dataeval 0.74.0__py3-none-any.whl → 0.74.2__py3-none-any.whl

dataeval/__init__.py

@@ -1,23 +1,36 @@
-__version__ = "0.74.0"
+__version__ = "0.74.2"
 
+import logging
 from importlib.util import find_spec
 
+logging.getLogger(__name__).addHandler(logging.NullHandler())
+
+
+def log_stderr(level: int = logging.DEBUG) -> None:
+    """
+    Helper for quickly adding a StreamHandler to the logger. Useful for
+    debugging.
+    """
+    import logging
+
+    logger = logging.getLogger(__name__)
+    handler = logging.StreamHandler()
+    handler.setFormatter(logging.Formatter("%(asctime)s %(levelname)s %(message)s"))
+    logger.addHandler(handler)
+    logger.setLevel(level)
+    logger.debug("Added a stderr logging handler to logger: %s", __name__)
+
+
 _IS_TORCH_AVAILABLE = find_spec("torch") is not None
 _IS_TORCHVISION_AVAILABLE = find_spec("torchvision") is not None
-_IS_TENSORFLOW_AVAILABLE = find_spec("tensorflow") is not None and find_spec("tensorflow_probability") is not None
 
 del find_spec
 
 from dataeval import detectors, metrics  # noqa: E402
 
-__all__ = ["detectors", "metrics"]
+__all__ = ["log_stderr", "detectors", "metrics"]
 
 if _IS_TORCH_AVAILABLE:
-    from dataeval import workflows
-
-    __all__ += ["workflows"]
-
-if _IS_TENSORFLOW_AVAILABLE or _IS_TORCH_AVAILABLE:
-    from dataeval import utils
+    from dataeval import utils, workflows
 
-    __all__ += ["utils"]
+    __all__ += ["utils", "workflows"]

dataeval/detectors/__init__.py

@@ -2,14 +2,6 @@
 Detectors can determine if a dataset or individual images in a dataset are indicative of a specific issue.
 """
 
-from dataeval import _IS_TENSORFLOW_AVAILABLE
-from dataeval.detectors import drift, linters
+from dataeval.detectors import drift, linters, ood
 
-__all__ = ["drift", "linters"]
-
-if _IS_TENSORFLOW_AVAILABLE:
-    from dataeval.detectors import ood
-
-    __all__ += ["ood"]
-
-del _IS_TENSORFLOW_AVAILABLE
+__all__ = ["drift", "linters", "ood"]

dataeval/detectors/drift/base.py

@@ -19,7 +19,7 @@ import numpy as np
 from numpy.typing import ArrayLike, NDArray
 
 from dataeval.interop import as_numpy
-from dataeval.output import OutputMetadata, set_metadata
+from dataeval.output import Output, set_metadata
 
 R = TypeVar("R")
 
@@ -43,7 +43,7 @@ class UpdateStrategy(ABC):
 
 
 @dataclass(frozen=True)
-class DriftBaseOutput(OutputMetadata):
+class DriftBaseOutput(Output):
     """
     Base output class for Drift detector classes
 
@@ -387,7 +387,7 @@ class BaseDriftUnivariate(BaseDrift):
         else:
             raise ValueError("`correction` needs to be either `bonferroni` or `fdr`.")
 
-    @set_metadata()
+    @set_metadata
     @preprocess_x
     @update_x_ref
     def predict(

dataeval/detectors/drift/mmd.py

@@ -161,7 +161,7 @@ class DriftMMD(BaseDrift):
         distance_threshold = torch.sort(mmd2_permuted, descending=True).values[idx_threshold]
         return p_val.numpy().item(), mmd2.numpy().item(), distance_threshold.numpy().item()
 
-    @set_metadata()
+    @set_metadata
     @preprocess_x
     @update_x_ref
     def predict(self, x: ArrayLike) -> DriftMMDOutput:

dataeval/detectors/linters/clusterer.py

@@ -11,12 +11,12 @@ from scipy.cluster.hierarchy import linkage
 from scipy.spatial.distance import pdist, squareform
 
 from dataeval.interop import to_numpy
-from dataeval.output import OutputMetadata, set_metadata
+from dataeval.output import Output, set_metadata
 from dataeval.utils.shared import flatten
 
 
 @dataclass(frozen=True)
-class ClustererOutput(OutputMetadata):
+class ClustererOutput(Output):
     """
     Output class for :class:`Clusterer` lint detector
 
@@ -495,7 +495,7 @@ class Clusterer:
         return exact_dupes, near_dupes
 
     # TODO: Move data input to evaluate from class
-    @set_metadata(["data"])
+    @set_metadata(state=["data"])
     def evaluate(self) -> ClustererOutput:
         """Finds and flags indices of the data for Outliers and :term:`duplicates<Duplicates>`
 

dataeval/detectors/linters/duplicates.py

@@ -9,7 +9,7 @@ from numpy.typing import ArrayLike
 
 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
+from dataeval.output import Output, set_metadata
 
 DuplicateGroup = list[int]
 DatasetDuplicateGroupMap = dict[int, DuplicateGroup]
@@ -17,7 +17,7 @@ TIndexCollection = TypeVar("TIndexCollection", DuplicateGroup, DatasetDuplicateG
 
 
 @dataclass(frozen=True)
-class DuplicatesOutput(Generic[TIndexCollection], OutputMetadata):
+class DuplicatesOutput(Generic[TIndexCollection], Output):
     """
     Output class for :class:`Duplicates` lint detector
 
@@ -89,7 +89,7 @@ class Duplicates:
     @overload
     def from_stats(self, hashes: Sequence[HashStatsOutput]) -> DuplicatesOutput[DatasetDuplicateGroupMap]: ...
 
-    @set_metadata(["only_exact"])
+    @set_metadata(state=["only_exact"])
     def from_stats(
         self, hashes: HashStatsOutput | Sequence[HashStatsOutput]
     ) -> DuplicatesOutput[DuplicateGroup] | DuplicatesOutput[DatasetDuplicateGroupMap]:
@@ -138,7 +138,7 @@ class Duplicates:
 
         return DuplicatesOutput(**duplicates)
 
-    @set_metadata(["only_exact"])
+    @set_metadata(state=["only_exact"])
     def evaluate(self, data: Iterable[ArrayLike]) -> DuplicatesOutput[DuplicateGroup]:
         """
         Returns duplicate image indices for both exact matches and near matches

dataeval/detectors/linters/outliers.py

@@ -14,7 +14,7 @@ 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
+from dataeval.output import Output, set_metadata
 
 IndexIssueMap = dict[int, dict[str, float]]
 OutlierStatsOutput = Union[DimensionStatsOutput, PixelStatsOutput, VisualStatsOutput]
@@ -22,7 +22,7 @@ TIndexIssueMap = TypeVar("TIndexIssueMap", IndexIssueMap, list[IndexIssueMap])
 
 
 @dataclass(frozen=True)
-class OutliersOutput(Generic[TIndexIssueMap], OutputMetadata):
+class OutliersOutput(Generic[TIndexIssueMap], Output):
     """
     Output class for :class:`Outliers` lint detector
 
@@ -159,7 +159,7 @@ class Outliers:
     @overload
     def from_stats(self, stats: Sequence[OutlierStatsOutput]) -> OutliersOutput[list[IndexIssueMap]]: ...
 
-    @set_metadata(["outlier_method", "outlier_threshold"])
+    @set_metadata(state=["outlier_method", "outlier_threshold"])
     def from_stats(
         self, stats: OutlierStatsOutput | DatasetStatsOutput | Sequence[OutlierStatsOutput]
     ) -> OutliersOutput[IndexIssueMap] | OutliersOutput[list[IndexIssueMap]]:
@@ -228,7 +228,7 @@ class Outliers:
 
         return OutliersOutput(output_list)
 
-    @set_metadata(["use_dimension", "use_pixel", "use_visual", "outlier_method", "outlier_threshold"])
+    @set_metadata(state=["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

@@ -2,21 +2,14 @@
 Out-of-distribution (OOD)` detectors identify data that is different from the data used to train a particular model.
 """
 
-from dataeval import _IS_TENSORFLOW_AVAILABLE, _IS_TORCH_AVAILABLE
+from dataeval import _IS_TORCH_AVAILABLE
 from dataeval.detectors.ood.base import OODOutput, OODScoreOutput
 
 __all__ = ["OODOutput", "OODScoreOutput"]
 
-if _IS_TENSORFLOW_AVAILABLE:
-    from dataeval.detectors.ood.ae import OOD_AE
-    from dataeval.detectors.ood.aegmm import OOD_AEGMM
-    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"]
-
-elif _IS_TORCH_AVAILABLE:
+if _IS_TORCH_AVAILABLE:
     from dataeval.detectors.ood.ae_torch import OOD_AE
 
-    __all__ += ["OOD_AE", "OODOutput"]
+    __all__ += ["OOD_AE"]
+
+del _IS_TORCH_AVAILABLE

dataeval/detectors/ood/base.py

@@ -18,12 +18,12 @@ import numpy as np
 from numpy.typing import ArrayLike, NDArray
 
 from dataeval.interop import to_numpy
-from dataeval.output import OutputMetadata, set_metadata
+from dataeval.output import Output, set_metadata
 from dataeval.utils.gmm import GaussianMixtureModelParams
 
 
 @dataclass(frozen=True)
-class OODOutput(OutputMetadata):
+class OODOutput(Output):
     """
     Output class for predictions from :class:`OOD_AE`, :class:`OOD_AEGMM`, :class:`OOD_LLR`,
     :class:`OOD_VAE`, and :class:`OOD_VAEGMM` out-of-distribution detectors
@@ -44,7 +44,7 @@ class OODOutput(OutputMetadata):
 
 
 @dataclass(frozen=True)
-class OODScoreOutput(OutputMetadata):
+class OODScoreOutput(Output):
     """
     Output class for instance and feature scores from :class:`OOD_AE`, :class:`OOD_AEGMM`,
     :class:`OOD_LLR`, :class:`OOD_VAE`, and :class:`OOD_VAEGMM` out-of-distribution detectors
@@ -153,7 +153,7 @@ class OODBaseMixin(Generic[TModel], ABC):
     @abstractmethod
     def _score(self, X: ArrayLike, batch_size: int = int(1e10)) -> OODScoreOutput: ...
 
-    @set_metadata()
+    @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.
@@ -176,7 +176,7 @@ class OODBaseMixin(Generic[TModel], ABC):
     def _threshold_score(self, ood_type: Literal["feature", "instance"] = "instance") -> np.floating:
         return np.percentile(self._ref_score.get(ood_type), self._threshold_perc)
 
-    @set_metadata()
+    @set_metadata
     def predict(
         self,
         X: ArrayLike,

dataeval/detectors/ood/metadata_ks_compare.py

@@ -2,7 +2,6 @@ from __future__ import annotations
 
 import numbers
 import warnings
-from dataclasses import dataclass
 from typing import Any, Mapping, NamedTuple
 
 import numpy as np
@@ -10,7 +9,7 @@ from numpy.typing import NDArray
 from scipy.stats import iqr, ks_2samp
 from scipy.stats import wasserstein_distance as emd
 
-from dataeval.output import OutputMetadata, set_metadata
+from dataeval.output import MappingOutput, set_metadata
 
 
 class MetadataKSResult(NamedTuple):
@@ -20,24 +19,24 @@ class MetadataKSResult(NamedTuple):
     pvalue: float
 
 
-@dataclass(frozen=True)
-class KSOutput(OutputMetadata):
+class KSOutput(MappingOutput[str, MetadataKSResult]):
     """
-    Output class for results of ks_2samp featurewise comparisons of new metadata to reference metadata.
+    Output dictionary class for results of ks_2samp featurewise comparisons of new metadata to reference metadata.
 
     Attributes
     ----------
-    mdc : dict[str, dict[str, float]]
-        dict keyed by metadata feature names. Each value contains four floats, which are the KS statistic itself, its
-        location within the range of the reference metadata, the shift of new metadata relative to reference, the
-        p-value from the KS two-sample test.
-
+    key: str
+        Metadata feature names
+    value: NamedTuple[float, float, float, float]
+        Each value contains four floats, which are:
+        - statistic: the KS statistic itself
+        - statistic_location: its location within the range of the reference metadata
+        - shift_magnitude: the shift of new metadata relative to reference
+        - pvalue: the p-value from the KS two-sample test
     """
 
-    mdc: dict[str, MetadataKSResult]
-
 
-@set_metadata()
+@set_metadata
 def meta_distribution_compare(
     md0: Mapping[str, list[Any] | NDArray[Any]], md1: Mapping[str, list[Any] | NDArray[Any]]
 ) -> KSOutput:

dataeval/interop.py

@@ -1,23 +1,31 @@
 from __future__ import annotations
 
+from types import ModuleType
+
+from dataeval.logging import LogMessage
+
 __all__ = ["as_numpy", "to_numpy", "to_numpy_iter"]
 
+import logging
 from importlib import import_module
 from typing import Any, Iterable, Iterator
 
 import numpy as np
 from numpy.typing import ArrayLike, NDArray
 
+_logger = logging.getLogger(__name__)
+
 _MODULE_CACHE = {}
 
 
-def _try_import(module_name):
+def _try_import(module_name) -> ModuleType | None:
     if module_name in _MODULE_CACHE:
         return _MODULE_CACHE[module_name]
 
     try:
         module = import_module(module_name)
     except ImportError:  # pragma: no cover - covered by test_mindeps.py
+        _logger.log(logging.INFO, f"Unable to import {module_name}.")
         module = None
 
     _MODULE_CACHE[module_name] = module
@@ -40,14 +48,18 @@ def to_numpy(array: ArrayLike | None, copy: bool = True) -> NDArray[Any]:
     if array.__class__.__module__.startswith("tensorflow"):
         tf = _try_import("tensorflow")
         if tf and tf.is_tensor(array):
+            _logger.log(logging.INFO, "Converting Tensorflow array to NumPy array.")
             return array.numpy().copy() if copy else array.numpy()  # type: ignore
 
     if array.__class__.__module__.startswith("torch"):
         torch = _try_import("torch")
         if torch and isinstance(array, torch.Tensor):
-            return array.detach().cpu().numpy().copy() if copy else array.detach().cpu().numpy()  # type: ignore
+            _logger.log(logging.INFO, "Converting PyTorch array to NumPy array.")
+            numpy = array.detach().cpu().numpy().copy() if copy else array.detach().cpu().numpy()  # type: ignore
+            _logger.log(logging.DEBUG, LogMessage(lambda: f"{str(array)} -> {str(numpy)}"))
+            return numpy
 
-    return np.array(array, copy=copy)
+    return np.array(array) if copy else np.asarray(array)
 
 
 def to_numpy_iter(iterable: Iterable[ArrayLike]) -> Iterator[NDArray[Any]]:

dataeval/logging.py

@@ -0,0 +1,16 @@
+from typing import Callable
+
+
+class LogMessage:
+    """
+    Deferred message callback for logging expensive messages.
+    """
+
+    def __init__(self, fn: Callable[..., str]):
+        self._fn = fn
+        self._str = None
+
+    def __str__(self) -> str:
+        if self._str is None:
+            self._str = self._fn()
+        return self._str

dataeval/metrics/bias/balance.py

@@ -14,14 +14,14 @@ from sklearn.feature_selection import mutual_info_classif, mutual_info_regressio
 
 from dataeval.metrics.bias.metadata_preprocessing import MetadataOutput
 from dataeval.metrics.bias.metadata_utils import get_counts, heatmap
-from dataeval.output import OutputMetadata, set_metadata
+from dataeval.output import Output, set_metadata
 
 with contextlib.suppress(ImportError):
     from matplotlib.figure import Figure
 
 
 @dataclass(frozen=True)
-class BalanceOutput(OutputMetadata):
+class BalanceOutput(Output):
     """
     Output class for :func:`balance` bias metric
 
@@ -117,7 +117,7 @@ def _validate_num_neighbors(num_neighbors: int) -> int:
     return num_neighbors
 
 
-@set_metadata("dataeval.metrics")
+@set_metadata
 def balance(
     metadata: MetadataOutput,
     num_neighbors: int = 5,

dataeval/metrics/bias/coverage.py

@@ -13,7 +13,7 @@ from scipy.spatial.distance import pdist, squareform
 
 from dataeval.interop import to_numpy
 from dataeval.metrics.bias.metadata_utils import coverage_plot
-from dataeval.output import OutputMetadata, set_metadata
+from dataeval.output import Output, set_metadata
 from dataeval.utils.shared import flatten
 
 with contextlib.suppress(ImportError):
@@ -21,7 +21,7 @@ with contextlib.suppress(ImportError):
 
 
 @dataclass(frozen=True)
-class CoverageOutput(OutputMetadata):
+class CoverageOutput(Output):
     """
     Output class for :func:`coverage` :term:`bias<Bias>` metric
 
@@ -67,7 +67,7 @@ class CoverageOutput(OutputMetadata):
         return fig
 
 
-@set_metadata()
+@set_metadata
 def coverage(
     embeddings: ArrayLike,
     radius_type: Literal["adaptive", "naive"] = "adaptive",

dataeval/metrics/bias/diversity.py

@@ -12,7 +12,7 @@ from numpy.typing import ArrayLike, NDArray
 
 from dataeval.metrics.bias.metadata_preprocessing import MetadataOutput
 from dataeval.metrics.bias.metadata_utils import diversity_bar_plot, get_counts, heatmap
-from dataeval.output import OutputMetadata, set_metadata
+from dataeval.output import Output, set_metadata
 from dataeval.utils.shared import get_method
 
 with contextlib.suppress(ImportError):
@@ -20,7 +20,7 @@ with contextlib.suppress(ImportError):
 
 
 @dataclass(frozen=True)
-class DiversityOutput(OutputMetadata):
+class DiversityOutput(Output):
     """
     Output class for :func:`diversity` :term:`bias<Bias>` metric
 
@@ -163,7 +163,7 @@ def diversity_simpson(
     return ev_index
 
 
-@set_metadata()
+@set_metadata
 def diversity(
     metadata: MetadataOutput,
     method: Literal["simpson", "shannon"] = "simpson",

dataeval/metrics/bias/metadata_preprocessing.py

@@ -11,7 +11,7 @@ from numpy.typing import ArrayLike, NDArray
 from scipy.stats import wasserstein_distance as wd
 
 from dataeval.interop import as_numpy, to_numpy
-from dataeval.output import OutputMetadata, set_metadata
+from dataeval.output import Output, set_metadata
 from dataeval.utils.metadata import merge_metadata
 
 TNum = TypeVar("TNum", int, float)
@@ -20,7 +20,7 @@ CONTINUOUS_MIN_SAMPLE_SIZE = 20
 
 
 @dataclass(frozen=True)
-class MetadataOutput(OutputMetadata):
+class MetadataOutput(Output):
     """
     Output class for :func:`metadata_binning` function
 
@@ -51,7 +51,7 @@ class MetadataOutput(OutputMetadata):
     total_num_factors: int
 
 
-@set_metadata()
+@set_metadata
 def metadata_preprocessing(
     raw_metadata: Iterable[Mapping[str, Any]],
     class_labels: ArrayLike | str,

dataeval/metrics/bias/parity.py

@@ -13,13 +13,13 @@ from scipy.stats.contingency import chi2_contingency, crosstab
 
 from dataeval.interop import as_numpy, to_numpy
 from dataeval.metrics.bias.metadata_preprocessing import MetadataOutput
-from dataeval.output import OutputMetadata, set_metadata
+from dataeval.output import Output, set_metadata
 
 TData = TypeVar("TData", np.float64, NDArray[np.float64])
 
 
 @dataclass(frozen=True)
-class ParityOutput(Generic[TData], OutputMetadata):
+class ParityOutput(Generic[TData], Output):
     """
     Output class for :func:`parity` and :func:`label_parity` :term:`bias<Bias>` metrics
 
@@ -116,7 +116,7 @@ def validate_dist(label_dist: NDArray[Any], label_name: str) -> None:
         )
 
 
-@set_metadata()
+@set_metadata
 def label_parity(
     expected_labels: ArrayLike,
     observed_labels: ArrayLike,
@@ -204,7 +204,7 @@ def label_parity(
     return ParityOutput(cs, p, None)
 
 
-@set_metadata()
+@set_metadata
 def parity(metadata: MetadataOutput) -> ParityOutput[NDArray[np.float64]]:
     """
     Calculate chi-square statistics to assess the linear relationship between multiple factors

dataeval/metrics/estimators/ber.py

@@ -20,12 +20,12 @@ from scipy.sparse import coo_matrix
 from scipy.stats import mode
 
 from dataeval.interop import as_numpy
-from dataeval.output import OutputMetadata, set_metadata
+from dataeval.output import Output, set_metadata
 from dataeval.utils.shared import compute_neighbors, get_classes_counts, get_method, minimum_spanning_tree
 
 
 @dataclass(frozen=True)
-class BEROutput(OutputMetadata):
+class BEROutput(Output):
     """
     Output class for :func:`ber` estimator metric
 
@@ -114,7 +114,7 @@ def knn_lowerbound(value: float, classes: int, k: int) -> float:
     return ((classes - 1) / classes) * (1 - np.sqrt(max(0, 1 - ((classes / (classes - 1)) * value))))
 
 
-@set_metadata()
+@set_metadata
 def ber(images: ArrayLike, labels: ArrayLike, k: int = 1, method: Literal["KNN", "MST"] = "KNN") -> BEROutput:
     """
     An estimator for Multi-class :term:`Bayes error rate<Bayes Error Rate (BER)>` using FR or KNN test statistic basis

dataeval/metrics/estimators/divergence.py

@@ -14,12 +14,12 @@ import numpy as np
 from numpy.typing import ArrayLike, NDArray
 
 from dataeval.interop import as_numpy
-from dataeval.output import OutputMetadata, set_metadata
+from dataeval.output import Output, set_metadata
 from dataeval.utils.shared import compute_neighbors, get_method, minimum_spanning_tree
 
 
 @dataclass(frozen=True)
-class DivergenceOutput(OutputMetadata):
+class DivergenceOutput(Output):
     """
     Output class for :func:`divergence` estimator metric
 
@@ -78,7 +78,7 @@ def divergence_fnn(data: NDArray[np.float64], labels: NDArray[np.int_]) -> int:
     return errors
 
 
-@set_metadata()
+@set_metadata
 def divergence(data_a: ArrayLike, data_b: ArrayLike, method: Literal["FNN", "MST"] = "FNN") -> DivergenceOutput:
     """
     Calculates the :term`divergence` and any errors between the datasets

dataeval/metrics/estimators/uap.py

@@ -14,11 +14,11 @@ from numpy.typing import ArrayLike
 from sklearn.metrics import average_precision_score
 
 from dataeval.interop import as_numpy
-from dataeval.output import OutputMetadata, set_metadata
+from dataeval.output import Output, set_metadata
 
 
 @dataclass(frozen=True)
-class UAPOutput(OutputMetadata):
+class UAPOutput(Output):
     """
     Output class for :func:`uap` estimator metric
 
@@ -31,7 +31,7 @@ class UAPOutput(OutputMetadata):
     uap: float
 
 
-@set_metadata()
+@set_metadata
 def uap(labels: ArrayLike, scores: ArrayLike) -> UAPOutput:
     """
     FR Test Statistic based estimate of the empirical mean precision for

dataeval/metrics/stats/base.py

@@ -15,7 +15,7 @@ import tqdm
 from numpy.typing import ArrayLike, NDArray
 
 from dataeval.interop import to_numpy_iter
-from dataeval.output import OutputMetadata
+from dataeval.output import Output
 from dataeval.utils.image import normalize_image_shape, rescale
 
 DTYPE_REGEX = re.compile(r"NDArray\[np\.(.*?)\]")
@@ -65,7 +65,7 @@ class SourceIndex(NamedTuple):
 
 
 @dataclass(frozen=True)
-class BaseStatsOutput(OutputMetadata):
+class BaseStatsOutput(Output):
     """
     Attributes
     ----------

dataeval/metrics/stats/boxratiostats.py

@@ -96,7 +96,7 @@ def calculate_ratios(key: str, box_stats: BaseStatsOutput, img_stats: BaseStatsO
     return out_stats
 
 
-@set_metadata()
+@set_metadata
 def boxratiostats(
     boxstats: TStatOutput,
     imgstats: TStatOutput,