ruff format

marginaleffects/__init__.py

@@ -1,8 +0,0 @@
-from .comparisons import avg_comparisons, comparisons
-from .datagrid import datagrid, datagridcf
-from .hypotheses import hypotheses
-from .plot_comparisons import plot_comparisons
-from .plot_predictions import plot_predictions
-from .plot_slopes import plot_slopes
-from .predictions import avg_predictions, predictions
-from .slopes import avg_slopes, slopes

marginaleffects/by.py

@@ -2,9 +2,13 @@
 
 
 def get_by(model, estimand, newdata, by=None, wts=None):
-
     # for predictions
-    if isinstance(by, list) and len(by) == 1 and by[0] == "group" and "group" not in estimand.columns:
+    if (
+        isinstance(by, list)
+        and len(by) == 1
+        and by[0] == "group"
+        and "group" not in estimand.columns
+    ):
         by = True
 
     if by is True:
@@ -20,12 +24,10 @@ def get_by(model, estimand, newdata, by=None, wts=None):
     else:
         out = pl.DataFrame({"estimate": estimand["estimate"]})
 
-
     by = [x for x in by if x in out.columns]
     if isinstance(by, list) and len(by) == 0:
         return out
 
-
     if wts is None:
         out = out.groupby(by, maintain_order=True).agg(pl.col("estimate").mean())
     else:

marginaleffects/classes.py

@@ -22,7 +22,7 @@ def __str__(self):
             "std_error": "Std.Error",
             "statistic": "z",
             "p_value": "P(>|z|)",
-            "s_value": "S"
+            "s_value": "S",
         }
 
         if hasattr(self, "conf_level"):
@@ -49,7 +49,7 @@ def __str__(self):
             valid = list(mapping.keys())
 
         valid = self.datagrid_explicit + valid
-        
+
         valid = [x for x in valid if x in self.columns]
         mapping = {key: mapping[key] for key in mapping if key in valid}
         tmp = self.select(valid).rename(mapping)

marginaleffects/comparisons.py

@@ -10,8 +10,13 @@
 from .estimands import estimands
 from .getters import get_coef, get_modeldata, get_predict
 from .hypothesis import get_hypothesis
-from .sanity import (sanitize_by, sanitize_hypothesis_null, sanitize_newdata,
-                     sanitize_variables, sanitize_vcov)
+from .sanity import (
+    sanitize_by,
+    sanitize_hypothesis_null,
+    sanitize_newdata,
+    sanitize_variables,
+    sanitize_vcov,
+)
 from .transform import get_transform
 from .uncertainty import get_jacobian, get_se, get_z_p_ci
 from .utils import get_pad, sort_columns, upcast
@@ -190,9 +195,7 @@ def inner(coefs, by, hypothesis, wts, nd):
 
         # estimates
         tmp = [
-            get_predict(model, get_coef(model), nd_X).rename(
-                {"estimate": "predicted"}
-            ),
+            get_predict(model, get_coef(model), nd_X).rename({"estimate": "predicted"}),
             get_predict(model, coefs, lo_X)
             .rename({"estimate": "predicted_lo"})
             .select("predicted_lo"),
@@ -270,7 +273,9 @@ def outer(x):
         J = get_jacobian(func=outer, coefs=get_coef(model))
         se = get_se(J, V)
         out = out.with_columns(pl.Series(se).alias("std_error"))
-        out = get_z_p_ci(out, model, conf_level=conf_level, hypothesis_null=hypothesis_null)
+        out = get_z_p_ci(
+            out, model, conf_level=conf_level, hypothesis_null=hypothesis_null
+        )
 
     out = get_transform(out, transform=transform)
     out = get_equivalence(out, equivalence=equivalence, df=np.inf)

marginaleffects/datagrid.py

@@ -10,7 +10,7 @@ def datagrid(
     newdata=None,
     FUN_numeric=lambda x: x.mean(),
     FUN_other=lambda x: x.mode()[0],  # mode can return multiple values
-    **kwargs
+    **kwargs,
 ):
     """
     Data grids
@@ -95,7 +95,6 @@ def datagrid(
     return out
 
 
-
 def datagridcf(model=None, newdata=None, **kwargs):
     """
     Data grids
@@ -122,20 +121,22 @@ def datagridcf(model=None, newdata=None, **kwargs):
         newdata = get_modeldata(model)
 
     if "rowid" not in newdata.columns:
-        newdata = newdata.with_columns(pl.Series(range(newdata.shape[0])).alias("rowid"))
-    newdata = newdata.rename({"rowid" : "rowidcf"})
+        newdata = newdata.with_columns(
+            pl.Series(range(newdata.shape[0])).alias("rowid")
+        )
+    newdata = newdata.rename({"rowid": "rowidcf"})
 
     # Create dataframe from kwargs
     dfs = [pl.DataFrame({k: v}) for k, v in kwargs.items()]
 
     # Perform cross join
-    df_cross = reduce(lambda df1, df2: df1.join(df2, how='cross'), dfs)
+    df_cross = reduce(lambda df1, df2: df1.join(df2, how="cross"), dfs)
 
     # Drop would-be duplicates
     newdata = newdata.drop(df_cross.columns)
 
-    result = newdata.join(df_cross, how = "cross")
+    result = newdata.join(df_cross, how="cross")
 
     result.datagrid_explicit = list(kwargs.keys())
 
-    return result
+    return result

marginaleffects/estimands.py

@@ -26,7 +26,7 @@ def prep(x):
     "dydxavg": lambda hi, lo, eps, x, y, w: prep(((hi - lo) / eps).mean()),
     "eyexavg": lambda hi, lo, eps, x, y, w: prep(((hi - lo) / eps * (x / y)).mean()),
     "eydxavg": lambda hi, lo, eps, x, y, w: prep((((hi - lo) / eps) / y).mean()),
-    "dyexavg": lambda hi, lo, eps, x, y, w: prep(((((hi - lo) / eps) * x)).mean()),
+    "dyexavg": lambda hi, lo, eps, x, y, w: prep((((hi - lo) / eps) * x).mean()),
     "dydxavgwts": lambda hi, lo, eps, x, y, w: prep(
         (((hi - lo) / eps) * w).sum() / w.sum()
     ),

marginaleffects/getters.py

@@ -22,7 +22,6 @@ def get_coef(model):
 
 
 def get_vcov(model, vcov=True):
-
     if isinstance(vcov, bool):
         if vcov is True:
             V = model.cov_params()
@@ -37,8 +36,10 @@ def get_vcov(model, vcov=True):
             raise ValueError(f"The model object has no {lab} attribute.")
 
     else:
-        raise ValueError('`vcov` must be a boolean or a string like "HC3", which corresponds to an attribute of the model object such as "vcov_HC3".')
-
+        raise ValueError(
+            '`vcov` must be a boolean or a string like "HC3", which corresponds to an attribute of the model object such as "vcov_HC3".'
+        )
+
     V = np.array(V)
 
     return V
@@ -87,4 +88,4 @@ def get_predict(model, params, newdata: pl.DataFrame):
             "The `predict()` method must return an array with 1 or 2 dimensions."
         )
     p = p.with_columns(pl.col("rowid").cast(pl.Int32))
-    return p
+    return p

marginaleffects/hypotheses.py

@@ -1,4 +1,3 @@
-import numpy as np
 import polars as pl
 
 from .classes import MarginaleffectsDataFrame
@@ -13,27 +12,27 @@
 def hypotheses(model, hypothesis=None, conf_level=0.95, vcov=True, equivalence=None):
     """
     (Non-)Linear Tests for Null Hypotheses, Joint Hypotheses, Equivalence, Non Superiority, and Non Inferiority.
-    
-    This function calculates uncertainty estimates as first-order approximate standard errors for linear or non-linear 
-    functions of a vector of random variables with known or estimated covariance matrix. It emulates the behavior of 
-    the excellent and well-established `car::deltaMethod` and `car::linearHypothesis` functions in R, but it supports 
-    more models; requires fewer dependencies; expands the range of tests to equivalence and superiority/inferiority; 
+
+    This function calculates uncertainty estimates as first-order approximate standard errors for linear or non-linear
+    functions of a vector of random variables with known or estimated covariance matrix. It emulates the behavior of
+    the excellent and well-established `car::deltaMethod` and `car::linearHypothesis` functions in R, but it supports
+    more models; requires fewer dependencies; expands the range of tests to equivalence and superiority/inferiority;
     and offers convenience features like robust standard errors.
-    
+
     To learn more, visit the package website: <https://marginaleffects.com/>
-    
-    Warning #1: Tests are conducted directly on the scale defined by the `type` argument. For some models, it can make 
-    sense to conduct hypothesis or equivalence tests on the `"link"` scale instead of the `"response"` scale which is 
+
+    Warning #1: Tests are conducted directly on the scale defined by the `type` argument. For some models, it can make
+    sense to conduct hypothesis or equivalence tests on the `"link"` scale instead of the `"response"` scale which is
     often the default.
-    
-    Warning #2: For hypothesis tests on objects produced by the `marginaleffects` package, it is safer to use the 
-    `hypothesis` argument of the original function. Using `hypotheses()` may not work in certain environments, in lists, 
+
+    Warning #2: For hypothesis tests on objects produced by the `marginaleffects` package, it is safer to use the
+    `hypothesis` argument of the original function. Using `hypotheses()` may not work in certain environments, in lists,
     or when working programmatically with *apply style functions.
-    
-    Warning #3: The tests assume that the `hypothesis` expression is (approximately) normally distributed, which for 
-    non-linear functions of the parameters may not be realistic. More reliable confidence intervals can be obtained using 
+
+    Warning #3: The tests assume that the `hypothesis` expression is (approximately) normally distributed, which for
+    non-linear functions of the parameters may not be realistic. More reliable confidence intervals can be obtained using
     the `inferences()` function with `method = "boot"`.
-    
+
     Parameters:
     model : object
         Model object estimated by `statsmodels`
@@ -45,28 +44,28 @@ def hypotheses(model, hypothesis=None, conf_level=0.95, vcov=True, equivalence=N
         Whether to use the covariance matrix in the hypothesis test. Default is True.
     equivalence : tuple, optional
         The equivalence range for the hypothesis test. Default is None.
-        
+
     Returns:
     MarginaleffectsDataFrame
         A DataFrame containing the results of the hypothesis tests.
-        
+
     Examples:
 
         # When `hypothesis` is `None`, `hypotheses()` returns a DataFrame of parameters
         hypotheses(model)
 
         # A different null hypothesis
         hypotheses(model, hypothesis = 3)
-        
+
         # Test of equality between coefficients
         hypotheses(model, hypothesis="param1 = param2")
-        
+
         # Non-linear function
         hypotheses(model, hypothesis="exp(param1 + param2) = 0.1")
-        
+
         # Robust standard errors
         hypotheses(model, hypothesis="param1 = param2", vcov="HC3")
-        
+
         # Equivalence, non-inferiority, and non-superiority tests
         hypotheses(model, equivalence=(0, 10))
     """
@@ -86,7 +85,9 @@ def fun(x):
         J = get_jacobian(fun, get_coef(model))
         se = get_se(J, V)
         out = out.with_columns(pl.Series(se).alias("std_error"))
-        out = get_z_p_ci(out, model, conf_level=conf_level, hypothesis_null=hypothesis_null)
+        out = get_z_p_ci(
+            out, model, conf_level=conf_level, hypothesis_null=hypothesis_null
+        )
     out = get_equivalence(out, equivalence=equivalence)
     out = sort_columns(out, by=None)
     out = MarginaleffectsDataFrame(out, conf_level=conf_level)

marginaleffects/hypothesis.py

@@ -71,7 +71,6 @@ def get_hypothesis(x, hypothesis):
         elif hypothesis == "revpairwise":
             hypmat = lincom_revpairwise(x)
         else:
-
             raise ValueError(msg)
         out = lincom_multiply(x, hypmat.to_numpy())
         out = out.with_columns(pl.Series(hypothesis.columns).alias("term"))