Hendrik/xgboost
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

text vignette

Browse Source
This commit is contained in:
El Potaeto 2015-02-22 00:17:37 +01:00
parent 56e9bff11f
commit 56068b5453
2 changed files with 16 additions and 12 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
R-package/vignettes/vignette.css
View File

@ -25,7 +25,7 @@ body{
line-height: 1; line-height: 1;
max-width: 800px; max-width: 800px;
padding: 20px; padding: 10px;
font-size: 17px; font-size: 17px;
text-align: justify; text-align: justify;
text-justify: inter-word; text-justify: inter-word;
@ -33,9 +33,10 @@ body{
p { p {
line-height: 150%; line-height: 140%;
/ max-width: 540px; / max-width: 540px;
max-width: 960px; max-width: 960px;
margin-bottom: 5px;
font-weight: 400; font-weight: 400;
/ color: #333333 / color: #333333
} }
@ -46,7 +47,7 @@ h1, h2, h3, h4 {
font-weight: 400; font-weight: 400;
} }
h2, h3, h4, h5, p { h2, h3, h4, h5 {
margin-bottom: 20px; margin-bottom: 20px;
padding: 0; padding: 0;
} }
@ -86,6 +87,7 @@ h6 {
font-variant:small-caps; font-variant:small-caps;
font-style: italic; font-style: italic;
} }
a { a {
color: #606AAA; color: #606AAA;
margin: 0; margin: 0;
@ -101,6 +103,7 @@ a:hover {
a:visited { a:visited {
color: gray; color: gray;
} }
ul, ol { ul, ol {
padding: 0; padding: 0;
margin: 0px 0px 0px 50px; margin: 0px 0px 0px 50px;
@ -138,9 +141,10 @@ code {
} }
p code { li code, p code {
background: #CDCDCD; background: #CDCDCD;
color: #606AAA; color: #606AAA;
padding: 0px 5px 0px 5px;
} }
code.r, code.cpp { code.r, code.cpp {

16
R-package/vignettes/xgboostPresentation.Rmd
View File

@ -22,8 +22,8 @@ This is an introductory document for using the \verb@xgboost@ package in *R*.
It is an efficient and scalable implementation of gradient boosting framework by @friedman2001greedy. Two solvers are included: It is an efficient and scalable implementation of gradient boosting framework by @friedman2001greedy. Two solvers are included:
- *linear model* - *linear* model ;
- *tree learning* algorithm - *tree learning* algorithm.
It supports various objective functions, including *regression*, *classification* and *ranking*. The package is made to be extendible, so that users are also allowed to define their own objective function easily. It supports various objective functions, including *regression*, *classification* and *ranking*. The package is made to be extendible, so that users are also allowed to define their own objective function easily.
@ -48,7 +48,7 @@ Installation
The first step is to install the package. The first step is to install the package.
For up-to-date version (which is *highly* recommended), install from Github: For up-to-date version (which is *highly* recommended), install from *Github*:
```{r installGithub, eval=FALSE} ```{r installGithub, eval=FALSE}
devtools::install_github('tqchen/xgboost',subdir='R-package') devtools::install_github('tqchen/xgboost',subdir='R-package')
@ -56,7 +56,7 @@ devtools::install_github('tqchen/xgboost',subdir='R-package')
> *Windows* user will need to install [RTools](http://cran.r-project.org/bin/windows/Rtools/) first. > *Windows* user will need to install [RTools](http://cran.r-project.org/bin/windows/Rtools/) first.
For stable version on CRAN, run: For stable version on *CRAN*, run:
```{r installCran, eval=FALSE} ```{r installCran, eval=FALSE}
install.packages('xgboost') install.packages('xgboost')
@ -194,11 +194,11 @@ print(paste("test-error=", err))
> We remind you that the algorithm has never seen the `test` data before. > We remind you that the algorithm has never seen the `test` data before.
Here, we have just computed a simple metric: the average error: Here, we have just computed a simple metric, the average error.
* `as.numeric(pred > 0.5)` applies our rule that when the probability (== prediction == regression) is over `0.5` the observation is classified as `1` and `0` otherwise ; 1. `as.numeric(pred > 0.5)` applies our rule that when the probability (== prediction == regression) is over `0.5` the observation is classified as `1` and `0` otherwise ;
* `probabilityVectorPreviouslyComputed != test$label` computes the vector of error between true data and computed probabilities ; 2. `probabilityVectorPreviouslyComputed != test$label` computes the vector of error between true data and computed probabilities ;
* `mean(vectorOfErrors)` computes the average error itself. 3. `mean(vectorOfErrors)` computes the average error itself.
The most important thing to remember is that **to do a classification basically, you just do a regression and then apply a threeshold**. The most important thing to remember is that **to do a classification basically, you just do a regression and then apply a threeshold**.