Skip to main content Skip to complementary content

tDecisionTreeModel properties for Apache Spark Batch

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

The Spark Batch tDecisionTreeModel 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 fields.

Click Edit schema 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 window.

Label column

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 classified.

Feature column

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 system

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.

Gain calculation method

An information gain is expected each time a node split occurs. From this drop-down list, select the measure used to define the best split out of each set of splits.

  • gini: it is about how often an element could be incorrectly labelled in a split.

  • entropy: it is about how unpredictable the information in each split is.

For further information about how each of the measures is calculated, see Impurity measures from the Spark documentation.

Maximum number of bins used for descritizing continuous features

Enter the numeric value to indicate the maximum number of bins used for splitting features.

The continuous features are automatically transformed to ordered discrete features.

Maximum depth of the tree

Enter the decision tree depth at which the training should stop adding new nodes. New nodes represent further tests on features on internal nodes and possible class labels held by leaf nodes.

For a tree of n depth, the number of internal nodes is 2n - 1. For example, depth 1 means 1 internal node plus 2 leaf nodes.

Generally speaking, a deeper decision tree is more expressive and thus potentially more accurate in predictions, but it is also more resource consuming and prone to overfitting.

Minimum information gain

Enter the minimum number of information gain to be expected from a parent node to its child nodes. When the number of information gain is less than this minimum number, node split is stopped.

The default value of the minimum number of information gain is 0.0, meaning that no further information is obtained by splitting a given node. As a result, the splitting could be stopped.

For further information about how the information gain is calculated, see Impurity and Information gain from the Spark documentation.

Minimum number of instances per node

Enter the minimum number of training instances a node should have to make it valid for further splitting.

The default value is 1, which means when a node has only 1 row of training data, it stops splitting.

Advanced settings

Maximum memory

Enter the maximum amount of memory (in MB) to be allocated to the training of the tree.

Checkpoint interval

Enter a number to indicate the checkpoint frequency. Every time at the end of the execution of this number of iterations the temporary model is saved.

Usage

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 depth of each decision tree, the maximum number of bins of splitting or the minimum number of information gain, but note that the training that stops too early could impact its performance.

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 matrix.

For examples about how the confusion matrix is used in a Talend 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 Configuration 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 tab.

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

    • When using Altus, specify the S3 bucket or the Azure Data Lake Storage for Job deployment in the Spark configuration tab.
    • 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-premises 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 Apache Spark Batch or tS3Configuration Apache Spark Batch.

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

This connection is effective on a per-Job basis.

Did this page help you?

If you find any issues with this page or its content – a typo, a missing step, or a technical error – please let us know!