https://github.com/wwbrannon/sqlscore

R utilities for generating SQL queries from model objects
https://github.com/wwbrannon/sqlscore

glm r sql

Last synced: 9 days ago
JSON representation

R utilities for generating SQL queries from model objects

• Host: GitHub
• URL: https://github.com/wwbrannon/sqlscore
• Owner: wwbrannon
• License: other
• Created: 2016-12-07T23:11:23.000Z (almost 8 years ago)
• Default Branch: master
• Last Pushed: 2019-03-17T16:48:36.000Z (over 5 years ago)
• Last Synced: 2024-09-28T22:24:35.635Z (about 1 month ago)
• Topics: glm, r, sql
• Language: R
• Size: 152 KB
• Stars: 13
• Watchers: 6
• Forks: 4
• Open Issues: 1
• Metadata Files:
  • Readme: README.Rmd
  • License: LICENSE

Awesome Lists containing this project

README

        


```{r set-options, echo=FALSE, cache=FALSE}

library(sqlscore)

library(knitr)

options(width=80)

knitr::opts_chunk$set(

  collapse = TRUE,

  comment = "#>",

  fig.path = "README-",

  warning = FALSE,

  message = FALSE,

  echo = TRUE,

  tidy = TRUE,

  size="small",

  linewidth=80

)

# From https://github.com/yihui/knitr-examples/blob/master/077-wrap-output.Rmd

hook_output = knit_hooks$get('output')

knit_hooks$set(output = function(x, options) {

  # this hook is used only when the linewidth option is not NULL

  if (!is.null(n <- options$linewidth)) {

    x = knitr:::split_lines(x)

    # any lines wider than n should be wrapped

    if (any(nchar(x) > n)) x = strwrap(x, width = n)

    x = paste(x, collapse = '\n')

  }

  hook_output(x, options)

})

```

[![CRAN_Status_Badge](https://www.r-pkg.org/badges/version/sqlscore)](https://cran.r-project.org/package=sqlscore)

[![Build Status](https://img.shields.io/travis/wwbrannon/sqlscore.svg?style=flat)](https://travis-ci.org/wwbrannon/sqlscore)

[![Coverage Status](https://codecov.io/gh/wwbrannon/sqlscore/branch/master/graph/badge.svg)](https://codecov.io/github/wwbrannon/sqlscore?branch=master)

[![Downloads](https://cranlogs.r-pkg.org/badges/sqlscore)](https://cran.r-project.org/package=sqlscore)

[![License](https://img.shields.io/:license-mit-blue.svg?style=flat)](https://wwbrannon.mit-license.org/)

# sqlscore

Utilities for scoring GLMs and related models in SQL. Use the create_statement

and select_statement functions to generate scoring queries from model objects.

The most important use case is for very large scoring datasets, especially those

which can't fit into memory or would require too much network or storage I/O if

scored the usual way in R.

The SQL-generating functions in sqlscore handle various formula operators, and

also take care of wrapping the model's linear predictor in the appropriate link

function. If needed, you can also specify a custom link function.

## Usage:

The SQL-generating functions create\_statement and select\_statement do what their

names suggest and generate CREATE TABLE and SELECT statements for model scoring.

If, for example, you have a database table of iris measurements, you can model the

sepal length and generate predictions as follows:

```

mod <- glm(Sepal.Length ~ Sepal.Width + Petal.Length + Petal.Width + Species,

           data=datasets::iris)

create_statement(mod, src_table="iris", dest_table="iris_scores", pk=c("id"))

```

```{r, echo = FALSE}

mod <- glm(Sepal.Length ~ Sepal.Width + Petal.Length + Petal.Width + Species,

           data=datasets::iris)

create_statement(mod, src_table="iris", dest_table="iris_scores")  

```

To get a SELECT statement that's not wrapped in a CREATE TABLE (so that, e.g.,

you can add your own database-specific pieces of SQL), use select_statement:

```

mod <- glm(Sepal.Length ~ Sepal.Width + Petal.Length + Petal.Width + Species,

           data=datasets::iris)

select_statement(mod, src_table="iris", pk=c("id"))

```

```{r, echo = FALSE}

mod <- glm(Sepal.Length ~ Sepal.Width + Petal.Length + Petal.Width + Species,

           data=datasets::iris)

select_statement(mod, src_table="iris")

```

Helper functions include linpred(), which generates an R call object representing

the linear predictor, and score_expression, an S3 generic that handles wrapping

the linear predictor in the response function.

## Supported model types:

Specific packages and models that are known to work include: glm and lm from

package:stats, cv.glmnet from package:glmnet, glmboost from package:mboost,

and bayesglm from package:arm.

Default S3 methods are for objects structured like those of class "glm", so

models not listed here may work if they resemble those objects, but are not 

guaranteed to.

## Installation:

Install the released version from CRAN:

```

install.packages('sqlscore')

```

Install the dev version from github:

```

install.packages("devtools")

devtools::install_github("wwbrannon/sqlscore")

```