ClickHouse/docs/zh/sql-reference/aggregate-functions/reference/categoricalinformationvalue.md

14 lines
382 B
Markdown
Raw Normal View History

2021-02-02 12:00:41 +00:00
---
toc_priority: 250
---
# categoricalInformationValue {#categoricalinformationvalue}
2021-02-03 15:22:18 +00:00
对于每个类别计算 `(P(tag = 1) - P(tag = 0))(log(P(tag = 1)) - log(P(tag = 0)))`
2021-02-02 12:00:41 +00:00
``` sql
categoricalInformationValue(category1, category2, ..., tag)
```
2021-02-03 15:22:18 +00:00
结果指示离散(分类)要素如何使用 `[category1, category2, ...]` 有助于使用学习模型预测`tag`的值。