熊猫到嵌套 YAML答案

【问题标题】：Pandas to nested YAML熊猫到嵌套 YAML
【发布时间】：2022-01-20 10:53:54
【问题描述】：

我正在尝试构建 yaml 内容并将其转储到文件中：

它应该是这样的

group:
  Base Configuration:
    id: 1
    group_name: Base Configuration
    group_members:
    - S3-WLP-Kafka-topics-test
    - S3-WLP-BATCH-API-503-P
    group_description: Lorem Ipsum is simply dummy text
  Entity Persistence:
    id: 2
    group_name: Entity Persistence
    group_members:
    - Entity-INPUT-DATA-PERSISTENCE
    - Entity-OUTPUT-DATA-PERSISTENCE
    group_description: Lorem Ipsum is not simply random text.
  API TESTS:
    id: 3
    group_name: API TESTS
    group_members:
    - API Configuration
    group_description: Standard chunk of Lorem Ipsum

我的 pandas 数据框如下所示：

id	group_name	group_members	group_description	level
1	Base Configuration	['S3-WLP-Kafka-topics-test', 'S3-WLP-BATCH-API-503-P']	Lorem Ipsum is simply dummy text	group
2	Entity Persistence	['Entity-INPUT-DATA-PERSISTENCE', 'Entity-OUTPUT-DATA-PERSISTENCE']	Lorem Ipsum is not simply random text.	group
3	API TESTS	['API Configuration']	Standard chunk of Lorem Ipsum	group

这是我实现这一目标的尝试。

        x = self.df_merged['level'].tolist()
        y = self.df_merged['group_name'].tolist()
        z = self.df_merged['id'].tolist()
        a = self.df_merged['group_description'].tolist()
        self.df_merged = self.df_merged.set_index(['level', 'group_name', 'id', 'group_description'])
        self.df_merged = self.df_merged.reindex(pd.MultiIndex.from_tuples(zip(x, y, z, a)))
        nested_dict = collections.defaultdict(dict)
        for keys, value in self.df_merged.group_members.iteritems():
            nested_dict[keys[0]][keys[1]] = value

这给了我：

defaultdict(<class 'dict'>, {'group': {'Base Configuration': ['S3-WLP-Kafka-topics-test', 'S3-WLP-BATCH-API-503-P'], 'Entity Persistence': ['Entity-INPUT-DATA-PERSISTENCE', 'Entity-OUTPUT-DATA-PERSISTENCE'], 'API TESTS': ['API Configuration']}})

虽然这不是我想要的，我的尝试也没有成功

【问题讨论】：

标签： python pandas dictionary multi-index

【解决方案1】：

如果其他人也在寻找相同的东西，我终于用groupby()和to_dict()实现了它

data = self.df_unchanged.groupby('level')[['group_name', 'id', 'group_description', 'group_members']].apply(
            lambda x: x.set_index('group_name').to_dict(orient='index')).to_dict()

最后转储到 yaml 文件：

    with open("test.yaml", "w") as yaml_file:
        yaml_file.write(yaml.dump(data, allow_unicode=True, default_flow_style=False, sort_keys=False))

【讨论】：