   av345e:#3132
3   ok132d;#8897
4   tn123h;#0980
5   as845;#657;#un567t;#456;#qw147;#123
6   ok132d;#8897
7   as845;#657;#un567t;#456;#qw147;#123
8   wd435;#6464;#ed532;#5454

o_dict= {"ad32s":"rupesh","ed532":"Frank","dr501u":"David","ok132d":"Ripal",
         "qw147":"ilesh","av345e":'carls'}

下面是我的代码：

def function01(row):
o_dict= {"ad32s":"rupesh","ed532":"Frank","dr501u":"David","ok132d":"Ripal","qw147":"ilesh","av345e":'carls'}
    if element.isalnum():
        if element in o_dict:
            return owner_dict[element]
        else:
            return "NOT IN DIC"
    else:
        continue
df['New_column'] = df.apply(lambda x: function01(x), axis=1)

如果字母数字值位于第一个位置，则此代码有效，但是如果其位于3或5位，则此代码不起作用。它适用于行0,2,3,4 6，但不适用于1,5,7,8。

O / p应该有两列，其字典中的值与groupid相匹配，否则应填充“ NOT IN DIC”。

我不确定我现在能做什么，是否有另一种方法来获取此值？是否有任何搜索功能可用于搜索此值？

感谢您的帮助:)

Answer 1

我发现在我的代码中，for循环仅适用于列表中的第一个值，并且它填充“ NOT IN DIC”而不检查其他值。我现在进行了以下更改，并获得了预期的输出。

def function01(row):
o_dict= {"ad32s":"rupesh","ed532":"Frank","dr501u":"David","ok132d":"Ripal","qw147":"ilesh","av345e":'carls'}
listA = row['Assigned'].split(";#")
listB = [i for i in listA if i.isdigit()==False]
for element in listA:
    if element in owner_dict:
        return owner_dict[element]
    else:
        continue    
return "NOT IN DIC"

df['New_column'] = df.apply(lambda x: function01(x), axis=1)

Answer 2

您可能想使用numpy.select

    import numpy
    import pandas

    d = {
        "GroupID": [
            "ad32s;#1214;#rf343;#4343",
            "wd435;#6464;#ed532;#5454",
            "av345e:#3132",
            "ok132d;#8897",
            "tn123h;#0980",
            "as845;#657;#un567t;#456;#qw147;#123",
            "ok132d;#8897",
            "as845;#657;#un567t;#456;#qw147;#123",
            "wd435;#6464;#ed532;#5454",
        ]
    }

    o_dict = {
        "ad32s": "rupesh",
        "ed532": "Frank",
        "dr501u": "David",
        "ok132d": "Ripal",
        "qw147": "ilesh",
        "av345e": "carls",
    }

    df = pandas.DataFrame.from_dict(d)

    values = []
    def fn(k):
        values.append(o_dict[k])
        return df["GroupID"].str.find(k) != -1
    conditions = list(map(fn, o_dict))

    df["New_column"] = numpy.select(conditions, values, default="NOT IN DIC")
    print(df)