Auto ML Binary class

This transform will use the selected algorithm (XGBoost, GBM, RandomForest, or Ensemble) to create models for classification problem in a dataset and returns the data with model metrics and prediction.

Parameters

The table gives a brief description about each parameter in Auto ML Binary class transform.

Name:

By default, the transform name is populated. You can also add a custom name for the transform.

Train dataset:

The train dataset for binary classification.

Test dataset:

The test dataset used for binary classification.

Target column:

The column on which predictions are made.

Build Platform:

The platform used for creating models. By default, H20 is used. This creates multiple models.

Algorithm:

The algorithm used to create models.

Max:

The maximum number of models you can use on the test data to validate the model performance.

Output Model Performance:

The file name with which the output is created. This file contains the model metrics.

Output Model prediction:

The file name with which the output is created with the model predictions.

Sample input for Auto ML binary class transform:

The output after running the Auto ML Binary class transform on the dataset appears as below for model performance:

The output for model prediction:

The dashboard output appears as below:

How to use it in Notebook

The following is the code snippet you must use in the Jupyter Notebook editor to run the **Auto ML Binary class**transform:

transform = Transform()
transform.name = "automl binary classification"
transform.templateId = automl.id
transform.variables = {
    "inputTrainDataset": train_w_fs.name,
    "inputTestDataset": test_w_fs.name,
    "targetCol": targetCol,
    "outputModelPerformance": "model_performance",
    "outputModelPredictions": "model_prediction",
}
recipe_auto_ml = project.addRecipe([train_w_fs, test_w_fs], name="auto_ml")
#recipe_auto_ml.prepareForLocal(transform, contextId="recipe_auto_ml")
recipe_auto_ml.addTransform(transform)
recipe_auto_ml.run()