Merge pull request #4910 from jbrockmendel/blanks

CLN: Blank Lines E301,E302,E303,E305,E306 in examples, tools, sm.base

examples/incomplete/wls_extended.py

@@ -47,7 +47,6 @@
 rsquared_adj = 1 -(wls_fit.nobs)/(wls_fit.df_resid)*(1-rsquared)
 
 
-
 #Trying to figure out what's going on in this example
 #----------------------------------------------------
 
@@ -80,6 +79,7 @@
 #@savefig wls_robust_compare.png
 plt.ylim([0,2000]);
 
+
 #What is going on? A more systematic look at the data
 #----------------------------------------------------
 
@@ -109,7 +109,6 @@ def getrsq(fitresult):
         endog = fitresult[1]
         nobs = resid.shape[0]
 
-
     rss = np.dot(resid, resid)
     tss = np.var(endog)*nobs
     return 1-rss/tss, rss, tss, tss-rss

examples/python/contrasts.py

@@ -65,9 +65,11 @@
 
 from patsy.contrasts import ContrastMatrix
 
+
 def _name_levels(prefix, levels):
     return ["[%s%s]" % (prefix, level) for level in levels]
 
+
 class Simple(object):
     def _simple_contrast(self, levels):
         nlevels = len(levels)

examples/python/formulas.py

@@ -106,6 +106,8 @@
 
 def log_plus_1(x):
     return np.log(x) + 1.
+
+
 res = sm.ols(formula='Lottery ~ log_plus_1(Literacy)', data=df).fit()
 print(res.params)
 

examples/python/glm_formula.py

@@ -28,6 +28,8 @@
 
 def double_it(x):
     return 2 * x
+
+
 formula = 'SUCCESS ~ double_it(LOWINC) + PERASIAN + PERBLACK + PERHISP + PCTCHRT +            PCTYRRND + PERMINTE*AVYRSEXP*AVSALK + PERSPENK*PTRATIO*PCTAF'
 mod2 = smf.glm(formula=formula, data=dta, family=sm.families.Binomial()).fit()
 mod2.summary()

examples/python/quantile_regression.py

@@ -43,10 +43,13 @@
 # For convenience, we place the quantile regression results in a Pandas DataFrame, and the OLS results in a dictionary.
 
 quantiles = np.arange(.05, .96, .1)
+
+
 def fit_model(q):
     res = mod.fit(q=q)
     return [q, res.params['Intercept'], res.params['income']] + res.conf_int().loc['income'].tolist()
-
+
+
 models = [fit_model(x) for x in quantiles]
 models = pd.DataFrame(models, columns=['q', 'a', 'b','lb','ub'])
 

statsmodels/base/_constraints.py

@@ -128,7 +128,6 @@ def reduce(self, params):
         return params.dot(self.transf_mat)
 
 
-
 def transform_params_constraint(params, Sinv, R, q):
     """find the parameters that statisfy linear constraint from unconstraint
 

statsmodels/base/_penalized.py

@@ -64,7 +64,6 @@ def loglike(self, params, pen_weight=None, **kwds):
 
         return llf
 
-
     def loglikeobs(self, params, pen_weight=None, **kwds):
         if pen_weight is None:
             pen_weight = self.pen_weight

statsmodels/base/_penalties.py

@@ -23,6 +23,7 @@
 
 import numpy as np
 
+
 def _check_wts(weights, wts):
     """helper function for deprecating `wts`
     """

statsmodels/base/_screening.py

@@ -420,7 +420,6 @@ def screen_exog_iterator(self, exog_iterator):
         exog_winner = np.column_stack(exog_winner)
         res_screen_final = self.screen_exog(exog_winner, maxiter=20)
 
-
         exog_winner_names = ['var%d_%d' % (bidx, idx)
                              for bidx, batch in enumerate(exog_idx)
                              for idx in batch]

statsmodels/base/data.py

@@ -176,7 +176,6 @@ def _handle_constant(self, hasconst):
                 self.k_constant = int(rank_orig == rank_augm)
                 self.const_idx = None
 
-
     @classmethod
     def _drop_nans(cls, x, nan_mask):
         return x[nan_mask]
@@ -569,6 +568,7 @@ def attach_ynames(self, result):
         else:
             return DataFrame(result, columns=self.ynames)
 
+
 def _make_endog_names(endog):
     if endog.ndim == 1 or endog.shape[1] == 1:
         ynames = ['y']

statsmodels/base/elastic_net.py

@@ -365,8 +365,6 @@ class RegularizedResultsWrapper(wrap.ResultsWrapper):
         'resid': 'rows',
         'fittedvalues': 'rows',
     }
-
     _wrap_attrs = _attrs
-
-wrap.populate_wrapper(RegularizedResultsWrapper,
+wrap.populate_wrapper(RegularizedResultsWrapper,  # noqa:E305
                       RegularizedResults)

statsmodels/base/l1_solvers_common.py

@@ -5,6 +5,7 @@
 import numpy as np
 from statsmodels.compat.python import range
 
+
 def qc_results(params, alpha, score, qc_tol, qc_verbose=False):
     """
     Theory dictates that one of two conditions holds:

statsmodels/base/optimizer.py

@@ -7,6 +7,7 @@
 import numpy as np
 from scipy import optimize
 
+
 def _check_method(method, methods):
     if method not in methods:
         message = "Unknown fit method %s" % method
@@ -213,7 +214,6 @@ def _fit_constrained(self, params):
         """
         pass
 
-
     def _fit_regularized(self, params):
         #TODO: code won't necessarily be general here. 3 options.
         # 1) setup for scipy.optimize.fmin_sqlsqp

statsmodels/base/tests/test_data.py

@@ -10,6 +10,7 @@
 from statsmodels.genmod import families
 from statsmodels.discrete.discrete_model import Logit
 
+
 #class TestDates(object):
 #    @classmethod
 #    def setup_class(cls):
@@ -20,6 +21,7 @@
 #        np.testing.assert_equal(data.wrap_output(self.dates_input, 'dates'),
 #                                self.dates_result)
 
+
 class TestArrays(object):
     @classmethod
     def setup_class(cls):
@@ -391,6 +393,7 @@ def test_endogexog(self):
         np.testing.assert_equal(self.data.endog, self.endog.values.squeeze())
         np.testing.assert_equal(self.data.exog, self.exog.values[:,None])
 
+
 def test_alignment():
     #Fix Issue #206
     from statsmodels.regression.linear_model import OLS
@@ -412,6 +415,7 @@ def test_alignment():
     # which index do we get??
     np.testing.assert_raises(ValueError, OLS, *(endog, exog))
 
+
 class TestMultipleEqsArrays(TestArrays):
     @classmethod
     def setup_class(cls):
@@ -498,6 +502,7 @@ def test_attach(self):
                                                  'columns_eq'),
                                 self.col_eq_result)
 
+
 class TestMissingArray(object):
     @classmethod
     def setup_class(cls):
@@ -564,6 +569,7 @@ def test_extra_kwargs_1d(self):
         weights = weights[idx]
         np.testing.assert_array_equal(data.weights, weights)
 
+
 class TestMissingPandas(object):
     @classmethod
     def setup_class(cls):
@@ -740,7 +746,6 @@ def test_hasconst(self):
                 assert_equal(res.model.k_constant, result[0])
                 assert_equal(res.model.data.k_constant, result[0])
 
-
     @classmethod
     def setup_class(cls):
         # create data
@@ -797,6 +802,7 @@ def mod(y, x):
     def _initialize(cls):
         cls.y = cls.y_bin
 
+
 class TestHasConstantLogit(CheckHasConstant):
 
     @classmethod

statsmodels/base/tests/test_generic_methods.py

@@ -77,7 +77,6 @@ def test_ttest_tvalues(self):
                 'Conf. Int. Low', 'Conf. Int. Upp.']
         assert_array_equal(summf.columns.values, cols)
 
-
     def test_ftest_pvalues(self):
         res = self.results
         use_t = res.use_t
@@ -568,7 +567,6 @@ def setup_class(cls):
         cls.data = test.data.drop([0,1,2])
         cls.initialize()
 
-
     def test_combined(self):
         res = self.res
         wa = res.wald_test_terms(skip_single=False, combine_terms=['Duration', 'Weight'])
@@ -582,7 +580,6 @@ def test_combined(self):
 
         compare_waldres(res, wa, [c_const, c_d, c_w, c_dw, c_duration, c_weight])
 
-
     def test_categories(self):
         # test only multicolumn terms
         res = self.res
@@ -632,7 +629,6 @@ def initialize(cls):
         mod = ols("np.log(Days+1) ~ C(Duration, Sum)*C(Weight, Sum)", cls.data)
         cls.res = mod.fit(use_t=False)
 
-
     def test_noformula(self):
         endog = self.res.model.endog
         exog = self.res.model.data.orig_exog
@@ -762,7 +758,6 @@ def setup_class(cls):
                            'C(Weight)[T.3]',
                            'C(Weight)[T.3] - C(Weight)[T.2]']
 
-
     def test_alpha(self):
         pw1 = self.res.t_test_pairwise(self.term_name, method='hommel',
                                        factor_labels='A B C'.split())
@@ -816,6 +811,7 @@ def setup_class(cls):
                            'C(Weight)[3] - C(Weight)[1]',
                            'C(Weight)[3] - C(Weight)[2]']
 
+
 class TestTTestPairwiseOLS4(CheckPairwise):
 
     @classmethod

statsmodels/base/tests/test_optimize.py

@@ -20,12 +20,15 @@
 def dummy_func(x):
     return x**2
 
+
 def dummy_score(x):
     return 2.*x
 
+
 def dummy_hess(x):
     return [[2.]]
 
+
 def test_full_output_false():
     # just a smoke test
 

statsmodels/base/tests/test_penalized.py

@@ -165,8 +165,6 @@ def _initialize(cls):
 
 
 class TestPenalizedPoissonOracle(CheckPenalizedPoisson):
-
-
     @classmethod
     def _initialize(cls):
         y, x = cls.y, cls.x
@@ -355,7 +353,6 @@ def _initialize(cls):
         cls.atol = 1e-12
         cls.k_params = cls.k_nonzero
 
-
     def test_cov_type(self):
         res1 = self.res1
         res2 = self.res2
@@ -526,7 +523,6 @@ def _initialize(cls):
         cls.atol = 1e-10
         cls.k_params = 4
 
-
     def test_deriv(self):
         res1 = self.res1
         res2 = self.res2
@@ -693,7 +689,6 @@ def _initialize(cls):
         # the corresponding Theil penweight seems to be 2 * nobs / sigma2_e
         cls.res2 = modp.fit(pen_weight=120.74564413221599 * 1000)
 
-
         pen = smpen.L2ContraintsPenalty(restriction=restriction)
         mod = GLMPenalized(y, x, family=family.Gaussian(),
                            penal=pen)

statsmodels/base/tests/test_penalties.py

@@ -12,6 +12,7 @@
 import statsmodels.base._penalties as smpen
 from statsmodels.tools.numdiff import approx_fprime, approx_hess
 
+
 class CheckPenalty(object):
 
     def test_symmetry(self):

statsmodels/base/tests/test_predict.py

@@ -31,7 +31,6 @@ def test_2d(self):
         assert_equal(pred.index, np.arange(len(pred)))
         assert_allclose(pred.values, fitted.values, rtol=1e-13)
 
-
     def test_1d(self):
         # one observation
         res = self.res
@@ -47,7 +46,6 @@ def test_1d(self):
         assert_equal(pred.index, np.arange(1))
         assert_allclose(pred, fittedm, rtol=1e-13)
 
-
         # Series
         pred = res.predict(data.mean())
         assert_equal(pred.index, np.arange(1))

statsmodels/base/tests/test_screening.py

@@ -49,6 +49,7 @@ def _get_poisson_data():
     y = np.random.poisson(mu)
     return y, x, idx_nonzero_true, beta
 
+
 def test_poisson_screening():
 
     np.random.seed(987865)
@@ -271,7 +272,6 @@ def test_glmlogit_screening():
 
     res_screen.results_final
 
-
     xnames = ['var%4d' % ii for ii in res_screen.idx_nonzero]
     xnames[0] = 'const'
 

statsmodels/base/tests/test_shrink_pickle.py

@@ -166,6 +166,7 @@ def setup(self):
         #TODO: temporary, fixed in master
         self.predict_kwds = dict(exposure=1, offset=0)
 
+
 class TestRemoveDataPickleNegativeBinomial(RemoveDataPickle):
 
     def setup(self):
@@ -176,6 +177,7 @@ def setup(self):
         mod = sm.NegativeBinomial(data.endog, data.exog)
         self.results = mod.fit(disp=0)
 
+
 class TestRemoveDataPickleLogit(RemoveDataPickle):
 
     def setup(self):
@@ -260,8 +262,11 @@ class TestPickleFormula4(TestPickleFormula2):
     def setup(self):
         self.results = sm.OLS.from_formula("Y ~ np.log(abs(A) + 1) + B * C", data=self.data).fit()
 
-# we need log in module namespace for the following test
+
+# we need log in module namespace for TestPickleFormula5
 from numpy import log
+
+
 class TestPickleFormula5(TestPickleFormula2):
 
     def setup(self):
@@ -281,7 +286,6 @@ def setup(self):
         self.results = model.fit_regularized(method='l1', disp=0, alpha=10)
 
 
-
 if __name__ == '__main__':
     for cls in [TestRemoveDataPickleOLS, TestRemoveDataPickleWLS,
                 TestRemoveDataPicklePoisson,