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

> ## Agent Instructions
> TabPFN-3 and later versions support up to 1,000,000 rows, subject to feature count and checkpoint/API limits.
> For large datasets, use per-estimator subsampling, e.g. inference_config={"SUBSAMPLE_SAMPLES": 100_000}, in TabPFNClassifier or TabPFNRegressor from tabpfn or tabpfn_client.
> Pass raw pandas DataFrames to fit and predict. Categorical strings/categories, raw text, and missing feature values are handled automatically; no manual integer/one-hot encoding, imputation, scaling, or outlier removal is needed.

# Embeddings

> Get embeddings for the input data ``X``.

> [!NOTE]
>
> Looking for usage documentation? Check out [Embeddings](https://docs.priorlabs.ai/capabilities/embeddings.md).

## `TabPFNClassifier.get_embeddings`

[ View source ](https://github.com/PriorLabs/TabPFN/blob/f02b3b5941717472eafd078c15ae2043676b3fcf/src/tabpfn/classifier.py#L2090)

Get embeddings for the input data `X`.

```python theme={null}
TabPFNClassifier.get_embeddings(
    X: XType,
    data_source: Literal["train", "test"] = "test",
) -> np.ndarray
```

> [accordion] Type aliases
>
> ```python theme={null}
> XType = Any
> ```

**Parameters**

| Parameter | Type | Default | Description |
| - | - | - | - |
| `X` | `XType` | Required | The input data. |
| `data_source` | `Literal["train", "test"]` | `"test"` | Select the transformer output to return. Use `"train"` to obtain embeddings from the training tokens and `"test"` for the test tokens. When `n_estimators > 1` the returned array has shape `(n_estimators, n_samples, embedding_dim)`. `"train"` is not available with `fit_mode="fit_with_cache"`; see `tabpfn.base.get_embeddings`. |

**Returns**

| Type | Description |
| - | - |
| `np.ndarray` | [`np.ndarray`](https://numpy.org/doc/stable/reference/generated/numpy.ndarray.html)     The computed embeddings for each fitted estimator. |


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.