Question

我有一个组ID列表：

letters = ['A', 'A/D', 'B', 'B/D', 'C', 'C/D', 'D']

和组的数据框：

groups = pd.DataFrame({'group': ['B', 'A/D', 'D', 'D', 'A']})

我想在数据框中创建一列，以提供组ID在列表中的位置，如下所示：

  group group_idx
0     B         2
1   A/D         1
2     D         6
3     D         6
4     A         0

我当前的解决方案是：

group_to_num = {hsg: i for i, hsg in enumerate(letters)}
groups['group_idx'] = groups.applymap(lambda x: group_to_num.get(x)).max(axis=1).fillna(-1).astype(np.int32)

但是看起来不太优雅。有没有更简单的方法？

Answer 1

使用map：

WordCountMapper.java:23: error: <identifier> expected
    Public void map(LongWritable key, Text value, Context context) throws IOException, InterruptedException {
          ^
WordCountMapper.java:23: error: invalid method declaration; return type required
    Public void map(LongWritable key, Text value, Context context) throws IOException, InterruptedException {
                ^
WordCountMapper.java:25: error: ';' expected
        String line = value.toString()
                                      ^
WordCountMapper.java:27: error: ';' expected
        StringTokenizer tokenizer = new StringTokenizer(line)
                                                             ^
WordCountMapper.java:35: error: illegal start of expression
    public void run(Context context) throws IOException, InterruptedException {
    ^
WordCountMapper.java:35: error: illegal start of expression
    public void run(Context context) throws IOException, InterruptedException {
           ^
WordCountMapper.java:35: error: ';' expected
    public void run(Context context) throws IOException, InterruptedException {
                   ^
WordCountMapper.java:35: error: ';' expected
    public void run(Context context) throws IOException, InterruptedException {
                                   ^
WordCountMapper.java:35: error: not a statement
    public void run(Context context) throws IOException, InterruptedException {
                                            ^
WordCountMapper.java:35: error: ';' expected
    public void run(Context context) throws IOException, InterruptedException {
                                                       ^
WordCountMapper.java:35: error: not a statement
    public void run(Context context) throws IOException, InterruptedException {
                                                         ^
WordCountMapper.java:35: error: ';' expected
    public void run(Context context) throws IOException, InterruptedException {
                                                                               ^

输出

import pandas as pd

letters = ['A', 'A/D', 'B', 'B/D', 'C', 'C/D', 'D']
group_to_num = {hsg: i for i, hsg in enumerate(letters)}
groups = pd.DataFrame({'group': ['B', 'A/D', 'D', 'D', 'A']})

groups['group_idx'] = groups.group.map(group_to_num)
print(groups)

Answer 2

您可以在数据框构造函数之后尝试合并：

groups.merge(pd.DataFrame(letters).reset_index(),left_on='group',right_on=0).\
           rename(columns={'index':'group_idx'}).drop(0,1)

   group  group_idx
0     B        2
1   A/D        1
2     D        6
3     D        6
4     A        0

熊猫系列或列表索引

2 个答案: