mypy: Fix import-untyped and most misc issues (#378)

Author: Avasam

pyproject.toml

@@ -21,7 +21,7 @@ tests = [
     "scipy-stubs",
     "typing_extensions",
 
-    # Untyped libraries, used to prevent "reportMissingImports" and get inferred typing
+    # Untyped libraries and partial stubs. Used to prevent "reportMissingImports" and get inferred typing
     "joblib",
     "networkx",
     "PyOpenGL",
@@ -142,15 +142,16 @@ python_version = "3.9" # Target oldest supported Python version
 strict = true
 check_untyped_defs = true # Strict check on all defs
 show_column_numbers = true
+# Not all imports in these stubs are gonna be typed
+# Don't infer symbols from untyped packages as Any
+follow_untyped_imports = true
 warn_unused_ignores = false # Change from pandas
 # Partial stubs are acceptable
 disallow_any_generics = false
 disallow_incomplete_defs = false
 disallow_untyped_defs = false
 # Suppressing errors
 disable_error_code = [
-    # Not all imports in these stubs are gonna be typed
-    "import-untyped",
     # mypy's overload implementation differs from pyright
     # `assert-type` issues is tests mostly comme from checking overloads
     # Since this project is specific to Pylance, just ignore them
@@ -159,6 +160,18 @@ disable_error_code = [
     # as they are inherited from the implementation.
     "override",
     # TODO
-    "assignment", # 773 errors in 172 files
-    "misc", # 692 errors in 132 files
+    "assignment", # 744 errors in 155 files
 ]
+
+[[tool.mypy.overrides]]
+# These modules are to be removed soon, not worth solving many issues
+module = ["matplotlib.*", "networkx.*"]
+disable_error_code = [
+    "assignment",
+    "misc",
+]
+[[tool.mypy.overrides]]
+module = ["sympy.*", "skimage.*", "sklearn.*"]
+# TODO: Too many untyped decorators still left
+# https://github.com/python/mypy/issues/19148
+disable_error_code = ["misc"]

stubs/skimage/__init__.pyi

@@ -4,7 +4,6 @@ from ._shared.version_requirements import ensure_python_version as ensure_python
 
 __version__: str = ...
 submodules: list = ...
-__getattr__ = ...
 __lazy_dir__ = ...
 
 def __dir__(): ...

stubs/skimage/data/__init__.pyi

@@ -1,5 +1,88 @@
-from .._shared import lazy as lazy
+from ._binary_blobs import binary_blobs
+from ._fetchers import (
+    astronaut,
+    brain,
+    brick,
+    camera,
+    cat,
+    cell,
+    cells3d,
+    checkerboard,
+    chelsea,
+    clock,
+    coffee,
+    coins,
+    colorwheel,
+    data_dir,
+    download_all,
+    eagle,
+    file_hash,
+    grass,
+    gravel,
+    horse,
+    hubble_deep_field,
+    human_mitosis,
+    immunohistochemistry,
+    kidney,
+    lbp_frontal_face_cascade_filename,
+    lfw_subset,
+    lily,
+    logo,
+    microaneurysms,
+    moon,
+    nickel_solidification,
+    page,
+    palisades_of_vogt as palisades_of_vogt,
+    protein_transport,
+    retina,
+    rocket,
+    shepp_logan_phantom,
+    skin,
+    stereo_motorcycle,
+    text,
+    vortex,
+)
 
-__getattr__ = ...
-__dir__ = ...
-__all__ = ...  # pyright: ignore[reportUnsupportedDunderAll] # TODO
+__all__ = [
+    "astronaut",
+    "binary_blobs",
+    "brain",
+    "brick",
+    "camera",
+    "cat",
+    "cell",
+    "cells3d",
+    "checkerboard",
+    "chelsea",
+    "clock",
+    "coffee",
+    "coins",
+    "colorwheel",
+    "data_dir",
+    "download_all",
+    "eagle",
+    "file_hash",
+    "grass",
+    "gravel",
+    "horse",
+    "hubble_deep_field",
+    "human_mitosis",
+    "immunohistochemistry",
+    "kidney",
+    "lbp_frontal_face_cascade_filename",
+    "lfw_subset",
+    "lily",
+    "logo",
+    "microaneurysms",
+    "moon",
+    "nickel_solidification",
+    "page",
+    "protein_transport",
+    "retina",
+    "rocket",
+    "shepp_logan_phantom",
+    "skin",
+    "stereo_motorcycle",
+    "text",
+    "vortex",
+]

stubs/skimage/data/_fetchers.pyi

@@ -1,5 +1,7 @@
 from os import PathLike
 
+def file_hash(fname, alg="sha256") -> str: ...
+
 legacy_data_dir = ...
 skimage_distribution_dir = ...
 
@@ -46,6 +48,7 @@ def hubble_deep_field(): ...
 def retina(): ...
 def shepp_logan_phantom(): ...
 def colorwheel(): ...
+def palisades_of_vogt(): ...
 def rocket(): ...
 def stereo_motorcycle(): ...
 def lfw_subset(): ...

stubs/sklearn/gaussian_process/kernels.pyi

@@ -23,29 +23,6 @@ class Hyperparameter(NamedTuple):
     value_type: str
     name: str
 
-    # A raw namedtuple is very memory efficient as it packs the attributes
-    # in a struct to get rid of the __dict__ of attributes in particular it
-    # does not copy the string for the keys on each instance.
-    # By deriving a namedtuple class just to introduce the __init__ method we
-    # would also reintroduce the __dict__ on the instance. By telling the
-    # Python interpreter that this subclass uses static __slots__ instead of
-    # dynamic attributes. Furthermore we don't need any additional slot in the
-    # subclass so we set __slots__ to the empty tuple.
-    __slots__ = ...
-
-    def __new__(
-        cls,
-        name: str,
-        value_type: str,
-        bounds: ndarray | str | tuple[float, int] | tuple[float, float],
-        n_elements: int = 1,
-        fixed=None,
-    ) -> Self: ...
-
-    # This is mainly a testing utility to check that two hyperparameters
-    # are equal.
-    def __eq__(self, other) -> bool: ...
-
 class Kernel(metaclass=ABCMeta):
     def get_params(self, deep: bool = True) -> dict: ...
     def set_params(self, **params) -> Self: ...

stubs/sklearn/metrics/_regression.pyi

@@ -56,10 +56,10 @@ def mean_squared_error(
     sample_weight: None | ArrayLike = None,
     multioutput: ArrayLike | Literal["raw_values", "uniform_average"] = "uniform_average",
 ) -> ndarray | Float: ...
+@overload
 @deprecated(
     "`squared` is deprecated in 1.4 and will be removed in 1.6. Use `root_mean_squared_error` instead to calculate the root mean squared error."
 )
-@overload
 def mean_squared_error(
     y_true: MatrixLike | ArrayLike,
     y_pred: MatrixLike | ArrayLike,
@@ -76,10 +76,10 @@ def mean_squared_log_error(
     sample_weight: None | ArrayLike = None,
     multioutput: ArrayLike | Literal["raw_values", "uniform_average"] = "uniform_average",
 ) -> float | ndarray: ...
+@overload
 @deprecated(
     "`squared` is deprecated in 1.4 and will be removed in 1.6. Use `root_mean_squared_log_error` instead to calculate the root mean squared logarithmic error."
 )
-@overload
 def mean_squared_log_error(
     y_true: MatrixLike | ArrayLike,
     y_pred: MatrixLike | ArrayLike,

stubs/sklearn/preprocessing/_data.pyi

@@ -112,12 +112,10 @@ class StandardScaler(OneToOneFeatureMixin, TransformerMixin, BaseEstimator):
     def transform(self, X: spmatrix, copy: None | bool = None) -> spmatrix: ...
     @overload
     def transform(self, X: ArrayLike, copy: None | bool = None) -> ndarray: ...
-    def transform(self, X: MatrixLike, copy: None | bool = None) -> ndarray | spmatrix: ...
     @overload
     def inverse_transform(self, X: spmatrix, copy: None | bool = None) -> spmatrix: ...
     @overload
     def inverse_transform(self, X: ArrayLike, copy: None | bool = None) -> ndarray: ...
-    def inverse_transform(self, X: MatrixLike | ArrayLike, copy: None | bool = None) -> ndarray | spmatrix: ...
 
 class MaxAbsScaler(OneToOneFeatureMixin, TransformerMixin, BaseEstimator):
     n_samples_seen_: int = ...
@@ -135,12 +133,10 @@ class MaxAbsScaler(OneToOneFeatureMixin, TransformerMixin, BaseEstimator):
     def transform(self, X: spmatrix) -> spmatrix: ...
     @overload
     def transform(self, X: ArrayLike) -> ndarray: ...
-    def transform(self, X: MatrixLike | ArrayLike) -> ndarray | spmatrix: ...
     @overload
     def inverse_transform(self, X: spmatrix) -> spmatrix: ...
     @overload
     def inverse_transform(self, X: ArrayLike) -> ndarray: ...
-    def inverse_transform(self, X: MatrixLike | ArrayLike) -> ndarray | spmatrix: ...
 
 def maxabs_scale(X: MatrixLike | ArrayLike, *, axis: Int = 0, copy: bool = True): ...
 
@@ -166,12 +162,10 @@ class RobustScaler(OneToOneFeatureMixin, TransformerMixin, BaseEstimator):
     def transform(self, X: spmatrix) -> spmatrix: ...
     @overload
     def transform(self, X: ArrayLike) -> ndarray: ...
-    def transform(self, X: MatrixLike | ArrayLike) -> ndarray | spmatrix: ...
     @overload
     def inverse_transform(self, X: spmatrix) -> spmatrix: ...
     @overload
     def inverse_transform(self, X: ArrayLike) -> ndarray: ...
-    def inverse_transform(self, X: MatrixLike | ArrayLike) -> ndarray | spmatrix: ...
 
 @overload
 def robust_scale(
@@ -186,7 +180,7 @@ def robust_scale(
 ) -> spmatrix: ...
 @overload
 def robust_scale(
-    X: ndarray,
+    X: ArrayLike,
     *,
     axis: Int = 0,
     with_centering: bool = True,
@@ -195,16 +189,6 @@ def robust_scale(
     copy: bool = True,
     unit_variance: bool = False,
 ) -> ndarray: ...
-def robust_scale(
-    X: MatrixLike,
-    *,
-    axis: Int = 0,
-    with_centering: bool = True,
-    with_scaling: bool = True,
-    quantile_range: tuple[float, float] = ...,
-    copy: bool = True,
-    unit_variance: bool = False,
-) -> ndarray | spmatrix: ...
 @overload
 def normalize(
     X: spmatrix,
@@ -241,14 +225,6 @@ def normalize(
     copy: bool = True,
     return_norm: Literal[False] = False,
 ) -> ndarray: ...
-def normalize(
-    X: MatrixLike | ArrayLike,
-    norm: Literal["l1", "l2", "max"] = "l2",
-    *,
-    axis: int = 1,
-    copy: bool = True,
-    return_norm: bool = False,
-) -> csr_matrix | tuple[ndarray | spmatrix, ndarray] | ndarray: ...
 
 class Normalizer(OneToOneFeatureMixin, TransformerMixin, BaseEstimator):
     feature_names_in_: ndarray = ...
@@ -262,7 +238,6 @@ class Normalizer(OneToOneFeatureMixin, TransformerMixin, BaseEstimator):
     def transform(self, X: spmatrix, copy: None | bool = None) -> spmatrix: ...
     @overload
     def transform(self, X: ArrayLike, copy: None | bool = None) -> ndarray: ...
-    def transform(self, X: MatrixLike | ArrayLike, copy: None | bool = None) -> ndarray | spmatrix: ...
 
 def binarize(X: MatrixLike | ArrayLike, *, threshold: Float = 0.0, copy: bool = True) -> ndarray | spmatrix: ...
 
@@ -278,7 +253,6 @@ class Binarizer(OneToOneFeatureMixin, TransformerMixin, BaseEstimator):
     def transform(self, X: spmatrix, copy: None | bool = None) -> spmatrix: ...
     @overload
     def transform(self, X: ArrayLike, copy: None | bool = None) -> ndarray: ...
-    def transform(self, X: MatrixLike | ArrayLike, copy: None | bool = None) -> ndarray | spmatrix: ...
 
 class KernelCenterer(ClassNamePrefixFeaturesOutMixin, TransformerMixin, BaseEstimator):
     feature_names_in_: ndarray = ...
@@ -294,7 +268,6 @@ class KernelCenterer(ClassNamePrefixFeaturesOutMixin, TransformerMixin, BaseEsti
 def add_dummy_feature(X: spmatrix, value: Float = 1.0) -> spmatrix: ...
 @overload
 def add_dummy_feature(X: ArrayLike, value: Float = 1.0) -> ndarray: ...
-def add_dummy_feature(X: MatrixLike | ArrayLike, value: Float = 1.0) -> ndarray | spmatrix: ...
 
 class QuantileTransformer(OneToOneFeatureMixin, TransformerMixin, BaseEstimator):
     feature_names_in_: ndarray = ...
@@ -324,12 +297,10 @@ class QuantileTransformer(OneToOneFeatureMixin, TransformerMixin, BaseEstimator)
     def transform(self, X: spmatrix) -> spmatrix: ...
     @overload
     def transform(self, X: ArrayLike) -> ndarray: ...
-    def transform(self, X: MatrixLike | ArrayLike) -> ndarray | spmatrix: ...
     @overload
     def inverse_transform(self, X: spmatrix) -> spmatrix: ...
     @overload
     def inverse_transform(self, X: ArrayLike) -> ndarray: ...
-    def inverse_transform(self, X: MatrixLike | ArrayLike) -> ndarray | spmatrix: ...
 
 @overload
 def quantile_transform(
@@ -355,17 +326,6 @@ def quantile_transform(
     random_state: RandomState | None | Int = None,
     copy: bool = True,
 ) -> ndarray: ...
-def quantile_transform(
-    X: MatrixLike | ArrayLike,
-    *,
-    axis: Int = 0,
-    n_quantiles: Int = 1000,
-    output_distribution: Literal["uniform", "normal"] = "uniform",
-    ignore_implicit_zeros: bool = False,
-    subsample: Int = ...,
-    random_state: RandomState | None | Int = None,
-    copy: bool = True,
-) -> ndarray | spmatrix: ...
 
 class PowerTransformer(OneToOneFeatureMixin, TransformerMixin, BaseEstimator):
     feature_names_in_: ndarray = ...

stubs/sklearn/utils/_isfinite.pyi

@@ -1,7 +1,17 @@
 from enum import IntEnum
+from typing import Final
 
 import numpy as np
 
-class FiniteStatus(IntEnum): ...
+__test__: dict
+
+class FiniteStatus(IntEnum):
+    all_finite = 0
+    has_nan = 1
+    has_infinite = 2
+
+all_finite: Final = FiniteStatus.all_finite
+has_nan: Final = FiniteStatus.has_nan
+has_infinite: Final = FiniteStatus.has_infinite
 
 def cy_isfinite(a: np.ndarray, allow_nan: bool = False) -> bool: ...

stubs/sympy-stubs/printing/defaults.pyi

@@ -0,0 +1,3 @@
+from sympy.core._print_helpers import Printable as Printable
+
+DefaultPrinting = Printable

stubs/transformers-stubs/models/auto/configuration_auto.pyi

@@ -5,7 +5,7 @@ from typing import Any, NoReturn, TypeVar
 
 from transformers.configuration_utils import PretrainedConfig
 
-_F = TypeVar("_F", bound=Callable[..., Any])
+_CallableT = TypeVar("_CallableT", bound=Callable)
 
 CONFIG_MAPPING_NAMES: OrderedDict[str, str]
 MODEL_NAMES_MAPPING: OrderedDict[str, str]
@@ -35,7 +35,9 @@ class _LazyLoadAllMappings(OrderedDict[str, str]):
     def __iter__(self) -> Iterator[str]: ...
     def __contains__(self, item: object) -> bool: ...
 
-def replace_list_option_in_docstrings(config_to_class=None, use_model_types: bool = True) -> Callable[[_F], _F]: ...
+def replace_list_option_in_docstrings(
+    config_to_class=None, use_model_types: bool = True
+) -> Callable[[_CallableT], _CallableT]: ...
 
 class AutoConfig:
     def __init__(self) -> None: ...