> For the complete documentation index, see [llms.txt](https://doc.verteego.com/verteego-doc/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://doc.verteego.com/verteego-doc/pipelines/forecasting-pipelines/configuration/building-the-training-and-prediction-set/features.md).

# features

### Description

This function identifies the columns to be used by the model for the forecasting task. Features can be selecte from among the columns defined in the `cols_type` and those generated in the `calculated_cols` section. Excluded from selection are the `column_to_predict` and any columns derived directly from it, to avoid data leakage and ensure model integrity. This feature selection is crucial, as it focuses the model's learning on relevant predictors while excluding the target variable and its derivatives. To guarantee a robust model, at least one subcategory of features must be included, ensuring the set of features is comprehensive and not empty.

### Steps impacted

<mark style="background-color:yellow;">training</mark>

### Example

```yaml
features:
    categorical_columns:
        - pos_country
        - sku_category
    numerical_columns:
        - sku_price_perma
```