Merge branch 'develop' into patch-1

ydataai · Oct 28, 2024 · 297e122 · 297e122
2 parents 9232eb0 + 5bbd589
commit 297e122
Show file tree

Hide file tree

Showing 4 changed files with 13 additions and 8 deletions.
diff --git a/src/ydata_profiling/config.py b/src/ydata_profiling/config.py
@@ -112,6 +112,8 @@ class TimeseriesVars(BaseModel):
     lags: List[int] = [1, 7, 12, 24, 30]
     significance: float = 0.05
     pacf_acf_lag: int = 100
+    autolag: Optional[str] = "AIC"
+    maxlag: Optional[int] = None
 
 
 class Univariate(BaseModel):

diff --git a/src/ydata_profiling/model/pandas/describe_timeseries_pandas.py b/src/ydata_profiling/model/pandas/describe_timeseries_pandas.py
@@ -16,12 +16,15 @@
 
 
 def stationarity_test(config: Settings, series: pd.Series) -> Tuple[bool, float]:
-    significance_threshold = config.vars.timeseries.significance
-
     # make sure the data has no missing values
-    adfuller_test = adfuller(series.dropna())
+    adfuller_test = adfuller(
+        series.dropna(),
+        autolag=config.vars.timeseries.autolag,
+        maxlag=config.vars.timeseries.maxlag,
+    )
     p_value = adfuller_test[1]
 
+    significance_threshold = config.vars.timeseries.significance
     return p_value < significance_threshold, p_value
 
 

diff --git a/src/ydata_profiling/model/typeset.py b/src/ydata_profiling/model/typeset.py
@@ -122,7 +122,7 @@ def get_relations() -> Sequence[TypeRelation]:
         @series_handle_nulls
         def contains_op(series: pd.Series, state: dict) -> bool:
             return (
-                not pdt.is_categorical_dtype(series)
+                not isinstance(series.dtype, pd.CategoricalDtype)
                 and pdt.is_string_dtype(series)
                 and series_is_string(series, state)
             )
@@ -205,9 +205,9 @@ def get_relations() -> Sequence[TypeRelation]:
         @series_not_empty
         @series_handle_nulls
         def contains_op(series: pd.Series, state: dict) -> bool:
-            is_valid_dtype = pdt.is_categorical_dtype(series) and not pdt.is_bool_dtype(
-                series
-            )
+            is_valid_dtype = isinstance(
+                series.dtype, pd.CategoricalDtype
+            ) and not pdt.is_bool_dtype(series)
             if is_valid_dtype:
                 return True
             return False

diff --git a/src/ydata_profiling/model/typeset_relations.py b/src/ydata_profiling/model/typeset_relations.py
@@ -31,7 +31,7 @@ def string_is_bool(series: pd.Series, state: dict, k: Dict[str, bool]) -> bool:
     def tester(s: pd.Series, state: dict) -> bool:
         return s.str.lower().isin(k.keys()).all()
 
-    if pdt.is_categorical_dtype(series):
+    if isinstance(series.dtype, pd.CategoricalDtype):
         return False
 
     return tester(series, state)