mirror of
https://github.com/ClickHouse/ClickHouse.git
synced 2024-11-21 23:21:59 +00:00
WIP update-aggregate-funcions-in-zh
This commit is contained in:
benbiti
parent
c1e5dc92a4
commit
c24207037f
@ -396,11 +396,6 @@ SELECT quantileTDigestWeighted(number, 1) FROM numbers(10)
|
||||
- [中位数](#median)
|
||||
- [分位数](#quantiles)
|
||||
|
||||
## quantiles(level1, level2, …)(x) {#quantiles}
|
||||
|
||||
所有分位数函数也有相应的函数: `quantiles`, `quantilesDeterministic`, `quantilesTiming`, `quantilesTimingWeighted`, `quantilesExact`, `quantilesExactWeighted`, `quantilesTDigest`。这些函数一次计算所列层次的所有分位数，并返回结果值的数组。
|
||||
|
||||
|
||||
## stochasticLinearRegression {#agg_functions-stochasticlinearregression}
|
||||
|
||||
该函数实现随机线性回归。 它支持自定义参数的学习率、L2正则化系数、微批，并且具有少量更新权重的方法（[Adam](https://en.wikipedia.org/wiki/Stochastic_gradient_descent#Adam) （默认）， [simple SGD](https://en.wikipedia.org/wiki/Stochastic_gradient_descent)， [Momentum](https://en.wikipedia.org/wiki/Stochastic_gradient_descent#Momentum)， [Nesterov](https://mipt.ru/upload/medialibrary/d7e/41-91.pdf)）。
|
||||
@ -473,57 +468,5 @@ evalMLMethod(model, param1, param2) FROM test_data
|
||||
- [stochasticLogisticRegression](#agg_functions-stochasticlogisticregression)
|
||||
- [线性回归和逻辑回归之间的区别](https://stackoverflow.com/questions/12146914/what-is-the-difference-between-linear-regression-and-logistic-regression)
|
||||
|
||||
## stochasticLogisticRegression {#agg_functions-stochasticlogisticregression}
|
||||
|
||||
该函数实现随机逻辑回归。 它可以用于二进制分类问题，支持与stochasticLinearRegression相同的自定义参数，并以相同的方式工作。
|
||||
|
||||
### 参数 {#agg_functions-stochasticlogisticregression-parameters}
|
||||
|
||||
参数与stochasticLinearRegression中的参数完全相同:
|
||||
`learning rate`, `l2 regularization coefficient`, `mini-batch size`, `method for updating weights`.
|
||||
欲了解更多信息，请参阅 [参数](#agg_functions-stochasticlinearregression-parameters).
|
||||
|
||||
``` text
|
||||
stochasticLogisticRegression(1.0, 1.0, 10, 'SGD')
|
||||
```
|
||||
|
||||
**1.** 安装
|
||||
|
||||
<!-- -->
|
||||
|
||||
参考stochasticLinearRegression相关文档
|
||||
|
||||
预测标签的取值范围为[-1, 1]
|
||||
|
||||
**2.** 预测
|
||||
|
||||
<!-- -->
|
||||
|
||||
使用已经保存的state我们可以预测标签为 `1` 的对象的概率。
|
||||
|
||||
``` sql
|
||||
WITH (SELECT state FROM your_model) AS model SELECT
|
||||
evalMLMethod(model, param1, param2) FROM test_data
|
||||
```
|
||||
|
||||
查询结果返回一个列的概率。注意 `evalMLMethod` 的第一个参数是 `AggregateFunctionState` 对象，接下来的参数是列的特性。
|
||||
|
||||
我们也可以设置概率的范围， 这样需要给元素指定不同的标签。
|
||||
|
||||
``` sql
|
||||
SELECT ans < 1.1 AND ans > 0.5 FROM
|
||||
(WITH (SELECT state FROM your_model) AS model SELECT
|
||||
evalMLMethod(model, param1, param2) AS ans FROM test_data)
|
||||
```
|
||||
|
||||
结果是标签。
|
||||
|
||||
`test_data` 是一个像 `train_data` 一样的表，但是不包含目标值。
|
||||
|
||||
**另请参阅**
|
||||
|
||||
- [随机指标线上回归](#agg_functions-stochasticlinearregression)
|
||||
- [线性回归和逻辑回归之间的差异](https://stackoverflow.com/questions/12146914/what-is-the-difference-between-linear-regression-and-logistic-regression)
|
||||
|
||||
|
||||
[原始文章](https://clickhouse.tech/docs/en/query_language/agg_functions/reference/) <!--hide-->
|
||||
|
||||
@ -4,40 +4,39 @@ toc_priority: 222
|
||||
|
||||
# stochasticLogisticRegression {#agg_functions-stochasticlogisticregression}
|
||||
|
||||
This function implements stochastic logistic regression. It can be used for binary classification problem, supports the same custom parameters as stochasticLinearRegression and works the same way.
|
||||
该函数实现随机逻辑回归。 它可以用于二进制分类问题，支持与stochasticLinearRegression相同的自定义参数，并以相同的方式工作。
|
||||
|
||||
### Parameters {#agg_functions-stochasticlogisticregression-parameters}
|
||||
### 参数 {#agg_functions-stochasticlogisticregression-parameters}
|
||||
|
||||
Parameters are exactly the same as in stochasticLinearRegression:
|
||||
参数与stochasticLinearRegression中的参数完全相同:
|
||||
`learning rate`, `l2 regularization coefficient`, `mini-batch size`, `method for updating weights`.
|
||||
For more information see [parameters](#agg_functions-stochasticlinearregression-parameters).
|
||||
欲了解更多信息，参见 [参数] (#agg_functions-stochasticlinearregression-parameters).
|
||||
|
||||
``` text
|
||||
``` sql
|
||||
stochasticLogisticRegression(1.0, 1.0, 10, 'SGD')
|
||||
```
|
||||
|
||||
**1.** Fitting
|
||||
**1.** 拟合
|
||||
|
||||
<!-- -->
|
||||
|
||||
See the `Fitting` section in the [stochasticLinearRegression](#stochasticlinearregression-usage-fitting) description.
|
||||
参考[stochasticLinearRegression](#stochasticlinearregression-usage-fitting) `拟合` 章节文档。
|
||||
|
||||
Predicted labels have to be in \[-1, 1\].
|
||||
预测标签的取值范围为\[-1, 1\]
|
||||
|
||||
**2.** Predicting
|
||||
**2.** 预测
|
||||
|
||||
<!-- -->
|
||||
|
||||
Using saved state we can predict probability of object having label `1`.
|
||||
|
||||
使用已经保存的state我们可以预测标签为 `1` 的对象的概率。
|
||||
``` sql
|
||||
WITH (SELECT state FROM your_model) AS model SELECT
|
||||
evalMLMethod(model, param1, param2) FROM test_data
|
||||
```
|
||||
|
||||
The query will return a column of probabilities. Note that first argument of `evalMLMethod` is `AggregateFunctionState` object, next are columns of features.
|
||||
查询结果返回一个列的概率。注意 `evalMLMethod` 的第一个参数是 `AggregateFunctionState` 对象，接下来的参数是列的特性。
|
||||
|
||||
We can also set a bound of probability, which assigns elements to different labels.
|
||||
我们也可以设置概率的范围， 这样需要给元素指定不同的标签。
|
||||
|
||||
``` sql
|
||||
SELECT ans < 1.1 AND ans > 0.5 FROM
|
||||
@ -45,11 +44,11 @@ stochasticLogisticRegression(1.0, 1.0, 10, 'SGD')
|
||||
evalMLMethod(model, param1, param2) AS ans FROM test_data)
|
||||
```
|
||||
|
||||
Then the result will be labels.
|
||||
结果是标签。
|
||||
|
||||
`test_data` is a table like `train_data` but may not contain target value.
|
||||
`test_data` 是一个像 `train_data` 一样的表，但是不包含目标值。
|
||||
|
||||
**See Also**
|
||||
**参见**
|
||||
|
||||
- [stochasticLinearRegression](../../../sql-reference/aggregate-functions/reference/stochasticlinearregression.md#agg_functions-stochasticlinearregression)
|
||||
- [Difference between linear and logistic regressions.](https://stackoverflow.com/questions/12146914/what-is-the-difference-between-linear-regression-and-logistic-regression)
|
||||
- [随机指标线性回归](../../../sql-reference/aggregate-functions/reference/stochasticlinearregression.md#agg_functions-stochasticlinearregression)
|
||||
- [线性回归和逻辑回归之间的差异](https://stackoverflow.com/questions/12146914/what-is-the-difference-between-linear-regression-and-logistic-regression)
|
||||
|
||||
Loading…
Reference in New Issue
Block a user