Preprocessing configuration

`ClassifierEnsembleConfig`

Configuration for a classifier ensemble member.

ClassifierEnsembleConfig(
    preprocess_config: PreprocessorConfig,
    add_fingerprint_feature: bool,
    polynomial_features: Literal["no", "all"] | int,
    feature_shift_count: int,
    feature_shift_decoder: Literal["shuffle", "rotate"] | None,
    outlier_removal_std: float | None,
    _model_index: int,
    passthrough_inf: bool,
    class_permutation: np.ndarray | None,
)

Fields

Name	Type	Default	Description
`preprocess_config`	`PreprocessorConfig`	Required	Preprocessor configuration to use.
`add_fingerprint_feature`	`bool`	Required	Whether to add fingerprint features.
`polynomial_features`	`Literal[“no”, “all”] \| int`	Required	Maximum number of polynomial features to add, if any.
`feature_shift_count`	`int`	Required	How much to shift the features columns.
`feature_shift_decoder`	`Literal[“shuffle”, “rotate”] \| None`	Required	How to shift features.
`outlier_removal_std`	`float \| None`	Required	Number of standard deviations from the mean to consider a sample an outlier. If `None`, no outliers are removed.
`passthrough_inf`	`bool`	Required	Whether to pass infinite values through to the model. When `True`, the preprocessing pipeline replaces infinities with NaN before preprocessing and restores them afterwards.
`class_permutation`	`np.ndarray \| None`	Required	—

`EnsembleConfig`

View source

Configuration for an ensemble member.

EnsembleConfig(
    preprocess_config: PreprocessorConfig,
    add_fingerprint_feature: bool,
    polynomial_features: Literal["no", "all"] | int,
    feature_shift_count: int,
    feature_shift_decoder: Literal["shuffle", "rotate"] | None,
    outlier_removal_std: float | None,
    _model_index: int,
    passthrough_inf: bool,
)

Fields

Name	Type	Default	Description
`preprocess_config`	`PreprocessorConfig`	Required	Preprocessor configuration to use.
`add_fingerprint_feature`	`bool`	Required	Whether to add fingerprint features.
`polynomial_features`	`Literal[“no”, “all”] \| int`	Required	Maximum number of polynomial features to add, if any.
`feature_shift_count`	`int`	Required	How much to shift the features columns.
`feature_shift_decoder`	`Literal[“shuffle”, “rotate”] \| None`	Required	How to shift features.
`outlier_removal_std`	`float \| None`	Required	Number of standard deviations from the mean to consider a sample an outlier. If `None`, no outliers are removed.
`passthrough_inf`	`bool`	Required	Whether to pass infinite values through to the model. When `True`, the preprocessing pipeline replaces infinities with NaN before preprocessing and restores them afterwards.

`FeatureSubsamplingMethod`

View source

Method for subsampling features if dataset exceeds max_features_per_estimator. Fields

Name	Type	Default	Description
`BALANCED`	—	`"balanced"`	—
`RANDOM`	—	`"random"`	—
`CONSTANT_AND_BALANCED`	—	`"constant_and_balanced"`	—
`GINI_FEATURE_IMPORTANCE`	—	`"gini_feature_importance"`	—
`AUTO`	—	`"auto"`	—

`PreprocessorConfig`

View source

Configuration for data preprocessing.

PreprocessorConfig(
    name: Literal["power", "safepower", "power_box", "safepower_box", "quantile_uni_coarse", "quantile_norm_coarse", "quantile_uni", "quantile_norm", "quantile_uni_fine", "quantile_norm_fine", "quantile_uni_extrapolate", "squashing_scaler_default", "squashing_scaler_max10", "robust", "kdi", "none", "kdi_random_alpha", "kdi_uni", "kdi_random_alpha_uni", "adaptive", "norm_and_kdi", "kdi_alpha_0.3_uni", "kdi_alpha_0.5_uni", "kdi_alpha_0.8_uni", "kdi_alpha_1.0_uni", "kdi_alpha_1.2_uni", "kdi_alpha_1.5_uni", "kdi_alpha_2.0_uni", "kdi_alpha_3.0_uni", "kdi_alpha_5.0_uni", "kdi_alpha_0.3", "kdi_alpha_0.5", "kdi_alpha_0.8", "kdi_alpha_1.0", "kdi_alpha_1.2", "kdi_alpha_1.5", "kdi_alpha_2.0", "kdi_alpha_3.0", "kdi_alpha_5.0"],
    categorical_name: Literal["none", "numeric", "onehot", "ordinal", "ordinal_shuffled", "ordinal_very_common_categories_shuffled"] = "none",
    append_original: bool | Literal["auto"] = False,
    max_features_per_estimator: int = 500,
    global_transformer_name: Literal["svd", "svd_quarter_components"] | None = None,
    max_onehot_cardinality: int | None = None,
    differentiable: bool = False,
)

Fields

Name	Type	Default	Description
`name`	Literal[“power”, “safepower”, “power_box”, “safepower_box”, “quantile_uni_coarse”, “quantile_norm_coarse”, “quantile_uni”, “quantile_norm”, “quantile_uni_fine”, “quantile_norm_fine”, “quantile_uni_extrapolate”, “squashing_scaler_default”, “squashing_scaler_max10”, “robust”, “kdi”, “none”, “kdi_random_alpha”, “kdi_uni”, “kdi_random_alpha_uni”, “adaptive”, “norm_and_kdi”, “kdi_alpha_0.3_uni”, “kdi_alpha_0.5_uni”, “kdi_alpha_0.8_uni”, “kdi_alpha_1.0_uni”, “kdi_alpha_1.2_uni”, “kdi_alpha_1.5_uni”, “kdi_alpha_2.0_uni”, “kdi_alpha_3.0_uni”, “kdi_alpha_5.0_uni”, “kdi_alpha_0.3”, “kdi_alpha_0.5”, “kdi_alpha_0.8”, “kdi_alpha_1.0”, “kdi_alpha_1.2”, “kdi_alpha_1.5”, “kdi_alpha_2.0”, “kdi_alpha_3.0”, “kdi_alpha_5.0”]	Required	Name of the preprocessor.
`categorical_name`	`Literal[“none”, “numeric”, “onehot”, “ordinal”, “ordinal_shuffled”, “ordinal_very_common_categories_shuffled”]`	`"none"`	Name of the categorical encoding method. Options: “none”, “numeric”, “onehot”, “ordinal”, “ordinal_shuffled”, “none”.
`append_original`	`bool \| Literal[“auto”]`	`False`	—
`max_features_per_estimator`	`int`	`500`	Maximum number of features per estimator. In case the dataset has more features than this, the features are subsampled for each estimator independently. If append to original is set to `True` we can still have more features.
`global_transformer_name`	`Literal[“svd”, “svd_quarter_components”] \| None`	`None`	Name of the global transformer to use.
`max_onehot_cardinality`	`int \| None`	`None`	Maximum number of unique values a categorical feature can have to be one-hot encoded. Features with higher cardinality are passed through unchanged to ordinal encoding. If `None`, all categorical features are one-hot encoded.
`differentiable`	`bool`	`False`	—

`RegressorEnsembleConfig`

View source

Configuration for a regression ensemble member.

RegressorEnsembleConfig(
    preprocess_config: PreprocessorConfig,
    add_fingerprint_feature: bool,
    polynomial_features: Literal["no", "all"] | int,
    feature_shift_count: int,
    feature_shift_decoder: Literal["shuffle", "rotate"] | None,
    outlier_removal_std: float | None,
    _model_index: int,
    passthrough_inf: bool,
    target_transform: TransformerMixin | Pipeline | None,
)

Fields

Name	Type	Default	Description
`preprocess_config`	`PreprocessorConfig`	Required	Preprocessor configuration to use.
`add_fingerprint_feature`	`bool`	Required	Whether to add fingerprint features.
`polynomial_features`	`Literal[“no”, “all”] \| int`	Required	Maximum number of polynomial features to add, if any.
`feature_shift_count`	`int`	Required	How much to shift the features columns.
`feature_shift_decoder`	`Literal[“shuffle”, “rotate”] \| None`	Required	How to shift features.
`outlier_removal_std`	`float \| None`	Required	Number of standard deviations from the mean to consider a sample an outlier. If `None`, no outliers are removed.
`passthrough_inf`	`bool`	Required	Whether to pass infinite values through to the model. When `True`, the preprocessing pipeline replaces infinities with NaN before preprocessing and restores them afterwards.
`target_transform`	`TransformerMixin \| Pipeline \| None`	Required	—

`SampleSubsamplingMethod`

View source

Method for subsampling rows per estimator when SUBSAMPLE_SAMPLES is set. Fields

Name	Type	Default	Description
`AUTO`	—	`"auto"`	—
`BALANCED`	—	`"balanced"`	—
`STRATIFIED`	—	`"stratified"`	—
`MAJORITY_DOWNSAMPLE`	—	`"majority_downsample"`	—

Reference

tabpfn

tabpfn-client

tabpfn-extensions

REST API

Preprocessing configuration

`ClassifierEnsembleConfig`

`EnsembleConfig`

`FeatureSubsamplingMethod`

`PreprocessorConfig`

`RegressorEnsembleConfig`

`SampleSubsamplingMethod`

​ClassifierEnsembleConfig

​EnsembleConfig

​FeatureSubsamplingMethod

​PreprocessorConfig

​RegressorEnsembleConfig

​SampleSubsamplingMethod

`ClassifierEnsembleConfig`

`EnsembleConfig`

`FeatureSubsamplingMethod`

`PreprocessorConfig`

`RegressorEnsembleConfig`

`SampleSubsamplingMethod`