fit

Feedback:

Use the fit operator to train a machine learning model on AQL search results.

Typically, a machine learning model consists of the following components:-

`algorithm`	Specifies the name of the machine learning algorithm. Supported algorithms include: ARIMA birch DBSCAN DecisionTreeRegressor DensityFunction IsolationForest LinearRegression LocalOutlierFactor LogisticRegression OrdinalEncoder RandomForest StandardScaler
`option_name`	Indicates the model parameters. The format is as follows: `option_name=value` Refer to the documentation and examples provided for the particular supported algorithm.
`field_to_predict`	Includes the target field name. Required for supervised models only. Not applicable to unsupervised algorithms.
`explanatory_fields`	Captures the input fields used to train the model. Required for supervised models. Not used for unsupervised models.
`kfold_cv`	Ensures k-fold cross validation (if the algorithm supports it). When specified, the validation result is returned. Substitute a number of folds for n (minimum value is 2). Note: This option cannot be used together with `into` option to save the model.
`into`	Use the `into` keyword to store the trained model as an artifact that can later be applied using the `apply` operator.
`'model_name'`	Denotes the name of the model artifact. The model name must be wrapped in single or double quotation marks.

The following AQL piped operators are used to manage machine learning models.

AQL Operator	Usage
fit	Train a machine learning model on AQL search results.
Example 1: Training with Context, Applying with Context	Compute predictions for the current search results based on a model that was learned using the `fit` operator.
listmodels	Return a list of machine learning models that were learned using the `fit` operator.
deletemodel	Delete a model that was generated by the `fit` operator.
ml_model_summary	Retrieve metadata about a model that was generated by the `fit` operator.

Example 1: Fit a linear regression model to data.

Copy

|fit LinearRegression species from petal_length petal_width sepal_length sepal_width into 'iris_linear_regression_model'

Example 2: Build a classification model.

Copy

|fit LinearRegression species from petal_length petal_width sepal_length sepal_width kfold_cv=5

Example 3: Apply trained model to new data.

Copy

|fit LogisticRegression species from petal_length petal_width sepal_length sepal_width into 'iris_logistic_regression_model'

Example 4: Generate predictions from model.

Copy

|fit LogisticRegression species from petal_length petal_width sepal_length sepal_width kfold_cv=5