replaced flake8 and isort with ruff (#197)

* replaced flake8 and isort with ruff

* added isort checks to ruff

* Enforce B008

* Enforce E731

* Enforce C413

* Enforce C414

* Enforce C416

* Enforce C419

* Enforce B006

* Enforce B007

---------

Co-authored-by: WalzDS <[email protected]>

.github/workflows/lint.yaml

@@ -9,7 +9,7 @@ on:
 
 jobs:    
   style:
-    name: Check style with flake8, black, and isort
+    name: Check style with ruff and black
     runs-on: ubuntu-latest
     steps:
       - uses: actions/checkout@v2
@@ -18,10 +18,9 @@ jobs:
           python-version: 3.9
       - name: style
         run: |
-          pip3 install flake8==5.0.4 black==22.10.0 isort==5.12.0
-          flake8 --verbose bofire tests
+          pip3 install ruff==0.0.265 black==22.10.0
           black --check bofire tests
-          isort . --check-only --verbose
+          ruff check .
 
   pyright:
     name: Typechecking with pyright

.pre-commit-config.yaml

@@ -7,21 +7,11 @@ repos:
         stages: [commit]
         entry: black
         types: [python]
-  - repo: https://github.com/pycqa/flake8
-    rev: 5.0.4
+  - repo: https://github.com/charliermarsh/ruff-pre-commit
+    # Ruff version.
+    rev: "v0.0.265"
     hooks:
-      - id: flake8
-        name: flake8
-        entry: flake8
-        types: [python]
-        require_serial: true
-  - repo: https://github.com/pycqa/isort
-    rev: 5.12.0
-    hooks:
-      - id: isort
-        name: isort
-        stages: [commit]
-        entry: isort
-        types: [python]
-
+      - id: ruff
+        args: [--fix, --exit-non-zero-on-fix]
+
 exclude: "README.md"

bofire/benchmarks/aspen_benchmark.py

@@ -35,7 +35,7 @@ def __init__(
         filename: str,
         domain: Domain,
         paths: Dict[str, str],
-        additional_output_keys: List = [],
+        additional_output_keys: Optional[List] = None,
         translate_into_aspen_readable: Optional[
             Callable[[Domain, pd.DataFrame], pd.DataFrame]
         ] = None,
@@ -60,7 +60,7 @@ def __init__(
 
         self.translate_into_aspen_readable = translate_into_aspen_readable
         self._domain = domain
-        self.additional_output_keys = additional_output_keys
+        self.additional_output_keys = additional_output_keys or []
 
         for key in self.domain.get_feature_keys():
             # Check, if every input and output variable has a path to Aspen provided.

bofire/benchmarks/multi.py

@@ -174,12 +174,14 @@ class SnarBenchmark(Benchmark):
     Solving of a differential equation system with varying intitial values.
     """
 
-    def __init__(self, C_i: Optional[np.ndarray] = np.ndarray((1, 1))):
+    def __init__(self, C_i: Optional[np.ndarray] = None):
         """Initializes multiobjective test function object of type SnarBenchmark.
 
         Args:
             C_i (Optional[np.ndarray]): Input concentrations. Defaults to [1, 1]
         """
+        if C_i is None:
+            C_i = np.array([1, 1])
         self.C_i = C_i
 
         # Decision variables
@@ -224,7 +226,7 @@ def _f(self, candidates: pd.DataFrame) -> pd.DataFrame:
         """
         stys = []
         e_factors = []
-        for i, candidate in candidates.iterrows():
+        for _, candidate in candidates.iterrows():
             tau = float(candidate["tau"])
             equiv_pldn = float(candidate["equiv_pldn"])
             conc_dfnb = float(candidate["conc_dfnb"])
@@ -297,11 +299,10 @@ def _integrand(self, t, C, T):
         T_ref = 90 + 273.71  # Convert to deg K
         T = T + 273.71  # Convert to deg K
         # Need to convert from 10^-2 M^-1s^-1 to M^-1min^-1
-        k = (
-            lambda k_ref, E_a, temp: 0.6
-            * k_ref
-            * np.exp(-E_a / R * (1 / temp - 1 / T_ref))
-        )
+
+        def k(k_ref, E_a, temp):
+            return 0.6 * k_ref * np.exp(-E_a / R * (1 / temp - 1 / T_ref))
+
         k_a = k(57.9, 33.3, T)
         k_b = k(2.70, 35.3, T)
         k_c = k(0.865, 38.9, T)

bofire/benchmarks/single.py

@@ -52,7 +52,7 @@ def __init__(
         lower: float = -32.768,
         upper: float = 32.768,
         best_possible_f: float = 0.0,
-        evaluated_points=[],
+        evaluated_points: Optional[list] = None,
     ):
         """Initializes benchmark function of type Ackley.
 
@@ -74,6 +74,8 @@ def __init__(
         self.lower = lower
         self.upper = upper
         self.best_possible_f = best_possible_f
+        if evaluated_points is None:
+            evaluated_points = []
         self.evaluated_points = evaluated_points
 
         input_feature_list = []

bofire/data_models/domain/domain.py

@@ -305,7 +305,7 @@ def get_nchoosek_combinations(self, exhaustive: bool = False):  # noqa: C901
                     used_features_list.extend(itertools.combinations(con.features, n))
 
                 if con.none_also_valid:
-                    used_features_list.append(tuple([]))
+                    used_features_list.append(())
             else:
                 used_features_list.extend(
                     itertools.combinations(con.features, con.max_count)

bofire/data_models/domain/features.py

@@ -433,7 +433,7 @@ def _validate_transform_specs(self, specs: TInputTransformSpecs):
             raise ValueError("Unknown features specified in transform specs.")
         # next check that all values are of type CategoricalEncodingEnum
         if not (
-            all([isinstance(enc, CategoricalEncodingEnum) for enc in specs.values()])
+            all(isinstance(enc, CategoricalEncodingEnum) for enc in specs.values())
         ):
             raise ValueError("Unknown transform specified.")
         # next check that only Categoricalwithdescriptor have the value DESCRIPTOR
@@ -477,12 +477,12 @@ def get_bounds(
         upper = []
 
         for feat in self.get():
-            l, u = feat.get_bounds(  # type: ignore
+            lo, up = feat.get_bounds(  # type: ignore
                 transform_type=specs.get(feat.key),  # type: ignore
                 values=experiments[feat.key] if experiments is not None else None,
             )
-            lower += l
-            upper += u
+            lower += lo
+            upper += up
         return lower, upper
 
 

bofire/data_models/features/categorical.py

@@ -42,7 +42,7 @@ def validate_categories_unique(cls, categories):
         Returns:
             List[str]: List of the categories
         """
-        categories = [name for name in categories]
+        categories = list(categories)
         if len(categories) != len(set(categories)):
             raise ValueError("categories must be unique")
         return categories
@@ -180,9 +180,7 @@ def get_possible_categories(self, values: pd.Series) -> list:
         Returns:
             list: list of possible categories
         """
-        return sorted(
-            list(set(list(set(values.tolist())) + self.get_allowed_categories()))
-        )
+        return sorted(set(list(set(values.tolist())) + self.get_allowed_categories()))
 
     def to_onehot_encoding(self, values: pd.Series) -> pd.DataFrame:
         """Converts values to a one-hot encoding.
@@ -368,7 +366,7 @@ def validate_categories_unique(cls, categories):
         Returns:
             List[str]: List of the categories
         """
-        categories = [name for name in categories]
+        categories = list(categories)
         if len(categories) != len(set(categories)):
             raise ValueError("categories must be unique")
         return categories
@@ -386,7 +384,7 @@ def validate_objective(cls, objective, values):
 
     def to_dict(self) -> Dict:
         """Returns the catergories and corresponding objective values as dictionary"""
-        return {cat: obj for cat, obj in zip(self.categories, self.objective)}
+        return dict(zip(self.categories, self.objective))
 
     def __call__(self, values: pd.Series) -> pd.Series:
         return values.map(self.to_dict()).astype(float)

bofire/data_models/features/descriptor.py

@@ -43,7 +43,7 @@ def descriptors_to_keys(cls, descriptors):
         Returns:
             List[str]: List of valid keys
         """
-        return [name for name in descriptors]
+        return list(descriptors)
 
     @root_validator(pre=False, skip_on_failure=True)
     def validate_list_lengths(cls, values):
@@ -104,7 +104,7 @@ def validate_descriptors(cls, descriptors):
         Returns:
             List[str]: List of the descriptors
         """
-        descriptors = [name for name in descriptors]
+        descriptors = list(descriptors)
         if len(descriptors) != len(set(descriptors)):
             raise ValueError("descriptors must be unique")
         return descriptors
@@ -133,7 +133,7 @@ def validate_values(cls, v, values):
         a = np.array(v)
         for i, d in enumerate(values["descriptors"]):
             if len(set(a[:, i])) == 1:
-                raise ValueError("No variation for descriptor {d}.")
+                raise ValueError(f"No variation for descriptor {d}.")
         return v
 
     def to_df(self):
@@ -142,7 +142,7 @@ def to_df(self):
         Returns:
             pd.DataFrame: tabular overview of the feature as DataFrame
         """
-        data = {cat: values for cat, values in zip(self.categories, self.values)}
+        data = dict(zip(self.categories, self.values))
         return pd.DataFrame.from_dict(data, orient="index", columns=self.descriptors)
 
     def fixed_value(
@@ -234,7 +234,7 @@ def to_descriptor_encoding(self, values: pd.Series) -> pd.DataFrame:
         """
         return pd.DataFrame(
             data=values.map(
-                {cat: value for cat, value in zip(self.categories, self.values)}
+                dict(zip(self.categories, self.values))
             ).values.tolist(),  # type: ignore
             columns=[f"{self.key}{_CAT_SEP}{d}" for d in self.descriptors],
             index=values.index,

bofire/plot/objective.py

@@ -13,7 +13,7 @@ def plot_objective_plotly(
     lower: float,
     upper: float,
     values: Optional[pd.Series] = None,
-    layout_options: Dict = {},
+    layout_options: Optional[Dict] = None,
 ):
     """Plot the assigned objective.
 
@@ -43,7 +43,7 @@ def plot_objective_plotly(
     else:
         fig = fig1
 
-    if len(layout_options) > 0:
+    if layout_options is not None:
         fig.update_layout(layout_options)
 
     return fig

bofire/plot/prior.py

@@ -1,4 +1,4 @@
-from typing import Dict
+from typing import Dict, Optional
 
 import numpy as np
 import plotly.express as px
@@ -12,7 +12,7 @@ def plot_prior_pdf_plotly(
     prior: AnyPrior,
     lower: float,
     upper: float,
-    layout_options: Dict = {},
+    layout_options: Optional[Dict] = None,
 ):
     """Plot the probability density function of the prior with plotly.
 
@@ -33,7 +33,7 @@ def plot_prior_pdf_plotly(
         y=np.exp(priors.map(prior).log_prob(torch.from_numpy(x)).numpy()),
     )
 
-    if len(layout_options) > 0:
+    if layout_options is not None:
         fig.update_layout(layout_options)
 
     return fig

bofire/strategies/doe/design.py

@@ -28,7 +28,7 @@ def find_local_max_ipopt(
     model_type: Union[str, Formula],
     n_experiments: Optional[int] = None,
     delta: float = 1e-7,
-    ipopt_options: Dict = {},
+    ipopt_options: Optional[Dict] = None,
     sampling: Optional[pd.DataFrame] = None,
     fixed_experiments: Optional[pd.DataFrame] = None,
     objective: OptimalityCriterionEnum = OptimalityCriterionEnum.D_OPTIMALITY,
@@ -41,7 +41,7 @@ def find_local_max_ipopt(
         n_experiments (int): Number of experiments. By default the value corresponds to
             the number of model terms - dimension of ker() + 3.
         delta (float): Regularization parameter. Default value is 1e-3.
-        ipopt_options (Dict): options for IPOPT. For more information see [this link](https://coin-or.github.io/Ipopt/OPTIONS.html)
+        ipopt_options (Dict, optional): options for IPOPT. For more information see [this link](https://coin-or.github.io/Ipopt/OPTIONS.html)
         sampling (Sampling, np.ndarray): Sampling class or a np.ndarray object containing the initial guess.
         fixed_experiments (pd.DataFrame): dataframe containing experiments that will be definitely part of the design.
             Values are set before the optimization.
@@ -76,11 +76,9 @@ def find_local_max_ipopt(
 
     # check that NChooseK constraints only impose an upper bound on the number of nonzero components (and no lower bound)
     assert all(
-        [
-            c.min_count == 0
-            for c in domain.constraints
-            if isinstance(c, NChooseKConstraint)
-        ]
+        c.min_count == 0
+        for c in domain.constraints
+        if isinstance(c, NChooseKConstraint)
     ), "NChooseKConstraint with min_count !=0 is not supported!"
 
     # determine number of experiments (only relevant if n_experiments is not provided by the user)
@@ -140,6 +138,8 @@ def find_local_max_ipopt(
             x0[i] = val
 
     # set ipopt options
+    if ipopt_options is None:
+        ipopt_options = {}
     _ipopt_options = {"maxiter": 500, "disp": 0}
     for key in ipopt_options.keys():
         _ipopt_options[key] = ipopt_options[key]

bofire/strategies/doe/utils.py

@@ -474,7 +474,7 @@ def nchoosek_constraints_as_bounds(
                 ]
 
                 # find and shuffle all combinations of elements of ind of length max_active
-                ind = np.array([c for c in combinations(ind, r=n_inactive)])
+                ind = np.array(list(combinations(ind, r=n_inactive)))
                 np.random.shuffle(ind)
 
                 # set bounds to zero in each experiments for the variables that should be inactive

bofire/surrogates/cloudpickle_module.py

@@ -5,8 +5,8 @@
 try:
     from pickle import Unpickler  # noqa: F401
 
-    from cloudpickle import CloudPickler as Pickler  # noqa: F401
     from cloudpickle import *  # noqa: F401, F403,
+    from cloudpickle import CloudPickler as Pickler  # noqa: F401
 
 except ModuleNotFoundError:
     warnings.warn("Cloudpicke is not available.")

bofire/surrogates/diagnostics.py

@@ -293,9 +293,9 @@ def validate_results(cls, v, values):
                     )
         # check columns of X
         if v[0].X is not None:
-            cols = sorted(list(v[0].X.columns))
+            cols = sorted(v[0].X.columns)
             for i in v:
-                if sorted(list(i.X.columns)) != cols:
+                if sorted(i.X.columns) != cols:
                     raise ValueError("Columns of X do not match.")
         return v
 

bofire/surrogates/mlp.py

@@ -132,7 +132,7 @@ def fit_mlp(
     optimizer = torch.optim.Adam(mlp.parameters(), lr=lr, weight_decay=weight_decay)
     for _ in range(n_epoches):
         current_loss = 0.0
-        for i, data in enumerate(train_loader, 0):
+        for data in train_loader:
             # Get and prepare inputs
             inputs, targets = data
             if len(targets.shape) == 1:
@@ -181,7 +181,7 @@ def _fit(self, X: pd.DataFrame, Y: pd.DataFrame):
 
         mlps = []
         subsample_size = round(self.subsample_fraction * X.shape[0])
-        for i in range(self.n_estimators):
+        for _ in range(self.n_estimators):
             # resample X and Y
             sample_idx = np.random.choice(X.shape[0], replace=True, size=subsample_size)
             tX = torch.from_numpy(transformed_X.values[sample_idx]).to(**tkwargs)

bofire/surrogates/surrogate.py

@@ -59,7 +59,7 @@ def validate_predictions(self, predictions: pd.DataFrame) -> pd.DataFrame:
         expected_cols = [
             f"{key}_{t}" for key in self.outputs.get_keys() for t in ["pred", "sd"]
         ]
-        if sorted(list(predictions.columns)) != sorted(expected_cols):
+        if sorted(predictions.columns) != sorted(expected_cols):
             raise ValueError(
                 f"Predictions are ill-formatted. Expected: {expected_cols}, got: {list(predictions.columns)}."
             )