---
name: "liblinear-train"
section: "1"
platform: "linux"
package: "liblinear-tools"
version: "2.3.0+dfsg-5+b2"
release: "Debian 13 (Trixie)"
platforms: ["linux"]
url: "https://new.carta.tech/liblinear-train"
---

# liblinear-train(1)

train a linear classifier and produce a model

## NAME

liblinear-train - train a linear classifier and produce a model

## SYNOPSIS

**liblinear-train** \[*options*\] *training\_set\_file* \[*model\_file*\]\

## DESCRIPTION

**liblinear-train** trains a linear classifier using liblinear and produces a model suitable for use with [**liblinear-predict**(1)](/liblinear-predict/1.md).

*training\_set\_file* is the file containing the data used for training. *model\_file* is the file to which the model will be saved. If *model\_file* is not provided, it defaults to *training\_set\_file.model*.

To obtain good performances, sometimes one needs to scale the data. This can be done with [**svm-scale**(1)](/svm-scale/1.md).

## OPTIONS

A summary of options is included below.

- **-s *type***

  Set the type of the solver (default **1**):

  ```
  for multi-class classification
  ```

  ```
  0 ... L2-regularized logistic regression (primal)

  1 ... L2-regularized L2-loss support vector classification (dual)

  2 ... L2-regularized L2-loss support vector classification (primal)

  3 ... L2-regularized L1-loss support vector classification (dual)

  4 ... multi-class support vector classification

  5 ... L1-regularized L2-loss support vector classification

  6 ... L1-regularized logistic regression

  7 ... L2-regularized logistic regression (dual)

  ```

  ```
  for regression
  ```

  ```

  11 ... L2-regularized L2-loss support vector regression (primal)

  12 ... L2-regularized L2-loss support vector regression (dual)

  13 ... L2-regularized L1-loss support vector regression (dual)
  ```

- **-c *cost***

  Set the parameter C (default: **1**)

- **-p *epsilon***

  Set the epsilon in loss function of epsilon-SVR (default: **0.1**)

- **-e *epsilon***

  Set the tolerance of the termination criterion

  \-s 0 and 2:

  ```

  |f'(w)|_2 <= epsilon*min(pos,neg)/l*|f'(w0)|_2, where f is
  the primal function and pos/neg are the number of positive/negative data
  (default: 0.01)
  ```

-

  \-s 11:

  ```
  |f'(w)|_2 <= epsilon*|f'(w0)|_2 (default 0.0001)
  ```

-

  \-s 1, 3, 4 and 7:

  ```
  Dual maximal violation <= epsilon; similar to libsvm (default: 0.1)
  ```

-

  \-s 5 and 6:

  ```
  |f'(w)|_inf <= epsilon*min(pos,neg)/l*|f'(w0)|_inf, where f is the primal
  function (default: 0.01)
  ```

-

  \-s 12 and 13:

  ```
  |f'(alpha)|_1 <= epsilon |f'(alpha0)|, where f is the dual function (default 0.1)
  ```

- **-B *bias***

  If *bias* \>= 0, then instance x becomes \[x; bias\]; if *bias* \< 0, then no bias term is added (default: **-1**)

- **-w*i* *weight***

  Weights adjust the parameter C of different classes (see README for details)

- **-v *n***

  *n*-fold cross validation mode

- **-C**

  Find parameters (C for -s 0, 2 and C, p for -s 11)

- **-q**

  Quiet mode (no outputs).

  Option -v randomly splits the data into n parts and calculates cross validation accuracy on them.

  Option -C conducts cross validation under different parameters and finds the best one. This option is supported only by -s 0, -s 2 (for finding C) and -s 11 (for finding C, p). If the solver is not specified, -s 2 is used.

## EXAMPLES

Train a linear SVM using L2-loss function:

```
 liblinear-train data_file
```

Train a logistic regression model:

```
 liblinear-train -s 0 data_file
```

Do five-fold cross-validation using L2-loss SVM, using a smaller stopping tolerance 0.001 instead of the default 0.1 for more accurate solutions:

```
 liblinear-train -v 5 -e 0.001 data_file
```

Conduct cross validation many times by L2-loss SVM and find the parameter C which achieves the best cross validation accuracy:

```
 train -C datafile
```

For parameter selection by -C, users can specify other solvers (currently -s 0, -s 2 and -s 11 are supported) and different number of CV folds. Further, users can use the -c option to specify the smallest C value of the search range. This option is useful when users want to rerun the parameter selection procedure from a specified C under a different setting, such as a stricter stopping tolerance -e 0.0001 in the above example. Similarly, for -s 11, users can use the -p option to specify the maximal p value of the search range.

```
 train -C -s 0 -v 3 -c 0.5 -e 0.0001 datafile
```

Train four classifiers:

positive    negative        Cp  Cn\
 class 1     class 2,3,4     20  10\
 class 2     class 1,3,4     50  10\
 class 3     class 1,2,4     20  10\
 class 4     class 1,2,3     10  10

```
 liblinear-train -c 10 -w1 2 -w2 5 -w3 2 four_class_data_file
```

If there are only two classes, we train ONE model. The C values for the two classes are 10 and 50:

```
 liblinear-train -c 10 -w3 1 -w2 5 two_class_data_file
```

Output probability estimates (for logistic regression only) using [**liblinear-predict**(1)](/liblinear-predict/1.md):

```
 liblinear-predict -b 1 test_file data_file.model output_file
```

## SEE ALSO

[**liblinear-predict**(1)](/liblinear-predict/1.md), [**svm-predict**(1)](/svm-predict/1.md), [**svm-train**(1)](/svm-train/1.md), [**svm-scale**(1)](/svm-scale/1.md)

## AUTHORS

liblinear-train was written by the LIBLINEAR authors at National Taiwan university for the LIBLINEAR Project.

This manual page was written by Christian Kastner \<ckk@debian.org\> for the Debian project (and may be used by others).