AutoML
Overview
You can use this Snap to simplify the exploration and tuning of machine learning models within a specified resource limit. A ML model is a mathematical tool used to predict or solve specific problems, such as forecasting sales, predicting customer churn, or determining loan repayment likelihood.
Transform-type Snap
Does not support Ultra Tasks
Prerequisites
None.
Limitations
- The Snap supports binary classification, multiclass classification, and regression problems.
- Generates a leaderboard ranking the top models with relevant metrics.
- Statistical data such as RMSLE may display as NaN for certain models and are excluded from reports.
Known issues
- Reports may exclude models with incomplete statistical data (e.g., NaN values in RMSLE).
Snap views
| View | Description | Examples of upstream and downstream Snaps |
|---|---|---|
| Input |
The Snap accepts one or two input views:
|
|
| Output |
The Snap produces one to three output views:
|
|
| Error |
Error handling is a generic way to handle errors without losing data or failing the Snap execution. You can handle the errors that the Snap might encounter when running the pipeline by choosing one of the following options from the When errors occur list under the Views tab. The available options are:
Learn more about Error handling in Pipelines. |
|
Snap settings
- Expression icon (
): JavaScript
syntax to access SnapLogic Expressions to set field values dynamically (if enabled). If
disabled, you can provide a static value. Learn more. - SnapGPT (
): Generates SnapLogic Expressions
based on natural language using SnapGPT. Learn
more. - Suggestion icon (
): Populates a
list of values dynamically based on your Account configuration. - Upload
: Uploads files. Learn more.
| Field / field set | Type | Description |
|---|---|---|
| Label | String |
Required. Specify a unique name for the Snap. Modify this to be more appropriate, especially if more than one of the same Snaps is in the pipeline. Default value: AutoML Example: Predict customer churn |
| Label field | String/Suggestion | Required.Specify the target or class field of the dataset that the model will use during the training process. This field represents the expected output values that the model learns to predict based on the input data. During inference, the model predicts this field as its output. Default value: N/A Example: $class |
| Time limit | Integer | Required.The maximum time in seconds to run the Snap. Set to 0 to use the number of models specified in Number of models without time restriction.Default value: 3600 Example: 3670 |
| Number of models | Integer | Required.The number of models to generate. If 0, the Snap builds as many models as possible within the time limit.Default value: 10 Example: 15 |
| Fold | Integer/Expression | Required. Enter the number of folds for cross-validation. This determines how the dataset is split for training and testing during model evaluation. Default value: 5 Example: 10 |
| Engine | String/Suggestion | Required.The engine used for model generation: Weka or H2O.Default value: H20 Example: Weka |
| Algorithms | String | Select algorithms for model building. Available options include: Standard, Tree, XGBoost, and NN. Default value: Standard, Tree, XGBoost, NN Example: Standard |
| Readable | Checkbox | When selected, the output model is made more interpretable, focusing on readability for end-users. Default status: Deselected |
| Use Random Seed | Checkbox | When selected, it ensures reproducibility by setting a fixed random seed for model training and evaluation. Default status: Deselected |
| Random Seed | String/Expression | Specify reproducibility by setting a fixed random seed for model training and evaluation. Default value: 12345 Example: 500 |
| Report title | String/Expression | Enter an optional title for the output report generated after the AutoML process is complete. Default value: AutoML Example: Predict customer churn |
| Snap execution | Dropdown list | Select one of the three modes in which the Snap executes.
Available options are:
|