---
name: "otbgui_TrainDimensionalityReduction"
section: "1"
platform: "linux"
package: "otb-bin-qt"
version: "6.6.1+dfsg-1+b1"
release: "Debian 10 (Buster)"
platforms: ["linux"]
url: "https://new.carta.tech/otbgui_TrainDimensionalityReduction"
---

# otbgui\_TrainDimensionalityReduction(1)

OTB TrainDimensionalityReduction application

## NAME

otbgui\_TrainDimensionalityReduction - OTB TrainDimensionalityReduction application

## DESCRIPTION

This is the Train Dimensionality Reduction (TrainDimensionalityReduction) application, version 6.6.0

Train a dimensionality reduction model Parameters:

- **-io**

  \<group\>          Input and output data

**-io**.vd                                  \<string\>         Input Vector Data  (mandatory) **-io**.out                                 \<string\>         Output model  (mandatory)

- **-io**.stats

  \<string\>         Input XML image statistics file  (optional, off by default)

**-feat**                                   \<string list\>    Field names to be used for training.  (mandatory)

- **-algorithm**

  \<string\>         algorithm to use for the training \[som/autoencoder/pca\] (mandatory, default value is som)

- **-algorithm**.som.s

  \<string list\>    Map size  (optional, off by default, default value is 10

10 )

- **-algorithm**.som.n

  \<string list\>    Neighborhood sizes  (optional, off by default, default value is 3

3 )

- **-algorithm**.som.ni

  \<int32\>          NumberIteration  (optional, off by default, default value is 5)

- **-algorithm**.som.bi

  \<float\>          BetaInit  (optional, off by default, default value is 1)

- **-algorithm**.som.bf

  \<float\>          BetaFinal  (optional, off by default, default value is 0.1)

- **-algorithm**.som.iv

  \<float\>          InitialValue  (optional, off by default, default value is 10)

- **-algorithm**.autoencoder.nbiter

  \<int32\>          Maximum number of iterations during training  (mandatory, default value is 100)

- **-algorithm**.autoencoder.nbiterfinetuning \<int32\>

  Maximum number of iterations during training  (mandatory, default value is 0)

- **-algorithm**.autoencoder.epsilon

  \<float\>          Epsilon  (mandatory, default value is 0)

- **-algorithm**.autoencoder.initfactor

  \<float\>          Weight initialization factor  (mandatory, default value is 1)

- **-algorithm**.autoencoder.nbneuron

  \<string list\>    Size  (mandatory)

- **-algorithm**.autoencoder.regularization

  \<string list\>    Strength of the regularization  (mandatory)

- **-algorithm**.autoencoder.noise

  \<string list\>    Strength of the noise  (mandatory)

- **-algorithm**.autoencoder.rho

  \<string list\>    Sparsity parameter  (mandatory)

- **-algorithm**.autoencoder.beta

  \<string list\>    Sparsity regularization strength  (mandatory)

- **-algorithm**.autoencoder.learningcurve

  \<string\>         Learning curve  (optional, off by default)

- **-algorithm**.pca.dim

  \<int32\>          Dimension of the output of the pca transformation  (mandatory, default value is 10)

- **-ram**

  \<int32\>          Available RAM (Mb)  (optional, off by default, default value is 128)

- **-inxml**

  \<string\>         Load otb application from xml file  (optional, off by default)

- **-progress**

  \<boolean\>        Report progress

- **-help**

  \<string list\>    Display long help (empty list), or help for given parameters keys

Use **-help** param1 \[... paramN\] to see detailed documentation of those parameters.

## EXAMPLES

otbgui\_TrainDimensionalityReduction -io.vd cuprite\_samples.sqlite -io.out mode.ae -algorithm pca -algorithm.pca.dim 8 -feat value\_0 value\_1 value\_2 value\_3 value\_4 value\_5 value\_6 value\_7 value\_8 value\_9