> ## 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.

# Constructor and configuration

> TabPFN regressor with a scikit-learn-compatible interface.

<Info>
  Looking for usage documentation? Check out [Regression](/capabilities/regression).
</Info>

<div className="python-reference-heading">
  <h2 id="constructor">
    `TabPFNRegressor`
  </h2>

  <a className="python-reference-source" href="https://github.com/PriorLabs/TabPFN/blob/c70b6ef0488858d32244c52222abfc0c5be207d6/src/tabpfn/regressor.py#L177" aria-label="View source for TabPFNRegressor"><span aria-hidden="true">\</></span> View source <span aria-hidden="true">↗</span></a>
</div>

TabPFN regressor with a scikit-learn-compatible interface.

Construct a TabPFN regressor.

This constructs a regressor using the latest model and settings. If you would
like to use a previous model version, use [`create_default_for_version()`](/api-reference/python/tabpfn/regressor/configuration#create-default-for-version)
instead. You can also use `model_path` to specify a particular model.

```python theme={null}
TabPFNRegressor(
    *,
    n_estimators: int | Literal["auto"] = "auto",
    auto_scale_n_estimators: bool = True,
    categorical_features_indices: Sequence[int] | None = None,
    softmax_temperature: float | Literal["auto"] = "auto",
    average_before_softmax: bool = False,
    model_path: str | Path | list[str] | list[Path] | Literal["auto"] | RegressorModelSpecs | list[RegressorModelSpecs] = "auto",
    device: DevicesSpecification = "auto",
    ignore_pretraining_limits: bool = False,
    inference_precision: _dtype | Literal["autocast", "auto"] = "auto",
    fit_mode: Literal["low_memory", "fit_preprocessors", "fit_with_cache", "batched"] = "fit_preprocessors",
    memory_saving_mode: MemorySavingMode = "auto",
    keep_cache_on_device: bool = True,
    kv_cache_precision: Literal["auto", "int8", "fp8"] | None = None,
    random_state: int | np.random.RandomState | np.random.Generator | None = 0,
    n_jobs: Annotated[int | None, deprecated("Use n_preprocessing_jobs")] = None,
    n_preprocessing_jobs: int = 1,
    inference_config: dict | InferenceConfig | None = None,
    differentiable_input: bool = False,
    eval_metric: str | RegressorEvalMetrics | None = None,
    tuning_config: dict | RegressorTuningConfig | None = None,
    show_progress_bar: bool = False,
)
```

<Accordion title="Type aliases">
  ```python theme={null}
  DevicesSpecification = torch.device | str | Sequence[torch.device | str] | Literal["auto"]
  MemorySavingMode = bool | Literal["auto"] | float | int
  ```
</Accordion>

**Parameters**

<div className="python-reference-table">
  | Parameter | Type | Default | Description |
  | - | - | - | - |
  | <span id="constructor--n-estimators" /><code className="python-reference-parameter">n\_<wbr />estimators</code> | <code className="python-reference-type">int \| Literal\["auto"]</code> | `"auto"` | The number of estimators in the TabPFN ensemble. We aggregate the predictions of `n_estimators`-many forward passes of TabPFN. Each forward pass has (slightly) different input data. Think of this as an ensemble of `n_estimators`-many "prompts" of the input data. With the default `"auto"`, the count comes from the checkpoint (`InferenceConfig.N_ESTIMATORS`), which is itself `"auto"` unless the checkpoint names a count. `"auto"` means `DEFAULT_N_ESTIMATORS`, raised on wide datasets so every feature is seen by some estimator (i.e. when the data has more than `max_features_per_estimator` features per estimator), to the smallest value that lets every feature appear in at least one ensemble member, emitting a warning when it does so. That auto-scaled value is capped at `MAX_AUTO_SCALED_N_ESTIMATORS`; beyond that some features may never be sampled unless you raise `n_estimators` yourself. An explicit integer — yours or the checkpoint's — is never overridden: if it is too small to cover every feature, a warning is emitted at fit time and the value is used as given. Your integer cannot be combined with an `N_ESTIMATORS` in `inference_config`, which is the other way of naming a count. |
  | <span id="constructor--auto-scale-n-estimators" /><code className="python-reference-parameter">auto\_<wbr />scale\_<wbr />n\_<wbr />estimators</code> | <code className="python-reference-type">bool</code> | `True` | Deprecated, removed in v9 — pass an explicit `n_estimators` instead. Only applies when `n_estimators="auto"`, where `False` keeps the auto value at `DEFAULT_N_ESTIMATORS` rather than raising it for feature coverage, exactly what passing that count as `n_estimators` does. Passing `False` emits a `FutureWarning` at fit time. |
  | <span id="constructor--categorical-features-indices" /><code className="python-reference-parameter">categorical\_<wbr />features\_<wbr />indices</code> | <code className="python-reference-type">Sequence\[int] \| None</code> | `None` | The indices of the columns that are suggested to be treated as categorical. If `None`, the model will infer the categorical columns. A column with pandas' `category` dtype counts as listed here. A string column declared this way is read as categorical whatever its cardinality, never as text; for a numeric one, we might ignore the suggestion to better fit the data seen during pre-training.<br /><br />!!! note     The indices are 0-based and should represent the data passed to     `.fit()`. If the data changes between the initializations of the     model and the `.fit()`, consider setting the     `.categorical_features_indices` attribute after the model was     initialized and before `.fit()`. |
  | <span id="constructor--softmax-temperature" /><code className="python-reference-parameter">softmax\_<wbr />temperature</code> | <code className="python-reference-type">float \| Literal\["auto"]</code> | `"auto"` | The temperature for the softmax function. This is used to control the confidence of the model's predictions. Lower values make the model's predictions more confident. This is only applied when predicting during a post-processing step. Set `softmax_temperature=1.0` for no effect.<br /><br />If `"auto"` (the default), the temperature is taken from the checkpoint (`InferenceConfig.SOFTMAX_TEMPERATURE`), which is `0.9` for every checkpoint released up to and including v8.5.0. Passing a float overrides the checkpoint for every model in the ensemble; it cannot be combined with a `SOFTMAX_TEMPERATURE` in `inference_config`, which is the other way of naming one. |
  | <span id="constructor--average-before-softmax" /><code className="python-reference-parameter">average\_<wbr />before\_<wbr />softmax</code> | <code className="python-reference-type">bool</code> | `False` | Only used if `n_estimators > 1`. Whether to average the predictions of the estimators before applying the softmax function. This can help to improve predictive performance when there are many classes or when calibrating the model's confidence. This is only applied when predicting during a post-processing.<br /><br />- If `True`, the predictions are averaged before applying the softmax   function. Thus, we average the logits of TabPFN and then apply the   softmax.<br />- If `False`, the softmax function is applied to each set of logits.   Then, we average the resulting probabilities of each forward pass. |
  | <span id="constructor--model-path" /><code className="python-reference-parameter">model\_<wbr />path</code> | <code className="python-reference-type">str \| <a href="https://docs.python.org/3/library/pathlib.html#pathlib.Path">Path</a> \| list\[str] \| list\[<a href="https://docs.python.org/3/library/pathlib.html#pathlib.Path">Path</a>] \| Literal\["auto"] \| <a href="/api-reference/python/tabpfn/model-persistence#regressormodelspecs">Regressor<wbr />Model<wbr />Specs</a> \| list\[<a href="/api-reference/python/tabpfn/model-persistence#regressormodelspecs">Regressor<wbr />Model<wbr />Specs</a>]</code> | `"auto"` | The path to the TabPFN model file, i.e., the pre-trained weights.<br /><br />- If `"auto"`, the model will be downloaded upon first use. This   defaults to your system cache directory, but can be overwritten   with the use of an environment variable `TABPFN_MODEL_CACHE_DIR`.<br />- If a path or a string of a path, the model will be loaded from   the user-specified location if available, otherwise it will be   downloaded to this location. Details on available checkpoints are   available in the repository README. |
  | <span id="constructor--device" /><code className="python-reference-parameter">device</code> | <code className="python-reference-type"><a href="https://github.com/PriorLabs/TabPFN/blob/c70b6ef0488858d32244c52222abfc0c5be207d6/src/tabpfn/utils.py#L101">Devices<wbr />Specification</a></code> | `"auto"` | The device(s) to use for inference. See the documentation of `.to()`. |
  | <span id="constructor--ignore-pretraining-limits" /><code className="python-reference-parameter">ignore\_<wbr />pretraining\_<wbr />limits</code> | <code className="python-reference-type">bool</code> | `False` | Whether to ignore the pre-training limits of the model. The TabPFN models have been pre-trained on a specific range of input data. If the input data is outside of this range, the model may not perform well. You may ignore our limits to use the model on data outside the pre-training range.<br /><br />- If `True`, the model will not raise an error if the input data is   outside the pre-training range. Also suppresses error when using   the model with a large dataset on CPU.<br />- If `False`, you can use the model outside the pre-training range, but   the model could perform worse.<br /><br />!!! note<br /><br />    For version 2.5, the pre-training limits are:<br /><br />    - 50\_000 samples/rows<br />    - 2\_000 features/columns (Note that for more than 500 features we         subsample 500 features per estimator. It is therefore important         to use a sufficiently large number of `n_estimators`.) |
  | <span id="constructor--inference-precision" /><code className="python-reference-parameter">inference\_<wbr />precision</code> | <code className="python-reference-type">\_dtype \| Literal\["autocast", "auto"]</code> | `"auto"` | The precision to use for inference. This can dramatically affect the speed and reproducibility of the inference. Higher precision can lead to better reproducibility but at the cost of speed. By default, we optimize for speed and use torch's mixed-precision autocast. The options are:<br /><br />- If `torch.dtype`, we force precision of the model and data to be   the specified torch.dtype during inference. This can is particularly   useful for reproducibility. Here, we do not use mixed-precision.<br />- If `"autocast"`, enable PyTorch's mixed-precision autocast. Ensure   that your device is compatible with mixed-precision.<br />- If `"auto"`, we determine whether to use autocast or not depending on   the device type. |
  | <span id="constructor--fit-mode" /><code className="python-reference-parameter">fit\_<wbr />mode</code> | <code className="python-reference-type">Literal\["low\_memory", "fit\_preprocessors", "fit\_with\_cache", "batched"]</code> | `"fit_preprocessors"` | Determine how the TabPFN model is "fitted". The mode determines how the data is preprocessed and cached for inference. This is unique to an in-context learning foundation model like TabPFN, as the "fitting" is technically the forward pass of the model. The options are:<br /><br />- If `"low_memory"`, the data is preprocessed on-demand during inference   when calling `.predict()` or `.predict_proba()`. This is the most   memory-efficient mode but can be slower for large datasets because   the data is (repeatedly) preprocessed on-the-fly.   Ideal with low GPU memory and/or a single call to `.fit()` and   `.predict()`.<br />- If `"fit_preprocessors"`, the data is preprocessed and cached once   during the `.fit()` call. During inference, the cached preprocessing   (of the training data) is used instead of re-computing it.   Ideal with low GPU memory and multiple calls to `.predict()` with   the same training data.<br />- If `"fit_with_cache"`, the data is preprocessed and cached once during   the `.fit()` call like in `fit_preprocessors`. Moreover, the   transformer key-value cache is also initialized, allowing for much   faster inference on the same data at a large cost of memory.   Ideal with very high GPU memory and multiple calls to `.predict()`   with the same training data.<br />- If `"batched"`, the already pre-processed data is iterated over in   batches. This can only be done after the data has been preprocessed   with the get\_preprocessed\_datasets function. This is primarily used   only for inference with the InferenceEngineBatchedNoPreprocessing   class in Fine-Tuning. The [`fit_from_preprocessed`](/api-reference/python/tabpfn/regressor/advanced#fit-from-preprocessed)() function sets this   attribute internally. |
  | <span id="constructor--memory-saving-mode" /><code className="python-reference-parameter">memory\_<wbr />saving\_<wbr />mode</code> | <code className="python-reference-type"><a href="https://github.com/PriorLabs/TabPFN/blob/c70b6ef0488858d32244c52222abfc0c5be207d6/src/tabpfn/constants.py#L21">Memory<wbr />Saving<wbr />Mode</a></code> | `"auto"` | Enable GPU/CPU memory saving mode. This can both avoid out-of-memory errors and improve fit+predict speed by reducing memory pressure.<br /><br />It saves memory by automatically batching certain model computations within TabPFN.<br /><br />- If "auto": memory saving mode is enabled/disabled automatically based     on a heuristic<br />- If `True`/`False`: memory saving mode is forced enabled/disabled.<br /><br />If speed is important to your application, you may wish to manually tune this option by comparing the time taken for fit+predict with it set to `False` and `True`.<br /><br />!!! warning     This does not batch the original input data. We still recommend to     batch the test set as necessary if you run out of memory. |
  | <span id="constructor--keep-cache-on-device" /><code className="python-reference-parameter">keep\_<wbr />cache\_<wbr />on\_<wbr />device</code> | <code className="python-reference-type">bool</code> | `True` | Only relevant when `fit_mode="fit_with_cache"`. If `True` (default), the key-value cache is kept on the inference device (e.g. GPU). Uses more device memory but gives lower latency. If `False`, the cache is stored on CPU. |
  | <span id="constructor--kv-cache-precision" /><code className="python-reference-parameter">kv\_<wbr />cache\_<wbr />precision</code> | <code className="python-reference-type">Literal\["auto", "int8", "fp8"] \| None</code> | `None` | Only relevant when `fit_mode="fit_with_cache"`. Resolved against what the model architecture supports. `None` (default) picks the architecture default (`"int8"` when it can quantize, e.g. TabPFN-3, else `"auto"`); `"int8"` quantizes the key-value cache to save memory; `"fp8"` stores it as 8-bit floats instead (same size, float rounding semantics; not supported on MPS); `"auto"` keeps the computed dtype. Requesting a quantized precision on an architecture that cannot quantize warns and falls back to `"auto"`. |
  | <span id="constructor--random-state" /><code className="python-reference-parameter">random\_<wbr />state</code> | <code className="python-reference-type">int \| np.random.Random<wbr />State \| np.random.Generator \| None</code> | `0` | Controls the randomness of the model. Pass an int for reproducible results and see the scikit-learn glossary for more information. If `None`, the randomness is determined by the system when calling `.fit()`.<br /><br />!!! warning     We depart from the usual scikit-learn behavior in that by default     we provide a fixed seed of `0`.<br /><br />!!! note     Even if a seed is passed, we cannot always guarantee reproducibility     due to PyTorch's non-deterministic operations and general numerical     instability. To get the most reproducible results across hardware,     we recommend using a higher precision as well (at the cost of a     much higher inference time). Likewise, for scikit-learn, consider     passing `USE_SKLEARN_16_DECIMAL_PRECISION=True` as kwarg. |
  | <span id="constructor--n-jobs" /><code className="python-reference-parameter">n\_<wbr />jobs</code> | <code className="python-reference-type">int \| None</code> | `None` | Deprecated, use `n_preprocessing_jobs` instead. This parameter never had any effect. |
  | <span id="constructor--n-preprocessing-jobs" /><code className="python-reference-parameter">n\_<wbr />preprocessing\_<wbr />jobs</code> | <code className="python-reference-type">int</code> | `1` | The number of worker processes to use for the preprocessing.<br /><br />If `1`, the preprocessing will be performed in the current process, parallelised across multiple CPU cores. If `>1` and `n_estimators > 1`, then different estimators will be dispatched to different processes.<br /><br />We strongly recommend setting this to 1, which has the lowest overhead and can often fully utilise the CPU. Values >1 can help if you have lots of CPU cores available, but can also be slower. |
  | <span id="constructor--inference-config" /><code className="python-reference-parameter">inference\_<wbr />config</code> | <code className="python-reference-type">dict \| <a href="/api-reference/python/tabpfn/inference-config#inferenceconfig">Inference<wbr />Config</a> \| None</code><br /><a href="/api-reference/python/tabpfn/inference-config#inferenceconfig"><code>InferenceConfig</code> options</a> | `None` | For advanced users, additional advanced arguments that adjust the behavior of the model interface. See [`tabpfn.inference_config.InferenceConfig`](/api-reference/python/tabpfn/inference-config#inferenceconfig) for details and options.<br /><br />- If `None`, the default [`InferenceConfig`](/api-reference/python/tabpfn/inference-config#inferenceconfig) is used.<br />- If `dict`, the key-value pairs are used to update the default   [`InferenceConfig`](/api-reference/python/tabpfn/inference-config#inferenceconfig). Raises an error if an unknown key is passed.<br />- If [`InferenceConfig`](/api-reference/python/tabpfn/inference-config#inferenceconfig), the object replaces the checkpoint's config   as a whole, so any field not set on it takes a class default   rather than the value the checkpoint declares. Deprecated. |
  | <span id="constructor--differentiable-input" /><code className="python-reference-parameter">differentiable\_<wbr />input</code> | <code className="python-reference-type">bool</code> | `False` | If true, preprocessing attempts to be end-to-end differentiable. Less relevant for standard regression fine-tuning compared to prompt-tuning. |
  | <span id="constructor--eval-metric" /><code className="python-reference-parameter">eval\_<wbr />metric</code> | <code className="python-reference-type">str \| <a href="/api-reference/python/tabpfn/inference-tuning#regressorevalmetrics">Regressor<wbr />Eval<wbr />Metrics</a> \| None</code> | `None` | Metric by which predictions will be evaluated on test data for temperature calibration. For currently supported metrics, see [`tabpfn.inference_tuning.RegressorEvalMetrics`](/api-reference/python/tabpfn/inference-tuning#regressorevalmetrics). |
  | <span id="constructor--tuning-config" /><code className="python-reference-parameter">tuning\_<wbr />config</code> | <code className="python-reference-type">dict \| <a href="/api-reference/python/tabpfn/inference-tuning#regressortuningconfig">Regressor<wbr />Tuning<wbr />Config</a> \| None</code> | `None` | The settings to use to tune the model's predictions for the specified `eval_metric`. See [`tabpfn.inference_tuning.RegressorTuningConfig`](/api-reference/python/tabpfn/inference-tuning#regressortuningconfig) for details and options. |
  | <span id="constructor--show-progress-bar" /><code className="python-reference-parameter">show\_<wbr />progress\_<wbr />bar</code> | <code className="python-reference-type">bool</code> | `False` | Whether to show a progress bar during inference. Defaults to `False`. |
</div>

**Usage guidance**

* TabPFN-3 and later versions support up to 1,000,000 rows, subject to
  feature count, checkpoint limits, and memory.
* For large datasets or limited memory, use per-estimator subsampling,
  e.g. `inference_config={"SUBSAMPLE_SAMPLES": 50_000}`.
* Pass raw pandas DataFrames to [`fit`](/api-reference/python/tabpfn/regressor/fitting-and-prediction#fit) and [`predict`](/api-reference/python/tabpfn/regressor/fitting-and-prediction#predict). Categorical
  strings/categories and missing feature values are handled automatically;
  no manual integer/one-hot encoding, imputation, scaling, or outlier
  removal is needed.

**Attributes**

| Attribute | Type | Description |
| - | - | - |
| `configs_` | <code className="python-reference-type">list\[<a href="https://github.com/PriorLabs/TabPFN/blob/c70b6ef0488858d32244c52222abfc0c5be207d6/src/tabpfn/architectures/interface.py#L18">Architecture<wbr />Config</a>]</code> | The configurations of the loaded models to be used for inference.<br /><br />The concrete type of these configs is defined by the architectures in use and should be inspected at runtime, but they will be subclasses of [`ArchitectureConfig`](https://github.com/PriorLabs/TabPFN/blob/c70b6ef0488858d32244c52222abfc0c5be207d6/src/tabpfn/architectures/interface.py#L18). |
| `models_` | <code className="python-reference-type">list\[<a href="https://github.com/PriorLabs/TabPFN/blob/c70b6ef0488858d32244c52222abfc0c5be207d6/src/tabpfn/architectures/interface.py#L166">Architecture</a>]</code> | The loaded models to be used for inference.<br /><br />The models can be different PyTorch modules, but will be subclasses of [`Architecture`](https://github.com/PriorLabs/TabPFN/blob/c70b6ef0488858d32244c52222abfc0c5be207d6/src/tabpfn/architectures/interface.py#L166). |
| `inference_config_` | <code className="python-reference-type"><a href="/api-reference/python/tabpfn/inference-config#inferenceconfig">Inference<wbr />Config</a></code> | Additional configuration of inference for expert users. |
| `devices_` | <code className="python-reference-type">tuple\[torch.device, ...]</code> | The devices determined to be used.<br /><br />The devices are determined based on the `device` argument to the constructor, and the devices available on the system. See the constructor documentation for details. |
| `feature_names_in_` | <code className="python-reference-type">npt.ND<wbr />Array\[Any]</code> | The feature names of the input data.<br /><br />May not be set if the input data does not have feature names, such as with a numpy array. |
| `n_features_in_` | <code className="python-reference-type">int</code> | The number of features in the input data used during [`fit()`](/api-reference/python/tabpfn/regressor/fitting-and-prediction#fit). |
| `n_train_samples_` | <code className="python-reference-type">int</code> | The number of training samples used during [`fit()`](/api-reference/python/tabpfn/regressor/fitting-and-prediction#fit). |
| `inferred_feature_schema_` | <code className="python-reference-type"><a href="https://github.com/PriorLabs/TabPFN/blob/c70b6ef0488858d32244c52222abfc0c5be207d6/src/tabpfn/preprocessing/datamodel.py#L145">Feature<wbr />Schema</a></code> | The inferred feature schema. This contains the feature modalities per column, using heuristics and user-provided indices for categorical features. |
| `n_outputs_` | <code className="python-reference-type">Literal\[1]</code> | The number of outputs the model supports. Only 1 for now |
| `znorm_space_bardist_` | <code className="python-reference-type"><a href="https://github.com/PriorLabs/TabPFN/blob/c70b6ef0488858d32244c52222abfc0c5be207d6/src/tabpfn/architectures/shared/bar_distribution.py#L471">Full<wbr />Support<wbr />Bar<wbr />Distribution</a></code> | The bar distribution of the target variable, used by the model. This is the bar distribution in the normalized target space. |
| `raw_space_bardist_` | <code className="python-reference-type"><a href="https://github.com/PriorLabs/TabPFN/blob/c70b6ef0488858d32244c52222abfc0c5be207d6/src/tabpfn/architectures/shared/bar_distribution.py#L471">Full<wbr />Support<wbr />Bar<wbr />Distribution</a></code> | The bar distribution in the raw target space, used for computing the predictions. |
| `use_autocast_` | <code className="python-reference-type">bool</code> | Whether torch's autocast should be used. |
| `forced_inference_dtype_` | <code className="python-reference-type">\_dtype \| None</code> | The forced inference dtype for the model based on `inference_precision`. |
| `executor_` | <code className="python-reference-type"><a href="https://github.com/PriorLabs/TabPFN/blob/c70b6ef0488858d32244c52222abfc0c5be207d6/src/tabpfn/inference.py#L93">Inference<wbr />Engine</a></code> | The inference engine used to make predictions. |
| `ordinal_encoder_` | <code className="python-reference-type"><a href="https://github.com/PriorLabs/TabPFN/blob/c70b6ef0488858d32244c52222abfc0c5be207d6/src/tabpfn/preprocessing/steps/preprocessing_helpers.py#L471">Order<wbr />Preserving<wbr />Column<wbr />Transformer</a></code> | The column transformer used to preprocess categorical data to be numeric. |
| `date_transformer_` | <code className="python-reference-type"><a href="https://github.com/PriorLabs/TabPFN/blob/c70b6ef0488858d32244c52222abfc0c5be207d6/src/tabpfn/preprocessing/datetimes.py#L52">Date<wbr />Transformer</a></code> | The transformer that converted every temporal column before validation. |
| `text_transformer_` | <code className="python-reference-type"><a href="https://github.com/PriorLabs/TabPFN/blob/c70b6ef0488858d32244c52222abfc0c5be207d6/src/tabpfn/preprocessing/text.py#L61">Text<wbr />Transformer</a></code> | The transformer that expanded every text column before validation. |
| `categorical_features_indices_` | <code className="python-reference-type">list\[int] \| None</code> | Declared categorical column positions after date/text expansion, including columns declared through pandas `category` dtype. Expanded source columns are removed and their generated features appended, so these positions can differ from those in the original fit input. |
| `eval_metric_` | <code className="python-reference-type"><a href="/api-reference/python/tabpfn/inference-tuning#regressorevalmetrics">Regressor<wbr />Eval<wbr />Metrics</a></code> | The validated evaluation metric to optimize for during prediction. |
| `ensemble_softmax_temperature_` | <code className="python-reference-type">float</code> | The temperature applied to the aggregated ensemble distribution at predict time, after the per-estimator `softmax_temperature`. This is `1.0`, a no-op, when no temperature calibration is done. |
| `softmax_temperature_` | <code className="python-reference-type">float</code> | The resolved per-estimator `softmax_temperature`, i.e. the one the checkpoint declares unless it was overridden. |

***

<div className="python-reference-heading">
  <h2 id="create-default-for-version">
    `TabPFNRegressor.create_default_for_version`
  </h2>

  <a className="python-reference-source" href="https://github.com/PriorLabs/TabPFN/blob/c70b6ef0488858d32244c52222abfc0c5be207d6/src/tabpfn/regressor.py#L620" aria-label="View source for TabPFNRegressor.create_default_for_version"><span aria-hidden="true">\</></span> View source <span aria-hidden="true">↗</span></a>
</div>

Construct a regressor that uses the given version of the model.

In addition to selecting the model, this also configures certain settings to the
default values associated with this model version.

Any kwargs will override the default settings.

```python theme={null}
TabPFNRegressor.create_default_for_version(
    version: ModelVersion,
    **overrides,
) -> Self
```

**Parameters**

<div className="python-reference-table">
  | Parameter | Type | Default | Description |
  | - | - | - | - |
  | <span id="create-default-for-version--version" /><code className="python-reference-parameter">version</code> | <code className="python-reference-type"><a href="/api-reference/python/tabpfn/model-versions#modelversion">Model<wbr />Version</a></code> | Required | — |
  | <span id="create-default-for-version--overrides" /><code className="python-reference-parameter">\*\*overrides</code> | —<br /><a href="/api-reference/python/tabpfn/regressor/configuration#constructor"><code>TabPFNRegressor</code> options</a> | — | — |
</div>

**Returns**

<div className="python-reference-table python-reference-returns">
  | Type | Description |
  | - | - |
  | <code className="python-reference-type"><a href="/api-reference/python/tabpfn/regressor/configuration#constructor">Self</a></code> | — |
</div>

***

<div className="python-reference-heading">
  <h2 id="estimator-type">
    `TabPFNRegressor.estimator_type`
  </h2>

  <a className="python-reference-source" href="https://github.com/PriorLabs/TabPFN/blob/c70b6ef0488858d32244c52222abfc0c5be207d6/src/tabpfn/regressor.py#L679" aria-label="View source for TabPFNRegressor.estimator_type"><span aria-hidden="true">\</></span> View source <span aria-hidden="true">↗</span></a>
</div>

The type of the model.

**Returns**

<div className="python-reference-table python-reference-returns">
  | Type | Description |
  | - | - |
  | <code className="python-reference-type">Literal\["regressor"]</code> | — |
</div>

***

<div className="python-reference-heading">
  <h2 id="model">
    `TabPFNRegressor.model_`
  </h2>

  <a className="python-reference-source" href="https://github.com/PriorLabs/TabPFN/blob/c70b6ef0488858d32244c52222abfc0c5be207d6/src/tabpfn/regressor.py#L684" aria-label="View source for TabPFNRegressor.model_"><span aria-hidden="true">\</></span> View source <span aria-hidden="true">↗</span></a>
</div>

The model used for inference.

This is set after the model is loaded and initialized.

**Returns**

<div className="python-reference-table python-reference-returns">
  | Type | Description |
  | - | - |
  | <code className="python-reference-type">Architecture</code> | — |
</div>

***

<div className="python-reference-heading">
  <h2 id="norm-bardist">
    `TabPFNRegressor.norm_bardist_`
  </h2>

  <a className="python-reference-source" href="https://github.com/PriorLabs/TabPFN/blob/c70b6ef0488858d32244c52222abfc0c5be207d6/src/tabpfn/regressor.py#L702" aria-label="View source for TabPFNRegressor.norm_bardist_"><span aria-hidden="true">\</></span> View source <span aria-hidden="true">↗</span></a>
</div>

WARNING: DEPRECATED. Please use `raw_space_bardist_` instead.
This attribute will be removed in a future version.

**Returns**

<div className="python-reference-table python-reference-returns">
  | Type | Description |
  | - | - |
  | <code className="python-reference-type">Full<wbr />Support<wbr />Bar<wbr />Distribution</code> | — |
</div>

***

<div className="python-reference-heading">
  <h2 id="bardist">
    `TabPFNRegressor.bardist_`
  </h2>

  <a className="python-reference-source" href="https://github.com/PriorLabs/TabPFN/blob/c70b6ef0488858d32244c52222abfc0c5be207d6/src/tabpfn/regressor.py#L725" aria-label="View source for TabPFNRegressor.bardist_"><span aria-hidden="true">\</></span> View source <span aria-hidden="true">↗</span></a>
</div>

WARNING: DEPRECATED. Please use `znorm_space_bardist_` instead.
This attribute will be removed in a future version.

**Returns**

<div className="python-reference-table python-reference-returns">
  | Type | Description |
  | - | - |
  | <code className="python-reference-type">Full<wbr />Support<wbr />Bar<wbr />Distribution</code> | — |
</div>

***

<div className="python-reference-heading">
  <h2 id="get-inference-config">
    `TabPFNRegressor.get_inference_config`
  </h2>

  <a className="python-reference-source" href="https://github.com/PriorLabs/TabPFN/blob/c70b6ef0488858d32244c52222abfc0c5be207d6/src/tabpfn/regressor.py#L747" aria-label="View source for TabPFNRegressor.get_inference_config"><span aria-hidden="true">\</></span> View source <span aria-hidden="true">↗</span></a>
</div>

Load the model if needed and return the active inference config.

Loads the model checkpoint without requiring fit data so the config can be
inspected before calling [`fit()`](/api-reference/python/tabpfn/regressor/fitting-and-prediction#fit). Any `inference_config` override
passed to the constructor is considered.

```python theme={null}
TabPFNRegressor.get_inference_config() -> InferenceConfig
```

**Returns**

<div className="python-reference-table python-reference-returns">
  | Type | Description |
  | - | - |
  | <code className="python-reference-type"><a href="/api-reference/python/tabpfn/inference-config#inferenceconfig">Inference<wbr />Config</a></code> | A deep copy of the active inference config. |
</div>

## Scikit-learn methods

See [`get_params`](https://scikit-learn.org/stable/modules/generated/sklearn.base.BaseEstimator.html#sklearn.base.BaseEstimator.get_params) and [`set_params`](https://scikit-learn.org/stable/modules/generated/sklearn.base.BaseEstimator.html#sklearn.base.BaseEstimator.set_params).


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