July 30, 2023

tSVMModel – Docs for ESB 7.x


Generates an SVM-based classifier model that can be used by tPredict to classify given elements.

tSVMModel applies
the SVM algorithm to analyze feature vectors typically prepared and provided by

It generates a binary classification model out of this analysis and
writes this model in memory or in a file system supported by this
component, such as HDFS or S3.

tSVMModel properties for Apache Spark Batch

These properties are used to configure tSVMModel running in the Spark Batch Job framework.

The Spark Batch
tSVMModel component belongs to the Machine Learning family.

This component is available in Talend Platform products with Big Data and
in Talend Data Fabric.

Basic settings

Schema and Edit schema

A schema is a row description. It defines the number of fields
(columns) to be processed and passed on to the next component. When you create a Spark
Job, avoid the reserved word line when naming the

Click Edit
to make changes to the schema. If the current schema is of the Repository type, three options are available:

  • View schema: choose this
    option to view the schema only.

  • Change to built-in property:
    choose this option to change the schema to Built-in for local changes.

  • Update repository connection:
    choose this option to change the schema stored in the repository and decide whether
    to propagate the changes to all the Jobs upon completion. If you just want to
    propagate the changes to the current Job, you can select No upon completion and choose this schema metadata
    again in the Repository Content


Select the input column used to provide classification labels. The records of this column
are used as the class names (Target in terms of classification) of the elements to be

Since a SVM model is a binary classification model, only two classes are expected, that is to
say, only two distinct values are expected from this column.

Vector to process

Select the input column used to provide features. Very often, this column is the output of
the feature engineering computations performed by tModelEncoder.

Save the model on file

Select this check box to store the model in a given file system. Otherwise, the model is
stored in memory. The button for browsing does not work with the Spark Local mode; if you are using the Spark Yarn
or the Spark Standalone mode, ensure that you have properly
configured the connection in a configuration component in the same Job, such as tHDFSConfiguration.

Step size

Enter the size (numerical value) of the initial step of the gradient
descent calculation. The default value 1.0 means that the whole data set is taken.

Selecting the best step size is often delicate in practice.

Generally speaking, when the feature points to be analyzed are very
muddled, it is recommended to increase the step size in order to cover
enough number of points in each iteration; however, bear in mind that
too large a step size can improperly increase the time of each

On the other hand, the smaller the step size is, the more slowly the
convergence occurs and a more accurate model you can expect.

Number of iterations

Enter the number of iterations you want the Job to perform to train the model.

Fraction of data to be used per

Enter the fraction (expressed in decimal) of the input data to be used
in each iteration to calculate the gradient.

The default value 1.0 means that the whole data set is taken.

Regularization parameter

Enter the regularization number to used by the Updater
in order to avoid overfitting in learning.

Updater function

Select the function to calculate the form of the hyperplane that
separates the two classes.

This function updates the weights of every point in each iteration so
as to perform the gradient step in a given direction to form the

For example, in a 2-dimension space, this hyperplane can be a line or
a set of lines if the points to be classified are linearly

The available functions are:

  • Simple: it does not use the

  • L1: it performs the L1

  • Squared L2: it performs the
    L2 regularization.

Gradient function

Select the loss function to calculate the margin between the
hyperplane and the nearest point of either class.

For further information about the loss functions available on this
drop-down list, see Loss function for classification.

Advanced settings

Use feature scaling

If your training data cannot converge, select this check box to make
tSVMModel reduce the condition numbers
heuristically by scaling the feature data.

Reducing the condition numbers can often improve the convergence


Select this check box to allow the tSVMModel to automatically calculate the intercept constants
and include them in the computation.

In general, intercept can guarantee that the residuals of your model
have a mean of zero.

Validate data before

Select this check box to check whether the vectors of the training
data are well formatted before starting the training.


Usage rule

This component is used as an end component and requires an input link.

You can accelerate the training process by adjusting the stopping conditions such as the
maximum number of iterations or the step size but note that the training
that stops too early could impact its accuracy.

Model evaluation

The parameters you need to set are free parameters and so their values may be provided by
previous experiments, empirical guesses or the like. They do not have any optimal values
applicable for all datasets.

Therefore, you need to train the classifier model you are generating with different sets
of parameter values until you can obtain the best confusion matrix. But note that you need
to write the evaluation code yourself to rank your model with scores.

You need to select the scores to be used depending on the algorithm you want to use to
train your classifier model. This allows you to build the most relevant confusion

For examples about how the confusion matrix is used in a
Job for classification, see Creating a classification model to filter spam.

For a general explanation about confusion matrix, see https://en.wikipedia.org/wiki/Confusion_matrix from Wikipedia.

Spark Connection

In the Spark
tab in the Run
view, define the connection to a given Spark cluster for the whole Job. In
addition, since the Job expects its dependent jar files for execution, you must
specify the directory in the file system to which these jar files are
transferred so that Spark can access these files:

  • Yarn mode (Yarn client or Yarn cluster):

    • When using Google Dataproc, specify a bucket in the
      Google Storage staging bucket
      field in the Spark configuration

    • When using HDInsight, specify the blob to be used for Job
      deployment in the Windows Azure Storage
      area in the Spark

    • When using Altus, specify the S3 bucket or the Azure
      Data Lake Storage for Job deployment in the Spark
    • When using Qubole, add a
      tS3Configuration to your Job to write
      your actual business data in the S3 system with Qubole. Without
      tS3Configuration, this business data is
      written in the Qubole HDFS system and destroyed once you shut
      down your cluster.
    • When using on-premise
      distributions, use the configuration component corresponding
      to the file system your cluster is using. Typically, this
      system is HDFS and so use tHDFSConfiguration.

  • Standalone mode: use the
    configuration component corresponding to the file system your cluster is
    using, such as tHDFSConfiguration or

    If you are using Databricks without any configuration component present
    in your Job, your business data is written directly in DBFS (Databricks

This connection is effective on a per-Job basis.

Related scenarios

No scenario is available for the Spark Batch version of this component

Document get from Talend https://help.talend.com
Thank you for watching.
Notify of
Inline Feedbacks
View all comments
Would love your thoughts, please comment.x