MNT: flake8: fix E704, E226; ignore = F824, W503, W504

Author: kernc

benchmark/__main__.py

@@ -124,14 +124,14 @@ def method_note(method):
         return ' \N{DAGGER}' if is_nonconstrained else ''
 
     header = f'{"Test function":24s}\t{"Method":24s}\tN Evals\tError %\tDuration'.expandtabs(4)
-    print(header, '—'*len(header), sep='\n')
+    print(header, '—' * len(header), sep='\n')
     for r in sorted(results, key=lambda r: (r['func'], r['error'], r['nfev'], r['method'])):
         print(f"{r['func']:24s}\t{r['method'] + method_note(r['method']):24s}\t{str(r['nfev']):>6s}{('' if r['success'] else '*')}\t{r['error']:7d}\t{r['duration']:5.2f}".expandtabs(4))  # noqa: E501
 
     print('\n')
 
     header = f'{"Method":24s}\t{"Correct":7s}\tN Evals\tError %\tDuration'.expandtabs(4)
-    print(header, '—'*len(header), sep='\n')
+    print(header, '—' * len(header), sep='\n')
 
     def key_func(r):
         return r['method']

benchmark/funcs.py

@@ -24,24 +24,24 @@
         'name': 'rosenbrock',
         'func': rosen,
         'constraints': lambda x: sum(x**2) <= 1.1 * len(x),
-        'bounds': [[-2., 2.1]]*10,
+        'bounds': [[-2., 2.1]] * 10,
         'tol': 1,
         'solution': 0,
         'codomain': [0, 3500],
     },
     {
         'name': 'bird',
-        'func': lambda x: sin(x[1]) * exp((1-cos(x[0]))**2) + cos(x[0]) * exp((1-sin(x[1]))**2) + (x[0] - x[1])**2,
-        'constraints': lambda x: (x[0] + 5)++2 + (x[1] + 5)**2 < 25,
+        'func': lambda x: sin(x[1]) * exp((1 - cos(x[0]))**2) + cos(x[0]) * exp((1 - sin(x[1]))**2) + (x[0] - x[1])**2,
+        'constraints': lambda x: (x[0] + 5)**2 + (x[1] + 5)**2 < 25,
         'bounds': [[-10., 0.], [-6.5, 0.]],
         'tol': 1,
         'solution': -106.77,
         'codomain': [-106.77, 97.7],
     },
     {
         'name': 'gomez-levy',
-        'func': lambda x: 4*x[0]**2 - 2.1*x[0]**4 + 1/3*x[0]**6 + x[0]*x[1] - 4*x[1]**2 + 4*x[1]**4,
-        'constraints': lambda x: -sin(4*pi*x[0]) + 2*sin(2*pi*x[1])**2 <= 1.5,
+        'func': lambda x: 4 * x[0]**2 - 2.1 * x[0]**4 + x[0]**6 / 3 + x[0] * x[1] - 4 * x[1]**2 + 4 * x[1]**4,
+        'constraints': lambda x: -sin(4 * pi * x[0]) + 2 * sin(2 * pi * x[1])**2 <= 1.5,
         'bounds': [[-1., .75], [-1., 1.]],
         'tol': .1,
         'solution': -1.0317,
@@ -78,7 +78,7 @@
     },
     {
         'name': '6-hump-camelback',
-        'func': lambda x: (4 - 2.1*x[0]**2 + x[0]**4/3)*x[0]**2 + x[0]*x[1] + (-4 + 4*x[1]**2)*x[1]**2,
+        'func': lambda x: (4 - 2.1 * x[0]**2 + x[0]**4 / 3) * x[0]**2 + x[0] * x[1] + (-4 + 4 * x[1]**2) * x[1]**2,
         'constraints': None,
         'bounds': [[-2., 2.], [-1., 1.]],
         'tol': 1,
@@ -87,7 +87,7 @@
     },
     {
         'name': 'griewank',
-        'func': lambda x: 1 + sum(x**2) / 4000 - prod(cos(x/sqrt(1 + arange(len(x))))),
+        'func': lambda x: 1 + sum(x**2) / 4000 - prod(cos(x / sqrt(1 + arange(len(x))))),
         'constraints': None,
         'bounds': [[-600., 590.], [-590., 600.]],
         'tol': 300,

sambo/_smbo.py

@@ -15,7 +15,8 @@
 )
 
 
-def _LCB(*, mean, std, kappa): return mean - np.outer(kappa, std)
+def _LCB(*, mean, std, kappa):
+    return mean - np.outer(kappa, std)
 
 
 class Optimizer:
@@ -243,7 +244,7 @@ def _fit(self):
     def _predict(self, X):
         means, stds, masks = [], [], []
         for estimator in self.estimators:
-            X_batched = [X[i:i+10_000] for i in range(0, len(X), 10_000)]
+            X_batched = [X[i:i + 10_000] for i in range(0, len(X), 10_000)]
             try:
                 mean, std = np.concatenate(
                     [estimator.predict(X, return_std=True) for X in X_batched], axis=1)

sambo/_test.py

@@ -78,7 +78,7 @@ def test_ask_tell_interface(self):
         np.testing.assert_array_equal(res.funv, np.arange(4))
 
     def test_multiple_runs_continue_ie_warm_start(self):
-        optimizer = Optimizer(fun=sum, x0=[0]*3, n_init=1)
+        optimizer = Optimizer(fun=sum, x0=[0] * 3, n_init=1)
         optimizer.run(max_iter=1)
         optimizer.run(max_iter=1)
         res = optimizer.run(max_iter=1)
@@ -325,7 +325,7 @@ def test_estimator_factory(self):
         }
         for estimator in BUILTIN_ESTIMATORS:
             with self.subTest(estimator=estimator):
-                res = smbo(lambda x: sum((x-2)**2), bounds=[(-100, 100)], estimator=estimator,
+                res = smbo(lambda x: sum((x - 2)**2), bounds=[(-100, 100)], estimator=estimator,
                            **dict(DEFAULT_KWARGS, **ESTIMATOR_KWARGS[estimator]))
                 self.assertLess(res.fun, 1, msg=res)
 
@@ -364,7 +364,7 @@ def test_make_doc_plots(self):
         }
         results = [
             minimize(
-                rosen, bounds=[(-2., 2.)]*2,
+                rosen, bounds=[(-2., 2.)] * 2,
                 constraints=lambda x: sum(x**2) <= 2**len(x),
                 max_iter=100, method=method, rng=2,
                 **KWARGS.get(method, {}),
@@ -418,7 +418,7 @@ def evaluate(x):
 
         results = []
         for estimator in BUILTIN_ESTIMATORS:
-            optimizer = Optimizer(fun=None, bounds=[(-2, 2)]*4, estimator=estimator, rng=0)
+            optimizer = Optimizer(fun=None, bounds=[(-2, 2)] * 4, estimator=estimator, rng=0)
 
             for i in range(30):
                 suggested_x = optimizer.ask(n_candidates=1)

sambo/_util.py

@@ -84,8 +84,10 @@ def wrapper(*args, **kwargs):
 
 @runtime_checkable
 class _SklearnLikeRegressor(Protocol):
-    def fit(self, X, y): pass
-    def predict(self, X, return_std=False): pass
+    def fit(self, X: np.ndarray, y: np.ndarray):
+        pass
+    def predict(self, X, return_std=False):  # noqa: E301
+        pass
 
 
 class OptimizeResult(_OptimizeResult):

sambo/plot.py

@@ -92,7 +92,7 @@ def plot_convergence(
         mins = np.minimum.accumulate(result.funv)
 
         ax.plot(range(1, nfev + 1), mins,
-                label=name, marker=next(MARKER), markevery=(.05 + .05*i, .2),
+                label=name, marker=next(MARKER), markevery=(.05 + .05 * i, .2),
                 linestyle='--', alpha=.7, markersize=6, lw=2)
 
     if true_minimum is not None:
@@ -195,7 +195,7 @@ def plot_regret(
                    for i in range(1, nfev + 1)]
 
         ax.plot(range(1, nfev + 1), regrets,
-                label=name, marker=next(MARKER), markevery=(.05 + .05*i, .2),
+                label=name, marker=next(MARKER), markevery=(.05 + .05 * i, .2),
                 linestyle='--', alpha=.7, markersize=6, lw=2)
 
     if name is not None:

setup.cfg

@@ -1,5 +1,9 @@
 [flake8]
 max-line-length = 120
+# F824 `nonlocal x` is unused: name is never assigned in scope
+# W503 Line break before a binary operator
+# W504 Line break after a binary operator -- https://www.flake8rules.com/rules/W504.html
+ignore = F824, W503, W504
 
 [mypy]
 warn_unused_ignores = True