dataeval 0.72.1__py3-none-any.whl → 0.72.2__py3-none-any.whl

dataeval/{_internal/detectors → detectors/linters}/clusterer.py

@@ -1,16 +1,18 @@
 from __future__ import annotations
 
+__all__ = ["ClustererOutput", "Clusterer"]
+
 from dataclasses import dataclass
-from typing import Iterable, NamedTuple, cast
+from typing import Any, Iterable, NamedTuple, cast
 
 import numpy as np
 from numpy.typing import ArrayLike, NDArray
 from scipy.cluster.hierarchy import linkage
 from scipy.spatial.distance import pdist, squareform
 
-from dataeval._internal.interop import to_numpy
-from dataeval._internal.metrics.utils import flatten
-from dataeval._internal.output import OutputMetadata, set_metadata
+from dataeval.interop import to_numpy
+from dataeval.output import OutputMetadata, set_metadata
+from dataeval.utils.shared import flatten
 
 
 @dataclass(frozen=True)
@@ -36,7 +38,7 @@ class ClustererOutput(OutputMetadata):
     potential_duplicates: list[list[int]]
 
 
-def extend_linkage(link_arr: NDArray) -> NDArray:
+def _extend_linkage(link_arr: NDArray) -> NDArray:
     """
     Adds a column to the linkage matrix link_arr that tracks the new id assigned
     to each row
@@ -60,10 +62,10 @@ def extend_linkage(link_arr: NDArray) -> NDArray:
     return arr
 
 
-class Cluster:
+class _Cluster:
     __slots__ = "merged", "samples", "sample_dist", "is_copy", "count", "dist_avg", "dist_std", "out1", "out2"
 
-    def __init__(self, merged: int, samples: NDArray, sample_dist: float | NDArray, is_copy: bool = False):
+    def __init__(self, merged: int, samples: NDArray, sample_dist: float | NDArray, is_copy: bool = False) -> None:
         self.merged = merged
         self.samples = np.array(samples, dtype=np.int32)
         self.sample_dist = np.array([sample_dist] if np.isscalar(sample_dist) else sample_dist)
@@ -85,8 +87,8 @@ class Cluster:
             self.out1 = dist > out1
             self.out2 = dist > out2
 
-    def copy(self) -> Cluster:
-        return Cluster(False, self.samples, self.sample_dist, True)
+    def copy(self) -> _Cluster:
+        return _Cluster(False, self.samples, self.sample_dist, True)
 
     def __repr__(self) -> str:
         _params = {
@@ -98,32 +100,32 @@ class Cluster:
         return f"{self.__class__.__name__}(**{repr(_params)})"
 
 
-class Clusters(dict[int, dict[int, Cluster]]):
-    def __init__(self, *args, **kwargs):
-        super().__init__(*args, **kwargs)
+class _Clusters(dict[int, dict[int, _Cluster]]):
+    def __init__(self, *args: dict[int, dict[int, _Cluster]]) -> None:
+        super().__init__(*args)
         self.max_level: int = 1
 
 
-class ClusterPosition(NamedTuple):
+class _ClusterPosition(NamedTuple):
     """Keeps track of a cluster's level and ID"""
 
     level: int
     cid: int
 
 
-class ClusterMergeEntry:
+class _ClusterMergeEntry:
     __slots__ = "level", "outer_cluster", "inner_cluster", "status"
 
-    def __init__(self, level: int, outer_cluster: int, inner_cluster: int, status: int):
+    def __init__(self, level: int, outer_cluster: int, inner_cluster: int, status: int) -> None:
         self.level = level
         self.outer_cluster = outer_cluster
         self.inner_cluster = inner_cluster
         self.status = status
 
-    def __lt__(self, value: ClusterMergeEntry) -> bool:
+    def __lt__(self, value: _ClusterMergeEntry) -> bool:
         return self.level.__lt__(value.level)
 
-    def __gt__(self, value: ClusterMergeEntry) -> bool:
+    def __gt__(self, value: _ClusterMergeEntry) -> bool:
         return self.level.__gt__(value.level)
 
 
@@ -153,36 +155,36 @@ class Clusterer:
     >>> cluster = Clusterer(dataset)
     """
 
-    def __init__(self, dataset: ArrayLike):
+    def __init__(self, dataset: ArrayLike) -> None:
         # Allows an update to dataset to reset the state rather than instantiate a new class
         self._on_init(dataset)
 
     def _on_init(self, dataset: ArrayLike):
-        self._data: NDArray = flatten(to_numpy(dataset))
+        self._data: NDArray[Any] = flatten(to_numpy(dataset))
         self._validate_data(self._data)
         self._num_samples = len(self._data)
 
-        self._darr: NDArray = pdist(self._data, metric="euclidean")
-        self._sqdmat: NDArray = squareform(self._darr)
-        self._larr: NDArray = extend_linkage(linkage(self._darr))
+        self._darr: NDArray[np.floating[Any]] = pdist(self._data, metric="euclidean")
+        self._sqdmat: NDArray[np.floating[Any]] = squareform(self._darr)
+        self._larr: NDArray[np.floating[Any]] = _extend_linkage(linkage(self._darr))
         self._max_clusters: int = np.count_nonzero(self._larr[:, 3] == 2)
 
         min_num = int(self._num_samples * 0.05)
-        self._min_num_samples_per_cluster = min(max(2, min_num), 100)
+        self._min_num_samples_per_cluster: int = min(max(2, min_num), 100)
 
-        self._clusters = None
-        self._last_good_merge_levels = None
+        self._clusters: _Clusters | None = None
+        self._last_good_merge_levels: dict[int, int] | None = None
 
     @property
-    def data(self) -> NDArray:
+    def data(self) -> NDArray[Any]:
         return self._data
 
     @data.setter
-    def data(self, x: ArrayLike):
+    def data(self, x: ArrayLike) -> None:
         self._on_init(x)
 
     @property
-    def clusters(self) -> Clusters:
+    def clusters(self) -> _Clusters:
         if self._clusters is None:
             self._clusters = self._create_clusters()
         return self._clusters
@@ -209,11 +211,11 @@ class Clusterer:
         if features < 1:
             raise ValueError(f"Samples should have at least 1 feature; got {features}")
 
-    def _create_clusters(self) -> Clusters:
+    def _create_clusters(self) -> _Clusters:
         """Generates clusters based on linkage matrix"""
         next_cluster_id = 0
-        cluster_map: dict[int, ClusterPosition] = {}  # Dictionary to associate new cluster ids with actual clusters
-        clusters: Clusters = Clusters()
+        cluster_map: dict[int, _ClusterPosition] = {}  # Dictionary to associate new cluster ids with actual clusters
+        clusters: _Clusters = _Clusters()
 
         # Walking through the linkage array to generate clusters
         for arr_i in self._larr:
@@ -240,7 +242,7 @@ class Clusterer:
                 # Update clusters to include previously skipped levels
                 clusters = self._fill_levels(clusters, left, right)
             elif left or right:
-                child, other_id = cast(tuple[ClusterPosition, int], (left, right_id) if left else (right, left_id))
+                child, other_id = cast(tuple[_ClusterPosition, int], (left, right_id) if left else (right, left_id))
                 cc = clusters[child.level][child.cid]
                 samples = np.concatenate([cc.samples, [other_id]])
                 sample_dist = np.concatenate([cc.sample_dist, sample_dist])
@@ -254,12 +256,12 @@ class Clusterer:
             if level not in clusters:
                 clusters[level] = {}
 
-            clusters[level][cid] = Cluster(merged, samples, sample_dist)
-            cluster_map[int(arr_i[-1])] = ClusterPosition(level, cid)
+            clusters[level][cid] = _Cluster(merged, samples, sample_dist)
+            cluster_map[int(arr_i[-1])] = _ClusterPosition(level, cid)
 
         return clusters
 
-    def _fill_levels(self, clusters: Clusters, left: ClusterPosition, right: ClusterPosition) -> Clusters:
+    def _fill_levels(self, clusters: _Clusters, left: _ClusterPosition, right: _ClusterPosition) -> _Clusters:
         # Sets each level's cluster info if it does not exist
         if left.level != right.level:
             (level, cid), max_level = (left, right[0]) if left[0] < right[0] else (right, left[0])
@@ -312,7 +314,7 @@ class Clusterer:
         mask2 = mask2_vals < one_std_check
         return np.logical_or(desired_merge, mask2)
 
-    def _generate_merge_list(self, cluster_matrix: NDArray) -> list[ClusterMergeEntry]:
+    def _generate_merge_list(self, cluster_matrix: NDArray) -> list[_ClusterMergeEntry]:
         """
         Runs through the clusters dictionary determining when clusters merge,
         and how close are those clusters when they merge.
@@ -329,7 +331,7 @@ class Clusterer:
         """
         intra_max = []
         merge_mean = []
-        merge_list: list[ClusterMergeEntry] = []
+        merge_list: list[_ClusterMergeEntry] = []
 
         for level, cluster_set in self.clusters.items():
             for outer_cluster, cluster in cluster_set.items():
@@ -356,7 +358,7 @@ class Clusterer:
                 # Calculate the corresponding distance stats
                 distance_stats_arr = aggregate_func(distances)
                 merge_mean.append(distance_stats_arr)
-                merge_list.append(ClusterMergeEntry(level, outer_cluster, inner_cluster, 0))
+                merge_list.append(_ClusterMergeEntry(level, outer_cluster, inner_cluster, 0))
 
         all_merge_indices = self._calc_merge_indices(merge_mean=merge_mean, intra_max=intra_max)
 
@@ -493,7 +495,7 @@ class Clusterer:
         return exact_dupes, near_dupes
 
     # TODO: Move data input to evaluate from class
-    @set_metadata("dataeval.detectors", ["data"])
+    @set_metadata(["data"])
     def evaluate(self) -> ClustererOutput:
         """Finds and flags indices of the data for Outliers and :term:`duplicates<Duplicates>`
 

dataeval/{_internal/detectors → detectors/linters}/duplicates.py

@@ -1,13 +1,15 @@
 from __future__ import annotations
 
+__all__ = ["DuplicatesOutput", "Duplicates"]
+
 from dataclasses import dataclass
-from typing import Generic, Iterable, Sequence, TypeVar
+from typing import Generic, Iterable, Sequence, TypeVar, overload
 
 from numpy.typing import ArrayLike
 
-from dataeval._internal.detectors.merged_stats import combine_stats, get_dataset_step_from_idx
-from dataeval._internal.metrics.stats.hashstats import HashStatsOutput, hashstats
-from dataeval._internal.output import OutputMetadata, set_metadata
+from dataeval.detectors.linters.merged_stats import combine_stats, get_dataset_step_from_idx
+from dataeval.metrics.stats.hashstats import HashStatsOutput, hashstats
+from dataeval.output import OutputMetadata, set_metadata
 
 DuplicateGroup = list[int]
 DatasetDuplicateGroupMap = dict[int, DuplicateGroup]
@@ -58,7 +60,7 @@ class Duplicates:
     >>> exact_dupes = Duplicates(only_exact=True)
     """
 
-    def __init__(self, only_exact: bool = False):
+    def __init__(self, only_exact: bool = False) -> None:
         self.stats: HashStatsOutput
         self.only_exact = only_exact
 
@@ -81,8 +83,16 @@ class Duplicates:
             "near": sorted(near),
         }
 
-    @set_metadata("dataeval.detectors", ["only_exact"])
-    def from_stats(self, hashes: HashStatsOutput | Sequence[HashStatsOutput]) -> DuplicatesOutput:
+    @overload
+    def from_stats(self, hashes: HashStatsOutput) -> DuplicatesOutput[DuplicateGroup]: ...
+
+    @overload
+    def from_stats(self, hashes: Sequence[HashStatsOutput]) -> DuplicatesOutput[DatasetDuplicateGroupMap]: ...
+
+    @set_metadata(["only_exact"])
+    def from_stats(
+        self, hashes: HashStatsOutput | Sequence[HashStatsOutput]
+    ) -> DuplicatesOutput[DuplicateGroup] | DuplicatesOutput[DatasetDuplicateGroupMap]:
         """
         Returns duplicate image indices for both exact matches and near matches
 
@@ -128,8 +138,8 @@ class Duplicates:
 
         return DuplicatesOutput(**duplicates)
 
-    @set_metadata("dataeval.detectors", ["only_exact"])
-    def evaluate(self, data: Iterable[ArrayLike]) -> DuplicatesOutput:
+    @set_metadata(["only_exact"])
+    def evaluate(self, data: Iterable[ArrayLike]) -> DuplicatesOutput[DuplicateGroup]:
         """
         Returns duplicate image indices for both exact matches and near matches
 

dataeval/{_internal/detectors → detectors/linters}/merged_stats.py

@@ -1,11 +1,13 @@
 from __future__ import annotations
 
+__all__ = []
+
 from copy import deepcopy
 from typing import Sequence, TypeVar
 
 import numpy as np
 
-from dataeval._internal.metrics.stats.base import BaseStatsOutput
+from dataeval.metrics.stats.base import BaseStatsOutput
 
 TStatsOutput = TypeVar("TStatsOutput", bound=BaseStatsOutput)
 

dataeval/{_internal/detectors → detectors/linters}/outliers.py

@@ -1,18 +1,20 @@
 from __future__ import annotations
 
+__all__ = ["OutliersOutput", "Outliers"]
+
 from dataclasses import dataclass
 from typing import Generic, Iterable, Literal, Sequence, TypeVar, Union, overload
 
 import numpy as np
 from numpy.typing import ArrayLike, NDArray
 
-from dataeval._internal.detectors.merged_stats import combine_stats, get_dataset_step_from_idx
-from dataeval._internal.metrics.stats.base import BOX_COUNT, SOURCE_INDEX
-from dataeval._internal.metrics.stats.datasetstats import DatasetStatsOutput, datasetstats
-from dataeval._internal.metrics.stats.dimensionstats import DimensionStatsOutput
-from dataeval._internal.metrics.stats.pixelstats import PixelStatsOutput
-from dataeval._internal.metrics.stats.visualstats import VisualStatsOutput
-from dataeval._internal.output import OutputMetadata, set_metadata
+from dataeval.detectors.linters.merged_stats import combine_stats, get_dataset_step_from_idx
+from dataeval.metrics.stats.base import BOX_COUNT, SOURCE_INDEX
+from dataeval.metrics.stats.datasetstats import DatasetStatsOutput, datasetstats
+from dataeval.metrics.stats.dimensionstats import DimensionStatsOutput
+from dataeval.metrics.stats.pixelstats import PixelStatsOutput
+from dataeval.metrics.stats.visualstats import VisualStatsOutput
+from dataeval.output import OutputMetadata, set_metadata
 
 IndexIssueMap = dict[int, dict[str, float]]
 OutlierStatsOutput = Union[DimensionStatsOutput, PixelStatsOutput, VisualStatsOutput]
@@ -37,7 +39,7 @@ class OutliersOutput(Generic[TIndexIssueMap], OutputMetadata):
 
     issues: TIndexIssueMap
 
-    def __len__(self):
+    def __len__(self) -> int:
         if isinstance(self.issues, dict):
             return len(self.issues)
         else:
@@ -157,10 +159,10 @@ class Outliers:
     @overload
     def from_stats(self, stats: Sequence[OutlierStatsOutput]) -> OutliersOutput[list[IndexIssueMap]]: ...
 
-    @set_metadata("dataeval.detectors", ["outlier_method", "outlier_threshold"])
+    @set_metadata(["outlier_method", "outlier_threshold"])
     def from_stats(
         self, stats: OutlierStatsOutput | DatasetStatsOutput | Sequence[OutlierStatsOutput]
-    ) -> OutliersOutput:
+    ) -> OutliersOutput[IndexIssueMap] | OutliersOutput[list[IndexIssueMap]]:
         """
         Returns indices of Outliers with the issues identified for each
 
@@ -195,7 +197,7 @@ class Outliers:
         {}
         """  # noqa: E501
         if isinstance(stats, DatasetStatsOutput):
-            outliers = self._get_outliers({k: v for o in stats.outputs() for k, v in o.dict().items()})
+            outliers = self._get_outliers({k: v for o in stats._outputs() for k, v in o.dict().items()})
             return OutliersOutput(outliers)
 
         if isinstance(stats, (DimensionStatsOutput, PixelStatsOutput, VisualStatsOutput)):
@@ -226,16 +228,7 @@ class Outliers:
 
         return OutliersOutput(output_list)
 
-    @set_metadata(
-        "dataeval.detectors",
-        [
-            "use_dimension",
-            "use_pixel",
-            "use_visual",
-            "outlier_method",
-            "outlier_threshold",
-        ],
-    )
+    @set_metadata(["use_dimension", "use_pixel", "use_visual", "outlier_method", "outlier_threshold"])
     def evaluate(self, data: Iterable[ArrayLike]) -> OutliersOutput[IndexIssueMap]:
         """
         Returns indices of Outliers with the issues identified for each

dataeval/detectors/ood/__init__.py

@@ -5,11 +5,11 @@ Out-of-distribution (OOD)` detectors identify data that is different from the da
 from dataeval import _IS_TENSORFLOW_AVAILABLE
 
 if _IS_TENSORFLOW_AVAILABLE:  # pragma: no cover
-    from dataeval._internal.detectors.ood.ae import OOD_AE
-    from dataeval._internal.detectors.ood.aegmm import OOD_AEGMM
-    from dataeval._internal.detectors.ood.base import OODOutput, OODScoreOutput
-    from dataeval._internal.detectors.ood.llr import OOD_LLR
-    from dataeval._internal.detectors.ood.vae import OOD_VAE
-    from dataeval._internal.detectors.ood.vaegmm import OOD_VAEGMM
+    from dataeval.detectors.ood.ae import OOD_AE
+    from dataeval.detectors.ood.aegmm import OOD_AEGMM
+    from dataeval.detectors.ood.base import OODOutput, OODScoreOutput
+    from dataeval.detectors.ood.llr import OOD_LLR
+    from dataeval.detectors.ood.vae import OOD_VAE
+    from dataeval.detectors.ood.vaegmm import OOD_VAEGMM
 
     __all__ = ["OOD_AE", "OOD_AEGMM", "OOD_LLR", "OOD_VAE", "OOD_VAEGMM", "OODOutput", "OODScoreOutput"]

dataeval/{_internal/detectors → detectors}/ood/ae.py

@@ -8,6 +8,8 @@ Licensed under Apache Software License (Apache 2.0)
 
 from __future__ import annotations
 
+__all__ = ["OOD_AE"]
+
 from typing import Callable
 
 import numpy as np
@@ -15,11 +17,10 @@ import tensorflow as tf
 import tf_keras as keras
 from numpy.typing import ArrayLike
 
-from dataeval._internal.detectors.ood.base import OODBase, OODScoreOutput
-from dataeval._internal.interop import as_numpy
-from dataeval._internal.models.tensorflow.autoencoder import AE
-from dataeval._internal.models.tensorflow.utils import predict_batch
-from dataeval._internal.output import set_metadata
+from dataeval.detectors.ood.base import OODBase, OODScoreOutput
+from dataeval.interop import as_numpy
+from dataeval.utils.tensorflow._internal.autoencoder import AE
+from dataeval.utils.tensorflow._internal.utils import predict_batch
 
 
 class OOD_AE(OODBase):
@@ -49,8 +50,7 @@ class OOD_AE(OODBase):
             loss_fn = keras.losses.MeanSquaredError()
         super().fit(as_numpy(x_ref), threshold_perc, loss_fn, optimizer, epochs, batch_size, verbose)
 
-    @set_metadata("dataeval.detectors")
-    def score(self, X: ArrayLike, batch_size: int = int(1e10)) -> OODScoreOutput:
+    def _score(self, X: ArrayLike, batch_size: int = int(1e10)) -> OODScoreOutput:
         self._validate(X := as_numpy(X))
 
         # reconstruct instances

dataeval/{_internal/detectors → detectors}/ood/aegmm.py

@@ -8,19 +8,20 @@ Licensed under Apache Software License (Apache 2.0)
 
 from __future__ import annotations
 
+__all__ = ["OOD_AEGMM"]
+
 from typing import Callable
 
 import tensorflow as tf
 import tf_keras as keras
 from numpy.typing import ArrayLike
 
-from dataeval._internal.detectors.ood.base import OODGMMBase, OODScoreOutput
-from dataeval._internal.interop import to_numpy
-from dataeval._internal.models.tensorflow.autoencoder import AEGMM
-from dataeval._internal.models.tensorflow.gmm import gmm_energy
-from dataeval._internal.models.tensorflow.losses import LossGMM
-from dataeval._internal.models.tensorflow.utils import predict_batch
-from dataeval._internal.output import set_metadata
+from dataeval.detectors.ood.base import OODGMMBase, OODScoreOutput
+from dataeval.interop import to_numpy
+from dataeval.utils.tensorflow._internal.autoencoder import AEGMM
+from dataeval.utils.tensorflow._internal.gmm import gmm_energy
+from dataeval.utils.tensorflow._internal.loss import LossGMM
+from dataeval.utils.tensorflow._internal.utils import predict_batch
 
 
 class OOD_AEGMM(OODGMMBase):
@@ -50,28 +51,7 @@ class OOD_AEGMM(OODGMMBase):
             loss_fn = LossGMM()
         super().fit(x_ref, threshold_perc, loss_fn, optimizer, epochs, batch_size, verbose)
 
-    @set_metadata("dataeval.detectors")
-    def score(self, X: ArrayLike, batch_size: int = int(1e10)) -> OODScoreOutput:
-        """
-        Compute the :term:`out of distribution<Out-of-distribution (OOD)>` score for a given dataset.
-
-        Parameters
-        ----------
-        X : ArrayLike
-            Input data to score.
-        batch_size : int, default 1e10
-            Number of instances to process in each batch.
-            Use a smaller batch size if your dataset is large or if you encounter memory issues.
-
-        Returns
-        -------
-        OODScoreOutput
-            An object containing the instance-level OOD score.
-
-        Note
-        ----
-        This model does not produce a feature level score like the OOD_AE or OOD_VAE models.
-        """
+    def _score(self, X: ArrayLike, batch_size: int = int(1e10)) -> OODScoreOutput:
         self._validate(X := to_numpy(X))
         _, z, _ = predict_batch(X, self.model, batch_size=batch_size)
         energy, _ = gmm_energy(z, self.gmm_params, return_mean=False)

dataeval/{_internal/detectors → detectors}/ood/base.py

@@ -8,6 +8,8 @@ Licensed under Apache Software License (Apache 2.0)
 
 from __future__ import annotations
 
+__all__ = ["OODOutput", "OODScoreOutput"]
+
 from abc import ABC, abstractmethod
 from dataclasses import dataclass
 from typing import Callable, Literal, cast
@@ -17,10 +19,10 @@ import tensorflow as tf
 import tf_keras as keras
 from numpy.typing import ArrayLike, NDArray
 
-from dataeval._internal.interop import to_numpy
-from dataeval._internal.models.tensorflow.gmm import GaussianMixtureModelParams, gmm_params
-from dataeval._internal.models.tensorflow.trainer import trainer
-from dataeval._internal.output import OutputMetadata, set_metadata
+from dataeval.interop import to_numpy
+from dataeval.output import OutputMetadata, set_metadata
+from dataeval.utils.tensorflow._internal.gmm import GaussianMixtureModelParams, gmm_params
+from dataeval.utils.tensorflow._internal.trainer import trainer
 
 
 @dataclass(frozen=True)
@@ -61,7 +63,7 @@ class OODScoreOutput(OutputMetadata):
     instance_score: NDArray[np.float32]
     feature_score: NDArray[np.float32] | None = None
 
-    def get(self, ood_type: Literal["instance", "feature"]) -> NDArray:
+    def get(self, ood_type: Literal["instance", "feature"]) -> NDArray[np.float32]:
         """
         Returns either the instance or feature score
 
@@ -107,6 +109,9 @@ class OODBase(ABC):
         self._validate(X)
 
     @abstractmethod
+    def _score(self, X: ArrayLike, batch_size: int = int(1e10)) -> OODScoreOutput: ...
+
+    @set_metadata()
     def score(self, X: ArrayLike, batch_size: int = int(1e10)) -> OODScoreOutput:
         """
         Compute the :term:`out of distribution<Out-of-distribution (OOD)>` scores for a given dataset.
@@ -124,6 +129,7 @@ class OODBase(ABC):
         OODScoreOutput
             An object containing the instance-level and feature-level OOD scores.
         """
+        return self._score(X, batch_size)
 
     def _threshold_score(self, ood_type: Literal["feature", "instance"] = "instance") -> np.floating:
         return np.percentile(self._ref_score.get(ood_type), self._threshold_perc)
@@ -131,12 +137,12 @@ class OODBase(ABC):
     def fit(
         self,
         x_ref: ArrayLike,
-        threshold_perc: float = 100.0,
-        loss_fn: Callable[..., tf.Tensor] | None = None,
-        optimizer: keras.optimizers.Optimizer = keras.optimizers.Adam,
-        epochs: int = 20,
-        batch_size: int = 64,
-        verbose: bool = True,
+        threshold_perc: float,
+        loss_fn: Callable[..., tf.Tensor],
+        optimizer: keras.optimizers.Optimizer,
+        epochs: int,
+        batch_size: int,
+        verbose: bool,
     ) -> None:
         """
         Train the model and infer the threshold value.
@@ -174,7 +180,7 @@ class OODBase(ABC):
         self._ref_score = self.score(x_ref, batch_size)
         self._threshold_perc = threshold_perc
 
-    @set_metadata("dataeval.detectors")
+    @set_metadata()
     def predict(
         self,
         X: ArrayLike,
@@ -218,12 +224,12 @@ class OODGMMBase(OODBase):
     def fit(
         self,
         x_ref: ArrayLike,
-        threshold_perc: float = 100.0,
-        loss_fn: Callable[..., tf.Tensor] | None = None,
-        optimizer: keras.optimizers.Optimizer = keras.optimizers.Adam,
-        epochs: int = 20,
-        batch_size: int = 64,
-        verbose: bool = True,
+        threshold_perc: float,
+        loss_fn: Callable[..., tf.Tensor],
+        optimizer: keras.optimizers.Optimizer,
+        epochs: int,
+        batch_size: int,
+        verbose: bool,
     ) -> None:
         # Train the model
         trainer(