Model Configuration

The configuration page is standard for all the ML Models except Tree-based, Naive Bayes, and Logistic Regression models. For the rest of the models, the properties are as follows.

Operation: Type of operation to be performed by the Analytics processor.

Training: Select the option training, if you want to train new models.

Prediction: Select the option prediction, if you want to give predictions over existing model.

Algorithm Type: Specifies whether the current algorithm is used for solving a classification problem or regression. Select the required algorithm from the drop-down list.

Classification Type: Type of Classification- Binary or Multiclass.

Model Name: Name of the model to be created when the training mode is selected or the name of an existing model to be used for the prediction when the prediction mode is selected.

Description: Summary or a brief description of the model.

Tags: Tags to be associated with the model.

Version Comments: A note about the model version.

Save Model On: Enables to save model on S3 or Gathr Storage (Gathr Store Output).

If Gathr Storage is selected, then the saved model will be available for download in Gathr Store Output.

Note: Gathr Store output records are available for 15 days from the created date.

If S3 is selected, then Specify S3 connection details, bucket name and path where the model should be saved.

Connection Name: Connections are the service identifiers. A connection name can be selected from the list if you have created and saved connection details for Amazon S3 earlier. Or create one as explained in the topic - Amazon S3 Connection →

Bucket Name: S3 target bucket name is to be specified.

Path: File or directory path of the bucket should be specified where the model needs to be saved.

The path must end with * in case of directory.

Example: outdir/*

ADD CONFIGURATION: Additional properties can be added using add configuration.