为数据帧中的所有链接元素行分配相同的唯一ID

Question

我的数据帧如下：

    No     IsRenew  PrevNo
0   IAB19   TRUE    -
1   IAB25   FALSE   - 
2   IAB56   TRUE    IAB19
3   IAB22   TRUE    IAB56
4   IAB81   TRUE    IAB22
5   IAB82   TRUE    -
6   IAB89   FALSE   IAB82

我想为每个组生成唯一的ID。 例如

        No     UniqueID
0       IAB19   1
1       IAB25   2
2       IAB56   1
3       IAB22   1
4       IAB81   1
5       IAB82   3
6       IAB89   3

我应该如何按它们分组或合并/加入它们以获得以上输出

Answer 1

使用networkx和^{}作为字典，然后使用^{}作为新列：

import networkx as nx

# Create the graph from the dataframe
g = nx.Graph()

#replace - to df['No'] values
df['PrevNo'] = df['PrevNo'].mask(df['PrevNo'] == '-', df['No'])
# if - are missing values
#df['PrevNo'] = df['PrevNo'].fillna(df['No'])
g.add_edges_from(df[['No','PrevNo']].itertuples(index=False))

connected_components = nx.connected_components(g)

# Find the component id of the nodes
node2id = {}
for cid, component in enumerate(connected_components):
    for node in component:
        node2id[node] = cid + 1

df['UniqueID'] = df['No'].map(node2id)
print (df)
      No  IsRenew PrevNo  UniqueID
0  IAB19     True  IAB19         1
1  IAB25    False  IAB25         2
2  IAB56     True  IAB19         1
3  IAB22     True  IAB56         1
4  IAB81     True  IAB22         1
5  IAB82     True  IAB82         3
6  IAB89    False  IAB82         3

编辑：通过问题的第二次编辑-使用defaultdict来编辑列表：

d = defaultdict(list)
for cid, component in enumerate(connected_components):
    for node in component:
        d[cid + 1].append(node)

print (dict(d))
{1: ['IAB56', 'IAB19', 'IAB22', 'IAB81'], 2: ['IAB25'], 3: ['IAB89', 'IAB82']}