python311Packages.tensorflow-datasets: 4.9.7 -> 4.9.8

diff --git a/tensorflow_datasets/image_classification/corruptions.py b/tensorflow_datasets/image_classification/corruptions.py

index 066c4460..cb9a6667 100644

--- a/tensorflow_datasets/image_classification/corruptions.py

+++ b/tensorflow_datasets/image_classification/corruptions.py

@@ -35,7 +35,7 @@ FROST_FILENAMES = []

 def _imagemagick_bin():

-  return 'imagemagick'  # pylint: disable=unreachable

+  return 'convert'  # pylint: disable=unreachable

 # /////////////// Corruption Helpers ///////////////

@@ -675,7 +675,7 @@ def spatter(x, severity=1):

     #     ker = np.array([[-1,-2,-3],[-2,0,0],[-3,0,1]], dtype=np.float32)

     #     ker -= np.mean(ker)

     ker = np.array([[-2, -1, 0], [-1, 1, 1], [0, 1, 2]])

-    dist = cv2.filter2D(dist, cv2.CVX_8U, ker)

+    dist = cv2.filter2D(dist, cv2.CV_8U, ker)

     dist = cv2.blur(dist, (3, 3)).astype(np.float32)

     m = cv2.cvtColor(liquid_layer * dist, cv2.COLOR_GRAY2BGRA)

{

  apache-beam,

  lib,

  buildPythonPackage,

  fetchFromGitHub,

  fetchpatch,

  # dependencies

  array-record,

  attrs,

  dill,

  dm-tree,

  future,

  immutabledict,

  importlib-resources,

  numpy,

  promise,

  protobuf,

  psutil,

  requests,

  simple-parsing,

  six,

  tensorflow-metadata,

  termcolor,

  tqdm,

  # tests

  apache-beam,

  beautifulsoup4,

  buildPythonPackage,

  click,

  datasets,

  dill,

  dm-tree,

  fetchFromGitHub,

  ffmpeg,

  future,

  imagemagick,

  importlib-resources,

  jax,

  jaxlib,

  jinja2,

  langdetect,

  lib,

  lxml,

  matplotlib,

  mlcroissant,

  mwparserfromhell,

  mwxml,

  networkx,

  nltk,

  numpy,

  opencv4,

  pandas,

  pillow,

  promise,

  protobuf,

  psutil,

  pycocotools,

  pydub,

  pytest-xdist,

  pytestCheckHook,

  requests,

  scikit-image,

  scipy,

  six,

  sortedcontainers,

  tensorflow,

  tensorflow-metadata,

  termcolor,

  tifffile,

  tqdm,

  zarr,

}:

buildPythonPackage rec {

  pname = "tensorflow-datasets";

  version = "4.9.7";

  format = "setuptools";

  version = "4.9.8";

  pyproject = true;

  src = fetchFromGitHub {

    owner = "tensorflow";

    repo = "datasets";

    tag = "v${version}";

    hash = "sha256-pnut5z3rEYIOOhsQT6uVjPdy+SqftKptSQMwxbMXoKA=";

    hash = "sha256-nqveZ+8b0f5sGIn6WufKeA37yEsZjzhCIbCfwMZ9JOM=";

  };

  patches = [

    # addresses https://github.com/tensorflow/datasets/issues/3673

    ./corruptions.patch

    # mlmlcroissant uses encoding_formats, not encoding_formats.

    # Backport https://github.com/tensorflow/datasets/pull/11037 until released.

    (fetchpatch {

      url = "https://github.com/tensorflow/datasets/commit/92cbcff725a1036569a515cc3356aa8480740451.patch";

      hash = "sha256-2hnMvQP83+eAJllce19aHujcoWQzUz3+LsasWCo4BtM=";

    })

  ];

  propagatedBuildInputs = [

  dependencies = [

    array-record

    attrs

    dill

    dm-tree

    future

    immutabledict

    importlib-resources

    numpy

    promise

    protobuf

    psutil

    requests

    simple-parsing

    six

    tensorflow-metadata

    termcolor

    langdetect

    lxml

    matplotlib

    mlcroissant

    mwparserfromhell

    mwxml

    networkx

    pytestCheckHook

    scikit-image

    scipy

    sortedcontainers

    tensorflow

    tifffile

    zarr

  ];

  pytestFlagsArray = [

    # AttributeError: 'NoneType' object has no attribute 'Table'

    "--deselect=tensorflow_datasets/core/file_adapters_test.py::test_read_write"

    "--deselect=tensorflow_datasets/text/c4_wsrs/c4_wsrs_test.py::C4WSRSTest"

  ];

  disabledTestPaths = [

    # Sandbox violations: network access, filesystem write attempts outside of build dir, ...

    "tensorflow_datasets/core/dataset_builder_test.py"

    "tensorflow_datasets/import_without_tf_test.py"

    "tensorflow_datasets/proto/build_tf_proto_test.py"

    "tensorflow_datasets/scripts/cli/build_test.py"

    "tensorflow_datasets/datasets/imagenet2012_corrupted/imagenet2012_corrupted_dataset_builder_test.py"

    # Requires `pretty_midi` which is not packaged in `nixpkgs`.

    "tensorflow_datasets/audio/groove.py"

    # Requires `gcld3` and `pretty_midi` which are not packaged in `nixpkgs`.

    "tensorflow_datasets/core/lazy_imports_lib_test.py"

    # AttributeError: 'NoneType' object has no attribute 'Table'

    "tensorflow_datasets/core/dataset_builder_beam_test.py"

    "tensorflow_datasets/core/dataset_builders/adhoc_builder_test.py"

    "tensorflow_datasets/core/split_builder_test.py"

    "tensorflow_datasets/core/writer_test.py"

    # Requires `tensorflow_io` which is not packaged in `nixpkgs`.

    "tensorflow_datasets/core/features/audio_feature_test.py"

    "tensorflow_datasets/image/lsun_test.py"

    # Requires `envlogger` which is not packaged in `nixpkgs`.

    "tensorflow_datasets/rlds/locomotion/locomotion_test.py"

    "tensorflow_datasets/rlds/robosuite_panda_pick_place_can/robosuite_panda_pick_place_can_test.py"

    # Fails with `TypeError: Constant constructor takes either 0 or 2 positional arguments`

    # deep in TF AutoGraph. Doesn't reproduce in Docker with Ubuntu 22.04 => might be related

    # to the differences in some of the dependencies?

    "tensorflow_datasets/text/c4_utils_test.py"

  ];

  meta = with lib; {

  meta = {

    description = "Library of datasets ready to use with TensorFlow";

    homepage = "https://www.tensorflow.org/datasets/overview";

    license = licenses.asl20;

    maintainers = with maintainers; [ ndl ];

    changelog = "https://github.com/tensorflow/datasets/releases/tag/v${version}";

    license = lib.licenses.asl20;

    maintainers = with lib.maintainers; [ ndl ];

  };

}