# Train a model

The **Train a model** task trains a machine learning model.

After the model has been trained, you can:
 * interpret it ([*Understand a model* task](task_analyze))
 * evaluate it ([*Evaluate a model* task](task_evaluate)),
 * make predictions with it ([*Make predictions with a model* task](task_predict))
 * get the code to use it in Colab ([*Export a model*
task](task_share))

The trained model is saved in the simple_ml_for_sheets folder of your Google Drive.

Use this task as follows:

1.  Open a sheet with the training examples values. The sheet should be in the
    [tabular format](sheet_format).
2.  Select the "Train a model" task.
3.  Give a name to the model. The model will be stored as a zip file in the
    "simple_machine_learning" directory at the root of your Google Drive.
4.  Select the **label**, input features (called "source columns") and the other parameters.
5.  Click "Train."
    
    After a few seconds, the training is done.

6.  Click "Close."