> ## Documentation Index
> Fetch the complete documentation index at: https://docs.woodwide.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Models

> Train, retrain, and manage your models.

Train a model on any dataset by specifying a model type. Retrain on new data to create a new version while keeping the same model ID.

<CardGroup cols={2}>
  <Card title="Train Model" icon="plus" href="/api-reference/models/train-model">
    Train a new model on a dataset.
  </Card>

  <Card title="Retrain Model" icon="rotate" href="/api-reference/models/retrain-model">
    Create a new version of an existing model with fresh data.
  </Card>

  <Card title="Get Model" icon="eye" href="/api-reference/models/get-model">
    Fetch details and status for a specific model.
  </Card>

  <Card title="List Models" icon="list" href="/api-reference/models/list-models">
    Retrieve all models in your account.
  </Card>
</CardGroup>

See [Capabilities](/capabilities/overview) for details on each model type.

***

## Selecting Input Columns

By default, all columns in the dataset are used as input features (except the target column for prediction models, which is used only as the label).

You can optionally pass `input_columns` when training to restrict which columns the model uses:

```python Python theme={null}
resp = requests.post(
    f"{base_url}/models/train",
    headers=headers,
    json={
        "model_name": "my_model",
        "model_type": "prediction",
        "dataset_id": dataset_id,
        "label_column": "target",
        "input_columns": ["age", "income", "region"],
    },
)
```