ruff update sachen

docs/conf.py

@@ -95,7 +95,7 @@ def substitute(matchobj):
     mathjax_path = ""
 else:
     extensions.append("sphinx.ext.mathjax")
-    mathjax_path = "https://cdn.jsdelivr.net/npm/mathjax@3/es5/" "tex-chtml.js"
+    mathjax_path = "https://cdn.jsdelivr.net/npm/mathjax@3/es5/tex-chtml.js"
 
 autodoc_default_options = {"members": True, "inherited-members": True, "special_members": True}
 # autodoc_typehints = 'description'  # Does not work as expected. Maybe try at future date again
@@ -202,6 +202,7 @@ def skip_properties(app, what, name, obj, skip, options):
     """This removes all properties from the documentation as they are expected to be documented in the docstring."""
     if isinstance(obj, property):
         return True
+    return None
 
 
 def setup(app):

pyproject.toml

@@ -45,7 +45,7 @@ memory-profiler = "^0.58.0"
 matplotlib = "^3.4.3"
 toml = "^0.10.2"
 Sphinx = "^6.1.3"
-ruff = "^0.0.235"
+ruff = "^0.0.286"
 
 
 [[tool.poetry.source]]
@@ -156,8 +156,6 @@ ignore = [
     "EM101",
     "EM102",
     "EM103",
-    # Multiline docstring summary
-    "D213",
     # Varaibles before return
     "RET504",
     # Abstract raise into inner function
@@ -169,7 +167,11 @@ ignore = [
     # df as varaible name
     "PD901",
     # melt over stack
-    "PD013"
+    "PD013",
+    # Avoid specifying long messages outside the exception class
+    "TRY003",
+    # To many arguments
+    "PLR0913"
 ]
 
 

tests/test_base.py

@@ -291,7 +291,7 @@ def test_nested_mutable_algorithm_copy():
     assert (
         joblib.hash(test_instance.mutable.get_params())
         == joblib.hash(nested_instance.get_params())
-        == joblib.hash({k: f for k, f in nested_params.items()})
+        == joblib.hash(dict(nested_params.items()))
     )
 
 

tests/test_dataset.py

@@ -61,7 +61,7 @@ def _create_valid_index(input_dict=None, columns_names=None):
 
 def _create_random_bool_map(n, seed):
     np.random.seed(seed)
-    return list(map(lambda x: x >= 0.5, np.random.rand(n)))
+    return [x >= 0.5 for x in np.random.rand(n)]
 
 
 class TestDataset:
@@ -340,7 +340,7 @@ def test_getitem_error_input(self, subscript, select_lvl, what_to_expect):
         "groupby_level", (["patients"], ["patients", "tests"], ["patients", "tests", "extra with space"])
     )
     @pytest.mark.parametrize(
-        "index,is_single_level",
+        ("index", "is_single_level"),
         (
             (
                 _create_valid_index(

tests/test_parameter_string_annot.py

@@ -37,9 +37,6 @@ def __init__(self, hyper: int, normal: str, custom_annotated: int, normal_no_ann
 
 
 def test_import_forward():
-    if TYPE_CHECKING:
-        pass
-
     class Test(BaseTpcpObject):
         hyper: HyperPara[int]
         normal: Para[renamed_optimize]
@@ -85,11 +82,11 @@ def __init__(self, hyper: int, normal: optimize.GridSearch, custom_annotated: in
 def test_test_str_based_forward():
     class Test(BaseTpcpObject):
         hyper: HyperPara[int]
-        normal: Para["Dataset"]
+        normal: Para[Dataset]
         custom_annotated: Annotated[HyperPara[int], "custom_metadata"]
         normal_no_annot: int
 
-        def __init__(self, hyper: int, normal: "Dataset", custom_annotated: int, normal_no_annot: int):
+        def __init__(self, hyper: int, normal: Dataset, custom_annotated: int, normal_no_annot: int):
             self.hyper = hyper
             self.normal = normal
             self.custom_annotated = custom_annotated

tpcp/_algorithm_utils.py

@@ -11,13 +11,14 @@
 
 from typing_extensions import Concatenate, ParamSpec
 
+from tpcp import Algorithm
 from tpcp._base import NOTHING, _get_annotated_fields_of_type
 from tpcp._hash import custom_hash
 from tpcp._parameters import _ParaTypes
 from tpcp.exceptions import PotentialUserErrorWarning
 
 if TYPE_CHECKING:
-    from tpcp import Algorithm, OptimizableAlgorithm, OptimizablePipeline
+    from tpcp import OptimizableAlgorithm, OptimizablePipeline
     from tpcp._algorithm import AlgorithmT
 
     OptimizableT = TypeVar("OptimizableT", OptimizablePipeline, OptimizableAlgorithm)
@@ -140,11 +141,9 @@ def _check_safe_run(algorithm: AlgorithmT, old_method: Callable, *args: Any, **k
     before_paras = algorithm.get_params()
     before_paras_hash = custom_hash(before_paras)
     output: AlgorithmT
-    if hasattr(old_method, "__self__"):
-        # In this case the method is already bound and we do not need to pass the algo as first argument
-        output = old_method(*args, **kwargs)
-    else:
-        output = old_method(algorithm, *args, **kwargs)
+
+    # In this case the method is already bound and we do not need to pass the algo as first argument
+    output = old_method(*args, **kwargs) if hasattr(old_method, "__self__") else old_method(algorithm, *args, **kwargs)
     after_paras = algorithm.get_params()
     after_paras_hash = custom_hash(after_paras)
     if not before_paras_hash == after_paras_hash:
@@ -220,6 +219,7 @@ def safe_wrapped(self: AlgorithmT, *args: P.args, **kwargs: P.kwargs) -> Algorit
                 f"`    _action_methods = ({action_method.__name__},)`\n\n"
                 "Or append it to the tuple, if it already exists.",
                 PotentialUserErrorWarning,
+                stacklevel=2,
             )
         return _check_safe_run(self, action_method, *args, **kwargs)
 
@@ -232,11 +232,9 @@ def _get_nested_opti_paras(algorithm: Algorithm, opti_para_names: List[str]) ->
     optimizable_paras = {}
     other_paras = {}
     for p, v in paras.items():
-        if p in opti_para_names:
-            optimizable_paras[p] = v
-        # For each optimizable parameter, we also add all children, as they are also allowed to change,
-        # if the parent is allowed to.
-        elif any(p.startswith(o + "__") for o in opti_para_names):
+        if p in opti_para_names or any(p.startswith(o + "__") for o in opti_para_names):
+            # For each optimizable parameter, we also add all children, as they are also allowed to change,
+            # if the parent is allowed to.
             optimizable_paras[p] = v
         else:
             other_paras[p] = v
@@ -249,7 +247,7 @@ def _get_nested_opti_paras(algorithm: Algorithm, opti_para_names: List[str]) ->
     return optimizable_paras, other_paras
 
 
-def _check_safe_optimize(  # noqa: C901
+def _check_safe_optimize(  # noqa: C901, PLR0912
     algorithm: OptimizableT, old_method: Callable, *args: Any, **kwargs: Any
 ) -> OptimizableT:
 
@@ -349,6 +347,7 @@ def _check_safe_optimize(  # noqa: C901
             f"({optimizable_paras}). "
             "This could indicate an implementation error of the `self_optimize` method.",
             PotentialUserErrorWarning,
+            stacklevel=2,
         )
     if other_returns != (NOTHING, NOTHING):
         return optimized_algorithm, other_returns
@@ -402,6 +401,7 @@ def safe_wrapped(self: OptimizableT, *args: P.args, **kwargs: P.kwargs) -> Optim
                 "The `make_optimize_safe` decorator is only meant for the `self_optimize` method, but you applied it "
                 f"to the `{self_optimize_method.__name__}` method.",
                 PotentialUserErrorWarning,
+                stacklevel=2,
             )
         try:
             return _check_safe_optimize(self, self_optimize_method, *args, **kwargs)

tpcp/_base.py

@@ -184,7 +184,7 @@ def _retry_eval_with_missing_locals(
 def _custom_get_type_hints(cls: Type[_BaseTpcpObject]) -> Dict[str, Any]:
     """Extract type hints while avoiding issues with forward references.
 
-    We automatically skip all douple_underscore methods.
+    We automatically skip all douple-underscore methods.
     """
     hints = {}
     for base in reversed(cls.__mro__):
@@ -194,11 +194,11 @@ def _custom_get_type_hints(cls: Type[_BaseTpcpObject]) -> Dict[str, Any]:
             if name.startswith("__"):
                 continue
             if value is None:
-                value = type(None)
+                value = type(None)  # noqa: PLW2901
             elif isinstance(value, str):
                 # NOTE: This does not check if the str is a valid expression.
                 #       This might not be an issue, but could lead to obscure error messages.
-                value = _retry_eval_with_missing_locals(value, base_globals)
+                value = _retry_eval_with_missing_locals(value, base_globals)  # noqa: PLW2901
             hints[name] = value
     return hints
 
@@ -212,7 +212,7 @@ def _extract_annotations(
         origin = get_origin(v)
         if origin is ClassVar:
             # If the parameter is a ClassVar, we go one level deeper and check, if its argument was annotated.
-            v = get_args(v)[0]
+            v = get_args(v)[0]  # noqa: PLW2901
             origin = get_origin(v)
         if origin is Annotated:
             for annot in get_args(v)[1:]:
@@ -265,7 +265,8 @@ def _validate_all_parent_parameters_implemented(cls: Type[_BaseTpcpObject]):
                 f"Missing parameters: {missing_params}\n"
                 "This might not be a problem, but indicates bad design and you might run into actual issues with some "
                 "of the validation magic `tpcp` does in the background. "
-                "We would recommend to implement all parameters of your parents in a subclass."
+                "We would recommend to implement all parameters of your parents in a subclass.",
+                stacklevel=2,
             )
 
 
@@ -495,7 +496,7 @@ def _set_comp_field(instance, field_name, params):
     # We first partition our field names to know to which index they belong
     comp_params: DefaultDict[str, Any] = defaultdict(dict)
     for key, value in params.items():
-        key, delim, sub_key = key.partition("__")
+        key, delim, sub_key = key.partition("__")  # noqa: PLW2901
         if delim:
             comp_params[key][sub_key] = value
         else:
@@ -537,7 +538,7 @@ def _set_params(instance: BaseTpcpObjectObjT, **params: Any) -> BaseTpcpObjectOb
 
     nested_params: DefaultDict[str, Any] = defaultdict(dict)  # grouped by prefix
     for key, value in params.items():
-        key, delim, sub_key = key.partition("__")
+        key, delim, sub_key = key.partition("__")  # noqa: PLW2901
         if key not in valid_params:
             raise ValueError(f"`{key}` is not a valid parameter name for {type(instance).__name__}.")
 
@@ -626,6 +627,7 @@ def _annotations_are_valid(
                     "Annotating a nested parameter (parameter like `nested_object__nest_para` as a simple "
                     "Parameter has no effect and the entire line should be removed.",
                     PotentialUserErrorWarning,
+                    stacklevel=2,
                 )
         elif k not in fields:
             raise ValueError(
@@ -728,7 +730,7 @@ def clone(algorithm: T, *, safe: bool = False) -> T:
             with Path(os.devnull).open("w") as devnull, contextlib.redirect_stdout(devnull):
                 return copy.deepcopy(algorithm)
         raise TypeError(
-            f"Cannot clone object '{repr(algorithm)}' (type {type(algorithm)}): "
+            f"Cannot clone object '{algorithm!r}' (type {type(algorithm)}): "
             "it does not seem to be a compatible algorithm/pipline class or general `tpcp` object as it does not "
             "inherit from `BaseTpcpObject` or `Algorithm` or `Pipeline`."
         )

tpcp/_utils/_score.py

@@ -124,7 +124,7 @@ def _score(
     return result
 
 
-def _optimize_and_score(  # noqa: C901
+def _optimize_and_score(
     optimizer: BaseOptimize,
     scorer: Scorer,
     train_set: Dataset,
@@ -229,7 +229,7 @@ def _optimize_and_score(  # noqa: C901
         # instance of the trained pipeline.
         result["optimizer"] = optimizer
     if return_parameters:
-        result["parameters"] = {**hyperparameters, **pure_parameters} or None
+        result["parameters"] = {**hyperparameters, **pure_parameters}
     return result
 
 

tpcp/optimize/_optimize.py

@@ -97,6 +97,7 @@ def optimize(self, dataset: DatasetT, **optimize_params: Any) -> Self:  # noqa:
                 "`DummyOptimize` does never call this method and skips any optimization steps! "
                 "Use `Optimize` if you actually want to optimize your pipeline.",
                 PotentialUserErrorWarning,
+                stacklevel=2,
             )
         self.optimized_pipeline_ = self.pipeline.clone()
         return self
@@ -830,6 +831,7 @@ def _store(key_name: str, array, weights=None, splits=False, rank=False):
                 warnings.warn(
                     f"One or more of the {key_name.split('_')[0]} scores are non-finite: {array_means}",
                     category=UserWarning,
+                    stacklevel=2,
                 )
             # Weighted std is not directly available in numpy
             array_stds = np.sqrt(np.average((array - array_means[:, np.newaxis]) ** 2, axis=1, weights=weights))
@@ -914,7 +916,8 @@ def _validate_return_optimized(return_optimized, multi_metric, results) -> Tuple
             "single score."
             "`return_optimized` is set to True. "
             "The only allowed string value for `return_optimized` in a single metric case is `-score`, "
-            "to invert the metric before score selection."
+            "to invert the metric before score selection.",
+            stacklevel=2,
         )
         return reverse, "score"
     raise ValueError("`return_optimized` must be a bool or explicitly `score` or `-score` in a single metric case.")