.. _TrainDimensionalityReduction: 

TrainDimensionalityReduction
============================





Train a dimensionality reduction model

Description
-----------

Trainer for dimensionality reduction algorithms (autoencoders, PCA, SOM). All input samples are used to compute the model, like other machine learning models.
The model can be used in the :ref:`ImageDimensionalityReduction` and :ref:`VectorDimensionalityReduction` applications.



Parameters
----------

.. contents:: :local:

.. |br| raw:: html

   <br />

.. |em| raw:: html

   &emsp;

Input and output data
^^^^^^^^^^^^^^^^^^^^^



This group of parameters allows setting input and output data.

**Input Vector Data** :code:`-io.vd vectorfile` *Mandatory* |br|
Input geometries used for training (note: all geometries from the layer will be used)

**Output model** :code:`-io.out filename [dtype]` *Mandatory* |br|
Output file containing the estimated model (.txt format).

**Input XML image statistics file** :code:`-io.stats filename [dtype]` |br|
XML file containing mean and variance of each feature.



------------

**Field names to be used for training** :code:`-feat string1 string2...` |br|
List of field names in the input vector data used as features for training.

**algorithm to use for the training** :code:`-algorithm [som|autoencoder|pca]` *Default value: som* |br|
Choice of the dimensionality reduction algorithm to use for the training.

* **OTB SOM** |br| This group of parameters allows setting SOM parameters. 

* **Shark Autoencoder** |br| This group of parameters allows setting Shark autoencoder parameters. 

* **Shark PCA** |br| This group of parameters allows setting Shark PCA parameters. 



OTB SOM options
^^^^^^^^^^^^^^^

**Map size** :code:`-algorithm.som.s string1 string2...` *Default value: 10 10* |br|
Sizes of the SOM map (one per dimension). For instance, [12;15] means a 2D map of size 12x15. Support2D to 5D maps.

**Neighborhood sizes** :code:`-algorithm.som.n string1 string2...` *Default value: 3 3* |br|
Sizes of the initial neighborhood in the SOM map (one per dimension). The number of sizes should be the same as the map sizes

**NumberIteration** :code:`-algorithm.som.ni int` *Default value: 5* |br|
Number of iterations for SOM learning

**BetaInit** :code:`-algorithm.som.bi float` *Default value: 1* |br|
Initial learning coefficient

**BetaFinal** :code:`-algorithm.som.bf float` *Default value: 0.1* |br|
Final learning coefficient

**InitialValue** :code:`-algorithm.som.iv float` *Default value: 10* |br|
Maximum initial neuron weight

Shark Autoencoder options
^^^^^^^^^^^^^^^^^^^^^^^^^

**Maximum number of iterations during training** :code:`-algorithm.autoencoder.nbiter int` *Default value: 100* |br|
The maximum number of iterations used during training.

**Maximum number of iterations during training** :code:`-algorithm.autoencoder.nbiterfinetuning int` *Default value: 0* |br|
The maximum number of iterations used during fine tuning of the whole network.

**Epsilon** :code:`-algorithm.autoencoder.epsilon float` *Default value: 0* |br|
Epsilon

**Weight initialization factor** :code:`-algorithm.autoencoder.initfactor float` *Default value: 1* |br|
Parameter that control the weight initialization of the autoencoder

**Size** :code:`-algorithm.autoencoder.nbneuron string1 string2...` *Mandatory* |br|
The number of neurons in each hidden layer.

**Strength of the regularization** :code:`-algorithm.autoencoder.regularization string1 string2...` *Mandatory* |br|
Strength of the L2 regularization used during training

**Strength of the noise** :code:`-algorithm.autoencoder.noise string1 string2...` *Mandatory* |br|
Strength of the noise

**Sparsity parameter** :code:`-algorithm.autoencoder.rho string1 string2...` *Mandatory* |br|
Sparsity parameter

**Sparsity regularization strength** :code:`-algorithm.autoencoder.beta string1 string2...` *Mandatory* |br|
Sparsity regularization strength

**Learning curve** :code:`-algorithm.autoencoder.learningcurve filename [dtype]` |br|
Learning error values

Shark PCA options
^^^^^^^^^^^^^^^^^

**Dimension of the output of the pca transformation** :code:`-algorithm.pca.dim int` *Default value: 10* |br|
Dimension of the output of the pca transformation.



------------

**Available RAM (MB)** :code:`-ram int` *Default value: 256* |br|
Available memory for processing (in MB).



Examples
--------

From the command-line:

.. code-block:: bash

    otbcli_TrainDimensionalityReduction -io.vd cuprite_samples.sqlite -io.out model.som -algorithm som -algorithm.som.s 10 10 -algorithm.som.n 3 3 -algorithm.som.ni 5 -algorithm.som.bi 1 -algorithm.som.bf 0.1 -algorithm.som.iv 10 -feat value_0 value_1 value_2 value_3 value_4 value_5 value_6 value_7 value_8 value_9


From Python:

.. code-block:: python

	import otbApplication

	app = otbApplication.Registry.CreateApplication("TrainDimensionalityReduction")

	app.SetParameterString("io.vd", "cuprite_samples.sqlite")
	app.SetParameterString("io.out", "model.som")
	app.SetParameterString("algorithm","som")
	app.SetParameterStringList("algorithm.som.s", ['10', '10'])
	app.SetParameterStringList("algorithm.som.n", ['3', '3'])
	app.SetParameterInt("algorithm.som.ni", 5)
	app.SetParameterFloat("algorithm.som.bi", 1)
	app.SetParameterFloat("algorithm.som.bf", 0.1)
	app.SetParameterFloat("algorithm.som.iv", 10)
	app.SetParameterStringList("feat", ['value_0', 'value_1', 'value_2', 'value_3', 'value_4', 'value_5', 'value_6', 'value_7', 'value_8', 'value_9'])

	app.ExecuteAndWriteOutput()





See also
--------

| :ref:`ImageDimensionalityReduction`, :ref:`VectorDimensionalityReduction`