dataeval 1.1.4__tar.gz → 1.2.0rc0__tar.gz

{dataeval-1.1.4 → dataeval-1.2.0rc0}/.gitignore

@@ -12,6 +12,10 @@ docs/build/
 docs/source/reference/autoapi
 docs/source/tutorials/notebooks/checkpoints/
 
+# Intersphinx inventories pre-fetched by docs/fetch-intersphinx-inventories.py
+# (cached in CI, keyed on docs/source/conf.py)
+docs/source/_intersphinx/
+
 # Jupyter cache is stored in orphan branches (docs-artifacts/<branch>)
 # See scripts/README.md for details
 .jupyter_cache/

{dataeval-1.1.4 → dataeval-1.2.0rc0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.5
 Name: dataeval
-Version: 1.1.4
+Version: 1.2.0rc0
 Summary: DataEval provides a simple interface to characterize image data and its impact on model performance across classification and object-detection tasks
 Project-URL: Homepage, https://dataeval.ai/
 Project-URL: Repository, https://github.com/aria-ml/dataeval/
@@ -13,13 +13,12 @@ Classifier: Development Status :: 5 - Production/Stable
 Classifier: Intended Audience :: Science/Research
 Classifier: Operating System :: OS Independent
 Classifier: Programming Language :: Python :: 3 :: Only
-Classifier: Programming Language :: Python :: 3.10
 Classifier: Programming Language :: Python :: 3.11
 Classifier: Programming Language :: Python :: 3.12
 Classifier: Programming Language :: Python :: 3.13
 Classifier: Programming Language :: Python :: 3.14
 Classifier: Topic :: Scientific/Engineering
-Requires-Python: >=3.10
+Requires-Python: >=3.11
 Requires-Dist: lightgbm>=4
 Requires-Dist: maite>=0.9.4
 Requires-Dist: numba>=0.61.0
@@ -48,13 +47,11 @@ Requires-Dist: onnx>=1.14.0; extra == 'onnx'
 Requires-Dist: onnxruntime>=1.17; extra == 'onnx'
 Provides-Extra: onnx-cu126
 Requires-Dist: onnx>=1.14.0; extra == 'onnx-cu126'
-Requires-Dist: onnxruntime-gpu<1.24,>=1.17; (python_version == '3.10' and extra != 'onnx-cu130') and extra == 'onnx-cu126'
-Requires-Dist: onnxruntime-gpu<1.27,>=1.17; (python_version >= '3.11' and extra != 'onnx-cu130') and extra == 'onnx-cu126'
+Requires-Dist: onnxruntime-gpu<1.27,>=1.17; (extra != 'onnx-cu130') and extra == 'onnx-cu126'
 Requires-Dist: onnxruntime-gpu<1.27,>=1.24; (python_version >= '3.14' and extra != 'onnx-cu130') and extra == 'onnx-cu126'
 Provides-Extra: onnx-cu130
 Requires-Dist: onnx>=1.14.0; extra == 'onnx-cu130'
-Requires-Dist: onnxruntime-gpu>=1.27; (python_version >= '3.11' and extra != 'onnx-cu126') and extra == 'onnx-cu130'
-Requires-Dist: onnxruntime>=1.17; (python_version == '3.10' and extra != 'onnx-cu126') and extra == 'onnx-cu130'
+Requires-Dist: onnxruntime-gpu>=1.27; (extra != 'onnx-cu126') and extra == 'onnx-cu130'
 Provides-Extra: ontology
 Requires-Dist: rdflib>=7.0; extra == 'ontology'
 Provides-Extra: opencv
@@ -94,7 +91,7 @@ DataEval addresses the critical need underlying every AI model -- the data.
 The difference between a great dataset and a poor dataset can have drastic
 consequences on AI model performance. Data collected in the wild is noisy,
 often imbalanced, and doesn't always cover the entire spectrum of conditions
-need for deployment. DataEval provides AI practitioners with a library of
+needed for deployment. DataEval provides AI practitioners with a library of
 rigorous, algorithm-backed metrics for performance estimation, bias analysis,
 dataset cleaning and assessment, and data distribution shifts. Throughout
 all stages of the machine learning lifecycle -- from initial data collection
@@ -107,22 +104,22 @@ communities.
 
 <!-- end needs -->
 
-### Target Audience
+### Target audience
 
 <!-- start JATIC interop -->
 
 DataEval is intended to help data scientists, developers, and T&E engineers
 who want to evaluate and enhance their datasets for optimum performance. For
-users of the JATI product suite, DataEval has native interoperability when
+users of the JATIC product suite, DataEval has native interoperability when
 using MAITE-compliant datasets and models.
 
 <!-- end JATIC interop -->
 
 ---
 
-## Getting Started
+## Getting started
 
-**Python versions:** 3.10 - 3.14
+**Python versions:** 3.11 - 3.14
 
 Choose your preferred method of installation below or follow our
 [installation guide](docs/source/getting-started/installation.md).
@@ -132,7 +129,7 @@ Choose your preferred method of installation below or follow our
 - [Installing with conda/mamba](#installing-with-conda)
 - [Installing from GitHub](#installing-from-github)
 
-### **Installing with pip**
+### Installing with pip
 
 You can install DataEval directly from pypi.org using the following command.
 
@@ -170,7 +167,7 @@ pip install dataeval
 > `--index-url` under pip, `--torch-backend` under `uv pip`, and use the extras
 > only for source installs.
 
-### **torchvision (optional)**
+### torchvision (optional)
 
 `torchvision` is not a DataEval dependency, and nothing imports it until you reach
 for `TorchvisionTransform` — the escape hatch for running a torchvision v2
@@ -189,13 +186,13 @@ pip install torchvision --index-url https://download.pytorch.org/whl/cu130
 > the same index — which is also why `pip install dataeval[cpu]` is the wrong way
 > to obtain it.
 
-### **Installing with uv**
+### Installing with uv
 
 ```bash
 uv pip install dataeval --torch-backend cpu   # or cu126 / cu130 / auto
 ```
 
-### **Installing with conda**
+### Installing with conda
 
 DataEval can be installed from conda-forge:
 
@@ -212,7 +209,7 @@ variant.
 micromamba create -f environment.yml
 ```
 
-### **Installing from GitHub**
+### Installing from GitHub
 
 To install DataEval from source locally on Ubuntu, pull the source down and
 change to the DataEval project directory.
@@ -227,21 +224,7 @@ cd dataeval
 > linting, type checking and docs tooling included. See
 > [Development Setup](./CONTRIBUTING.md#development-setup) for the options it takes.
 
-#### **Using Poetry**
-
-Install DataEval.
-
-```bash
-poetry install
-```
-
-Enable Poetry's virtual environment.
-
-```bash
-poetry env activate
-```
-
-#### **Using uv**
+#### Using uv
 
 Install DataEval with dependencies for development.
 
@@ -262,7 +245,7 @@ DataEval has two input paths depending on which part of the library you are usin
 **`dataeval.core`** provides stateless functions that operate directly on NumPy
 arrays — embeddings, labels, image hashes, and statistics. No dataset object is
 required. Call these functions with arrays and get results back directly. Examples
-include `compute_stats`, `label_errors`, `divergence_mst`, and `ber_knn`.
+include `image_stats`, `label_errors`, `divergence_mst`, and `ber_knn`.
 
 **`dataeval.quality`, `dataeval.bias`, `dataeval.shift`, and `dataeval.performance`**
 provide stateful evaluator classes (`Duplicates`, `Outliers`,
@@ -364,7 +347,7 @@ class MyObjectDetectionDataset(od.Dataset):
 
         self.metadata = mp.DatasetMetadata(
             id="my_object_detection_dataset",
-            index2label={i: f"class_{i}" for i in np.unique(labels)},  # example mapping
+            index2label={i: f"class_{i}" for i in sorted({lab for labs in labels for lab in labs})},  # example mapping
         )
 
     def __len__(self) -> int:
@@ -447,19 +430,19 @@ from dataeval.quality import Duplicates
 extractor = TorchExtractor(Flatten())
 
 # Find near-duplicates using only embedding-based clustering.
-# An aggressive cluster_threshold of 1.5 should produce detections
+# An aggressive cluster_sensitivity of 1.5 should produce detections
 # of near duplicates even with a simple Flatten extractor.
 evaluator = Duplicates(
     flags=ImageStats.NONE,
     cluster_algorithm="hdbscan",
-    cluster_threshold=1.5,
+    cluster_sensitivity=1.5,
     extractor=extractor,
     batch_size=64,
 )
 result = evaluator.evaluate(dataset)
 
-# Near duplicates are grouped into sets of indices that are within
-# the specified cluster_threshold in embedding space.
+# Near duplicates are grouped into sets of indices whose pairwise distances
+# fall below `cluster_sensitivity * cluster_std` in embedding space.
 print(result)
 ```
 
@@ -488,22 +471,22 @@ guide to find the right evaluator for your situation.
 
 Know which tool to use, then check out [What data does each tool need?](./docs/source/getting-started/input-requirements.md)
 for a quick-reference table of every algorithm's inputs, and the
-[Functional Overview](./docs/source/reference/FunctionalOverview.md) for task applicability.
+[Functional overview](./docs/source/reference/FunctionalOverview.md) for task applicability.
 
 Want to just explore the documentation? The [Where to go next](./docs/source/getting-started/where-to-go-next.md)
 page allows you to jump around between the different areas of the documentation with small summaries of what each page covers.
 
 ---
 
-## Contact Us
+## Contact us
 
 If you have any questions, feel free to reach out to [us](mailto:dataeval@ariacoustics.com)!
 
-## Acknowledgement
+## Acknowledgment
 
-### CDAO Funding Acknowledgement
+### CDAO funding acknowledgment
 
-<!-- start acknowledgement -->
+<!-- start acknowledgment -->
 
 This material is based upon work supported by the Chief Digital and Artificial
 Intelligence Office under Contract No. W519TC-23-9-2033. The views and
@@ -511,4 +494,4 @@ conclusions contained herein are those of the author(s) and should not be
 interpreted as necessarily representing the official policies or endorsements,
 either expressed or implied, of the U.S. Government.
 
-<!-- end acknowledgement -->
+<!-- end acknowledgment -->

{dataeval-1.1.4 → dataeval-1.2.0rc0}/README.md

@@ -31,7 +31,7 @@ DataEval addresses the critical need underlying every AI model -- the data.
 The difference between a great dataset and a poor dataset can have drastic
 consequences on AI model performance. Data collected in the wild is noisy,
 often imbalanced, and doesn't always cover the entire spectrum of conditions
-need for deployment. DataEval provides AI practitioners with a library of
+needed for deployment. DataEval provides AI practitioners with a library of
 rigorous, algorithm-backed metrics for performance estimation, bias analysis,
 dataset cleaning and assessment, and data distribution shifts. Throughout
 all stages of the machine learning lifecycle -- from initial data collection
@@ -44,22 +44,22 @@ communities.
 
 <!-- end needs -->
 
-### Target Audience
+### Target audience
 
 <!-- start JATIC interop -->
 
 DataEval is intended to help data scientists, developers, and T&E engineers
 who want to evaluate and enhance their datasets for optimum performance. For
-users of the JATI product suite, DataEval has native interoperability when
+users of the JATIC product suite, DataEval has native interoperability when
 using MAITE-compliant datasets and models.
 
 <!-- end JATIC interop -->
 
 ---
 
-## Getting Started
+## Getting started
 
-**Python versions:** 3.10 - 3.14
+**Python versions:** 3.11 - 3.14
 
 Choose your preferred method of installation below or follow our
 [installation guide](docs/source/getting-started/installation.md).
@@ -69,7 +69,7 @@ Choose your preferred method of installation below or follow our
 - [Installing with conda/mamba](#installing-with-conda)
 - [Installing from GitHub](#installing-from-github)
 
-### **Installing with pip**
+### Installing with pip
 
 You can install DataEval directly from pypi.org using the following command.
 
@@ -107,7 +107,7 @@ pip install dataeval
 > `--index-url` under pip, `--torch-backend` under `uv pip`, and use the extras
 > only for source installs.
 
-### **torchvision (optional)**
+### torchvision (optional)
 
 `torchvision` is not a DataEval dependency, and nothing imports it until you reach
 for `TorchvisionTransform` — the escape hatch for running a torchvision v2
@@ -126,13 +126,13 @@ pip install torchvision --index-url https://download.pytorch.org/whl/cu130
 > the same index — which is also why `pip install dataeval[cpu]` is the wrong way
 > to obtain it.
 
-### **Installing with uv**
+### Installing with uv
 
 ```bash
 uv pip install dataeval --torch-backend cpu   # or cu126 / cu130 / auto
 ```
 
-### **Installing with conda**
+### Installing with conda
 
 DataEval can be installed from conda-forge:
 
@@ -149,7 +149,7 @@ variant.
 micromamba create -f environment.yml
 ```
 
-### **Installing from GitHub**
+### Installing from GitHub
 
 To install DataEval from source locally on Ubuntu, pull the source down and
 change to the DataEval project directory.
@@ -164,21 +164,7 @@ cd dataeval
 > linting, type checking and docs tooling included. See
 > [Development Setup](./CONTRIBUTING.md#development-setup) for the options it takes.
 
-#### **Using Poetry**
-
-Install DataEval.
-
-```bash
-poetry install
-```
-
-Enable Poetry's virtual environment.
-
-```bash
-poetry env activate
-```
-
-#### **Using uv**
+#### Using uv
 
 Install DataEval with dependencies for development.
 
@@ -199,7 +185,7 @@ DataEval has two input paths depending on which part of the library you are usin
 **`dataeval.core`** provides stateless functions that operate directly on NumPy
 arrays — embeddings, labels, image hashes, and statistics. No dataset object is
 required. Call these functions with arrays and get results back directly. Examples
-include `compute_stats`, `label_errors`, `divergence_mst`, and `ber_knn`.
+include `image_stats`, `label_errors`, `divergence_mst`, and `ber_knn`.
 
 **`dataeval.quality`, `dataeval.bias`, `dataeval.shift`, and `dataeval.performance`**
 provide stateful evaluator classes (`Duplicates`, `Outliers`,
@@ -301,7 +287,7 @@ class MyObjectDetectionDataset(od.Dataset):
 
         self.metadata = mp.DatasetMetadata(
             id="my_object_detection_dataset",
-            index2label={i: f"class_{i}" for i in np.unique(labels)},  # example mapping
+            index2label={i: f"class_{i}" for i in sorted({lab for labs in labels for lab in labs})},  # example mapping
         )
 
     def __len__(self) -> int:
@@ -384,19 +370,19 @@ from dataeval.quality import Duplicates
 extractor = TorchExtractor(Flatten())
 
 # Find near-duplicates using only embedding-based clustering.
-# An aggressive cluster_threshold of 1.5 should produce detections
+# An aggressive cluster_sensitivity of 1.5 should produce detections
 # of near duplicates even with a simple Flatten extractor.
 evaluator = Duplicates(
     flags=ImageStats.NONE,
     cluster_algorithm="hdbscan",
-    cluster_threshold=1.5,
+    cluster_sensitivity=1.5,
     extractor=extractor,
     batch_size=64,
 )
 result = evaluator.evaluate(dataset)
 
-# Near duplicates are grouped into sets of indices that are within
-# the specified cluster_threshold in embedding space.
+# Near duplicates are grouped into sets of indices whose pairwise distances
+# fall below `cluster_sensitivity * cluster_std` in embedding space.
 print(result)
 ```
 
@@ -425,22 +411,22 @@ guide to find the right evaluator for your situation.
 
 Know which tool to use, then check out [What data does each tool need?](./docs/source/getting-started/input-requirements.md)
 for a quick-reference table of every algorithm's inputs, and the
-[Functional Overview](./docs/source/reference/FunctionalOverview.md) for task applicability.
+[Functional overview](./docs/source/reference/FunctionalOverview.md) for task applicability.
 
 Want to just explore the documentation? The [Where to go next](./docs/source/getting-started/where-to-go-next.md)
 page allows you to jump around between the different areas of the documentation with small summaries of what each page covers.
 
 ---
 
-## Contact Us
+## Contact us
 
 If you have any questions, feel free to reach out to [us](mailto:dataeval@ariacoustics.com)!
 
-## Acknowledgement
+## Acknowledgment
 
-### CDAO Funding Acknowledgement
+### CDAO funding acknowledgment
 
-<!-- start acknowledgement -->
+<!-- start acknowledgment -->
 
 This material is based upon work supported by the Chief Digital and Artificial
 Intelligence Office under Contract No. W519TC-23-9-2033. The views and
@@ -448,4 +434,4 @@ conclusions contained herein are those of the author(s) and should not be
 interpreted as necessarily representing the official policies or endorsements,
 either expressed or implied, of the U.S. Government.
 
-<!-- end acknowledgement -->
+<!-- end acknowledgment -->

{dataeval-1.1.4 → dataeval-1.2.0rc0}/pyproject.toml

@@ -12,7 +12,7 @@ authors = [
   { name = "Ryan Wood", email = "ryan.wood@ariacoustics.com" },
   { name = "Scott Swan", email = "scott.swan@ariacoustics.com" },
 ]
-requires-python = ">=3.10"
+requires-python = ">=3.11"
 readme = "README.md"
 license = "MIT"
 maintainers = [
@@ -23,7 +23,6 @@ classifiers = [
   "Operating System :: OS Independent",
   "Intended Audience :: Science/Research",
   "Programming Language :: Python :: 3 :: Only",
-  "Programming Language :: Python :: 3.10",
   "Programming Language :: Python :: 3.11",
   "Programming Language :: Python :: 3.12",
   "Programming Language :: Python :: 3.13",
@@ -53,18 +52,19 @@ litert = ["ai-edge-litert>=2.0; python_version <= '3.14'"]
 opencv = ["opencv-python-headless>=4.8.0"]
 onnx = [
   "onnx>=1.14.0",
+  # "onnxmltools>=1.16", # required for LightGBM ONNX export
   "onnxruntime>=1.17",
 ]
 onnx-cu126 = [
   "onnx>=1.14.0",
-  "onnxruntime-gpu>=1.17,<1.24; python_version == '3.10' and extra != 'onnx-cu130'",
-  "onnxruntime-gpu>=1.17,<1.27; python_version >= '3.11' and extra != 'onnx-cu130'",  # 1.27+ requires CUDA 13.0
+  # "onnxmltools>=1.16", # required for LightGBM ONNX export
+  "onnxruntime-gpu>=1.17,<1.27; extra != 'onnx-cu130'",  # 1.27+ requires CUDA 13.0
   "onnxruntime-gpu>=1.24,<1.27; python_version >= '3.14' and extra != 'onnx-cu130'",  # 1.24+ for cp314 wheels
 ]
 onnx-cu130 = [
   "onnx>=1.14.0",
-  "onnxruntime>=1.17; python_version == '3.10' and extra != 'onnx-cu126'",  # CUDA 13.0 is not supported on Python 3.10
-  "onnxruntime-gpu>=1.27; python_version >= '3.11' and extra != 'onnx-cu126'",
+  # "onnxmltools>=1.16", # required for LightGBM ONNX export
+  "onnxruntime-gpu>=1.27; extra != 'onnx-cu126'",
 ]
 ontology = ["rdflib>=7.0"]
 
@@ -97,7 +97,6 @@ base = [
 lock = [
   { include-group = "base" },
   "pyproject2conda>=0.22",
-  "poetry==2.2.0; python_version<'3.14'",
 ]
 lint = [
   "ruff>=0.15",
@@ -109,34 +108,34 @@ docsync = [
 ]
 test = [
   "coverage[toml]>=7.6",
-  "filelock>=3.20.3",
-  "pytest>=8.3",
+  "pytest>=7.3.1",  # floor capped by the JATIC program standard; uv.lock resolves current
   "pytest-cov>=6.1",
   "pytest-xdist>=3.6.1",
 ]
 test-onnx = [
   { include-group = "test" },
+  "onnxmltools>=1.16",
   "onnxscript>=0.6.0",
 ]
 verify = [
-  "pytest>=8.3",
+  "pytest>=7.3.1",  # floor capped by the JATIC program standard; uv.lock resolves current
   "pyyaml>=6.0",
 ]
 type = [
   "pyright[nodejs]>=1.1.400",
-  "types-requests",
-  "types-pyyaml",
+  "requests>=2.31",
+  "pyyaml>=6.0",
 ]
 docs = [
-  "numpy>=2.0.2",
+  "numpy>=1.24.2",  # floor capped by the JATIC program standard; uv.lock resolves current
   "certifi>=2024.07.04",
-  "dataeval-plots[plotly]>=0.0.6",
+  "dataeval-plots[plotly]>=0.0.15",
   "ipykernel>=6.26.0",
   "ipywidgets>=8.1.1",
   "jinja2>=3.1.6",
   "jupyter-client>=8.6.0",
   "jupyter-cache>=1.0",
-  "maite-datasets>=0.0.18",
+  "maite-datasets[hf]>=0.0.21",
   "myst-nb>=1.0",
   "opencv-python-headless>=4.8.0",
   "pandas>=2.0.0",
@@ -154,22 +153,6 @@ docs = [
   "jupytext>=1.19.1",
   "ultralytics>=8.0.0",
 ]
-security = [  # keep in sync with [tool.uv.constraint-dependencies]
-  "cryptography>=46.0.5",    # CVE-2026-26007: Missing Subgroup Validation for SECT Curves
-  "filelock>=3.20.3",        # GHSA-w853-jp5j-5j7f, GHSA-qmgc-5h2g-mvrw
-  "pillow>=12.1.1",          # CVE-2026-25990: OOB write via PSD image
-  "protobuf>=6.33.5",        # GHSA-7gcm-g887-7qv7
-  "setuptools>=82.0.0",      # CVE-2026-23949: (jaraco_context) path traversal in tarball()
-                             # CVE-2026-24049: (wheel) privilege escalation via unpack
-  # FIPS: opencv builds newer than 4.12.0.88 are not approved for FIPS-enabled
-  # hosts. All three distributions below ship the same `cv2` module and are
-  # interchangeable to a resolver, so capping only one leaves the others free to
-  # supply a newer cv2 (ultralytics pulls opencv-python). Cap all three.
-  # Do not raise this ceiling without a compliance review.
-  "opencv-python-headless<=4.12.0.88",
-  "opencv-python<=4.12.0.88",
-  "opencv-contrib-python<=4.12.0.88",
-]
 dev = [
   { include-group = "base" },
   { include-group = "lint" },
@@ -179,7 +162,7 @@ dev = [
   "nox>=2025.5.1",
   "nox-uv>=0.6.2",
   "pip>=25.1",
-  "jupyter",
+  "jupyter>=1.0",
 ]
 
 [tool.uv]
@@ -197,8 +180,12 @@ conflicts = [
 ]
 constraint-dependencies = [
     "cryptography>=46.0.5",    # CVE-2026-26007: Missing Subgroup Validation for SECT Curves
-    "filelock>=3.20.3",        # GHSA-w853-jp5j-5j7f, GHSA-qmgc-5h2g-mvrw
-    "pillow>=12.1.1",          # CVE-2026-25990: OOB write via PSD image
+    "filelock>=4.2.1",         # CVE-2026-112940: (GHSA-v4x8-3wr5-3wq8) ReadWriteLock lets SQLite reopen files by name
+                               # GHSA-w853-jp5j-5j7f, GHSA-qmgc-5h2g-mvrw (fixed earlier, in 3.20.3)
+    "pillow>=13.0.0",          # Pillow 13.0.0 security release (decoder/encoder OOB reads and writes, DoS):
+                               # CVE-2026-112395, CVE-2026-112397, CVE-2026-112400,
+                               # CVE-2026-112403, CVE-2026-112404, CVE-2026-112895,
+                               # CVE-2026-112896, CVE-2026-112897, CVE-2026-112899
     "protobuf>=6.33.5",        # GHSA-7gcm-g887-7qv7
     "setuptools>=82.0.0",      # CVE-2026-23949: (jaraco_context) path traversal in tarball()
                                # CVE-2026-24049: (wheel) privilege escalation via unpack
@@ -242,17 +229,12 @@ torchvision = [
 [tool.uv.extra-build-dependencies]
 numba = ["tbb>=2021.6"]
 
-[tool.poetry]
-version = "0.0.0"  # overwritten by poetry-dynamic-versioning
-
-[[tool.poetry.source]]
-name = "pytorch-cpu"
-url = "https://download.pytorch.org/whl/cpu"
-priority = "supplemental"
-
-[tool.poetry.dependencies]
-torch = { version = ">=2.2.0", source = "pytorch-cpu" }
-torchvision = { version = ">=0.17.0", source = "pytorch-cpu" }
+[tool.hatch.build]
+# CI keeps the uv cache in the project directory. `uv sync` extracts packages into it while hatchling walks the project to build
+# dataeval, and a directory that disappears mid-walk fails the build with FileNotFoundError. Without skip-excluded-dirs,
+# hatchling filters excluded paths only after it has walked into them.
+skip-excluded-dirs = true
+exclude = [".uv-cache"]
 
 [tool.hatch.build.targets.sdist]
 include = ["src/dataeval"]
@@ -269,12 +251,6 @@ source = "vcs"
 [tool.hatch.build.hooks.vcs]
 version-file = "src/dataeval/_version.py"
 
-[tool.poetry-dynamic-versioning]
-enable = true
-vcs = "git"
-style = "pep440"
-pattern = "^v?(?P<base>\\d+\\.\\d+\\.\\d+)"
-
 [tool.pyproject2conda]
 # maite is only published on conda-forge; the defaults channel cannot resolve it.
 channels = ["conda-forge"]
@@ -301,8 +277,7 @@ reportMissingImports = false
 [tool.pytest.ini_options]
 testpaths = ["tests"]
 filterwarnings = [
-  "ignore:The default value of normalize_pixel_values changed:FutureWarning",
-  "ignore:Clustering metrics expect discrete values but received continuous values:UserWarning",
+  "ignore:The default value of hash_radius will change:FutureWarning",
 ]
 addopts = [
   "--pythonwarnings=ignore::DeprecationWarning",
@@ -314,6 +289,9 @@ markers = [
   "required: marks tests for required features",
   "optional: marks tests for optional features",
   "cuda: marks tests that require cuda",
+  "ffmpeg: marks integration tests that run against ffmpeg-generated video fixtures",
+  "performance: marks verification tests that time workloads or measure memory",
+  "infrastructure: marks tests of the documentation and repository tooling, which need the source checkout and git",
 ]
 
 [tool.coverage.run]
@@ -358,7 +336,7 @@ exclude = [
 ]
 line-length = 120
 indent-width = 4
-target-version = "py310"
+target-version = "py311"
 extend-include = ["*.ipynb"]
 
 [tool.ruff.lint]
@@ -398,8 +376,11 @@ docstring-code-format = true
 docstring-code-line-length = "dynamic"
 
 [tool.typos.default]
+locale = "en-us"
 extend-ignore-re = [
-  "`[0-9a-fA-F]{7,40}`"
+  "`[0-9a-fA-F]{7,40}`",
+  "get_cluster_labelling_at_cut",  # vendored fast_hdbscan API name
+  "MISPmicrosectime",  # MISB ST 0604 timestamp identifier
 ]
 
 [tool.typos.default.extend-words]
@@ -409,17 +390,20 @@ iy = "iy"
 fpr = "fpr"
 Pont = "Pont"
 setp = "setp"
+writeable = "writeable"
+aeroplane = "aeroplane"  # Pascal VOC class label
 
 [tool.typos.files]
 extend-exclude = [
   "requirements.*.txt",
   "uv.lock",
-  "poetry.lock"
+  "*.svg",  # generated figures: path data and random element ids
 ]
 
 [tool.codespell]
-skip = 'pyproject.toml,./*env*,./output,./docs/build,./docs/source/.jupyter_cache,./docs/source/*/data,CHANGELOG.md,uv.lock,requirements.*.txt,*.html,*.lock,*.ipynb,./graphify-out,cosign.pub'
-ignore-words-list = ["Hart","FPR", "MOT", "mot", "reencode"]
+skip = 'pyproject.toml,./*env*,./specs,./output,./docs/build,./docs/source/.jupyter_cache,./docs/source/*/data,CHANGELOG.md,uv.lock,requirements.*.txt,*.html,*.lock,*.ipynb,./graphify-out,cosign.pub'
+builtin = "clear,rare,en-GB_to_en-US"
+ignore-words-list = ["Hart","FPR", "MOT", "mot", "reencode", "writeable", "coo"]
 
 [build-system]
 requires = ["hatchling", "hatch-vcs"]

{dataeval-1.1.4 → dataeval-1.2.0rc0}/src/dataeval/__init__.py

@@ -43,8 +43,10 @@ def log(level: int = logging.DEBUG, handler: logging.Handler | None = None) -> N
     """
     Add a handler to the logger quickly for debugging.
 
-    Calling this more than once is idempotent: a handler equal to one already
-    attached to the logger is not added again, so log lines are not duplicated.
+    Calling this more than once is idempotent. With the default handler, a later call
+    reuses the handler an earlier call attached. A custom handler equal to one already
+    attached to the logger is not added again. Either way, log lines are not
+    duplicated, and each call sets the level.
 
     Parameters
     ----------
@@ -58,15 +60,21 @@ def log(level: int = logging.DEBUG, handler: logging.Handler | None = None) -> N
     """
     import logging
 
+    from dataeval._log import ConsoleHandler
+
     _logger = logging.getLogger(_ROOT)
     if handler is None:
-        handler = logging.StreamHandler()
+        handler = next((h for h in _logger.handlers if isinstance(h, ConsoleHandler)), None)
+    if handler is None:
+        handler = ConsoleHandler()
         handler.setFormatter(
             logging.Formatter(
                 "%(asctime)s %(levelname)-8s %(name)s.%(filename)s:%(lineno)s - %(funcName)10s() | %(message)s",
             ),
         )
-    if handler not in _logger.handlers:
+    attached = handler not in _logger.handlers
+    if attached:
         _logger.addHandler(handler)
     _logger.setLevel(level)
-    _logger.debug("Added logging handler %s to logger: %s", handler, _ROOT)
+    if attached:
+        _logger.debug("Added logging handler %s to logger: %s", handler, _ROOT)