---
name: "svmocas"
section: "1"
platform: "linux"
package: "libocas-tools"
version: "0.97+dfsg-8+b2"
release: "Debian 13 (Trixie)"
platforms: ["linux"]
url: "https://new.carta.tech/svmocas"
---

# svmocas(1)

train a binary linear SVM classifier

## NAME

svmocas - train a binary linear SVM classifier

## SYNOPSIS

**svmocas** \[*options*\] *example\_file* *model\_file*

## DESCRIPTION

**svmocas** is a program that trains a binary linear SVM classifier using the Optimized Cutting Plane Algorithm for Support Vector Machines (OCAS) and produces a model file.

*example\_file* is a file with training examples in SVM^light format, and *model\_file* is the file in which to store the learned linear rule **f(x)=w'\*x+w0**. *model\_file* contains **d** lines, where **d** is the number of data dimensions. The first n lines are coordinates of **w** and the last line is **w0**.

## OPTIONS

A summary of options is included below.

**General options:**

- **-h**

  Show summary of options.

- **-v *(0\|1)***

  Set the verbosity level (default: **1**)

**Learning options:**

- **-c *float***

  Regularization constant C. (default: **1**)

- **-C *constants\_file***

  If specified, each example has a different regularization constant, taken from the text file *constants\_file*. Each line of the text file must contain a single constant (positive double) for the corresponding example. If **-C** is used, then the **-c** option is ignored.

- **-b *(0\|1)***

  Value of the L2-bias feature. A value of 0 implies not having bias. (default: **0**)

- **-n *integer***

  Use only the first *integer* examples for training. By default, *integer* equals the number of examples in *example\_file*.

**Optimization options:**

- **-m *(0\|1)***

  Solver to be used:

  ```
  0 ... standard cutting plane (equivalent to BMRM, SVM^perf)

  1 ... OCAS (default)
  ```

- **-s *integer***

  Cache size for cutting planes. (default: **2000**)

- **-p *integer***

  Number of threads. (default: **1**)

**Stopping conditions:**

- **-a *float***

  Absolute tolerance TolAbs: halt if **QP-QD \<= TolAbs**. (default: **0**)

- **-r *float***

  Relative tolerance TolAbs: halt if **QP-QD \<= abs(QP)\*TolRel**. (default: **0.01**)

- **-q *float***

  Desired objective value QPValue: halt is **QP \<= QPValue**. (default: **0**)

- **-t *float***

  Halts if the solver time (loading time is not counted) exceeds the time given in seconds. (default: **infinity**)

## EXAMPLES

Train the binary SVM classifier from *riply\_trn.light*, with the regularization constant C=10, bias switched on, verbosity switched off, and save model to *svmocas.model*:

```
 svmocas -c 10 -b 1 -v 0 riply_trn.light svmocas.model
```

Compute the testing error of the classifier stored in *svmocas.model* with [**linclassif**(1)](/linclassif/1.md) using testing examples from *riply\_tst.light* and save the predicted labels to *riply\_tst.pred*:

```
 linclassif -e -o riply_tst.pred riply_tst.light svmocas.model
```

## SEE ALSO

[**msvmocas**(1)](/msvmocas/1.md), [**linclassif**(1)](/linclassif/1.md).

## AUTHORS

svmocas was written by Vojtech Franc \<xfrancv@cmp.felk.cvut.cz\> and Soeren Sonnenburg \<Soeren.Sonnenburg@tu-berlin.de\>.

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