changes wording of datetime transformer features (#355)

Author: solegalli

feature_engine/datetime/datetime.py

@@ -39,11 +39,10 @@ class DatetimeFeatures(BaseEstimator, TransformerMixin):
     - "quarter"
     - "semester"
     - "year"
-    - "week_of_the_month"
-    - "week_of_the_year"
-    - "day_of_the_week"
-    - "day_of_the_month"
-    - "day_of_the_year"
+    - "week"
+    - "day_of_week"
+    - "day_of_month"
+    - "day_of_year"
     - "weekend"
     - "month_start"
     - "month_end"
@@ -68,7 +67,7 @@ class DatetimeFeatures(BaseEstimator, TransformerMixin):
 
     features_to_extract: list, default=None
         The list of date features to extract. If None, the following features will be
-        extracted: "month", "year", "day_of_the_week", "day_of_the_month", "hour",
+        extracted: "month", "year", "day_of_week", "day_of_month", "hour",
         "minute" and "second". If "all", all supported features will be extracted.
         Alternatively, you can pass a list with the names of the supported features
         you want to extract.

feature_engine/datetime/datetime_constants.py

@@ -5,10 +5,10 @@
     "quarter",
     "semester",
     "year",
-    "week_of_the_year",
-    "day_of_the_week",
-    "day_of_the_month",
-    "day_of_the_year",
+    "week",
+    "day_of_week",
+    "day_of_month",
+    "day_of_year",
     "weekend",
     "month_start",
     "month_end",
@@ -28,10 +28,10 @@
     "quarter": "_quarter",
     "semester": "_semester",
     "year": "_year",
-    "week_of_the_year": "_woty",
-    "day_of_the_week": "_dotw",
-    "day_of_the_month": "_dotm",
-    "day_of_the_year": "_doty",
+    "week": "_week",
+    "day_of_week": "_day_of_week",
+    "day_of_month": "_day_of_month",
+    "day_of_year": "_day_of_year",
     "weekend": "_weekend",
     "month_start": "_month_start",
     "month_end": "_month_end",
@@ -51,10 +51,10 @@
     "quarter": lambda x: x.dt.quarter,
     "semester": lambda x: np.where(x.dt.month <= 6, 1, 2).astype(np.int64),
     "year": lambda x: x.dt.year,
-    "week_of_the_year": lambda x: x.dt.isocalendar().week.astype(np.int64),
-    "day_of_the_week": lambda x: x.dt.dayofweek,
-    "day_of_the_month": lambda x: x.dt.day,
-    "day_of_the_year": lambda x: x.dt.dayofyear,
+    "week": lambda x: x.dt.isocalendar().week.astype(np.int64),
+    "day_of_week": lambda x: x.dt.dayofweek,
+    "day_of_month": lambda x: x.dt.day,
+    "day_of_year": lambda x: x.dt.dayofyear,
     "weekend": lambda x: np.where(x.dt.dayofweek <= 4, 0, 1).astype(np.int64),
     "month_start": lambda x: x.dt.is_month_start.astype(np.int64),
     "month_end": lambda x: x.dt.is_month_end.astype(np.int64),
@@ -72,8 +72,8 @@
 FEATURES_DEFAULT = [
     "month",
     "year",
-    "day_of_the_week",
-    "day_of_the_month",
+    "day_of_week",
+    "day_of_month",
     "hour",
     "minute",
     "second",

tests/test_datetime/conftest.py

@@ -10,10 +10,10 @@ def df_datetime_transformed(df_datetime):
         "datetime_range_quarter": [1, 1, 1, 1],
         "datetime_range_semester": [1, 1, 1, 1],
         "datetime_range_year": [2020, 2020, 2020, 2020],
-        "datetime_range_woty": [9, 9, 9, 9],
-        "datetime_range_dotw": [0, 1, 2, 3],
-        "datetime_range_dotm": [24, 25, 26, 27],
-        "datetime_range_doty": [55, 56, 57, 58],
+        "datetime_range_week": [9, 9, 9, 9],
+        "datetime_range_day_of_week": [0, 1, 2, 3],
+        "datetime_range_day_of_month": [24, 25, 26, 27],
+        "datetime_range_day_of_year": [55, 56, 57, 58],
         "datetime_range_weekend": [0, 0, 0, 0],
         "datetime_range_month_start": [0, 0, 0, 0],
         "datetime_range_month_end": [0, 0, 0, 0],
@@ -30,10 +30,10 @@ def df_datetime_transformed(df_datetime):
         "date_obj1_quarter": [1, 1, 2, 2],
         "date_obj1_semester": [1, 1, 1, 1],
         "date_obj1_year": [2010, 1945, 2100, 1999],
-        "date_obj1_woty": [53, 8, 24, 20],
-        "date_obj1_dotw": [4, 5, 0, 0],
-        "date_obj1_dotm": [1, 24, 14, 17],
-        "date_obj1_doty": [1, 55, 165, 137],
+        "date_obj1_week": [53, 8, 24, 20],
+        "date_obj1_day_of_week": [4, 5, 0, 0],
+        "date_obj1_day_of_month": [1, 24, 14, 17],
+        "date_obj1_day_of_year": [1, 55, 165, 137],
         "date_obj1_weekend": [0, 1, 0, 0],
         "date_obj1_month_start": [1, 0, 0, 0],
         "date_obj1_month_end": [0, 0, 0, 0],
@@ -50,10 +50,10 @@ def df_datetime_transformed(df_datetime):
         "date_obj2_quarter": [4, 4, 2, 1],
         "date_obj2_semester": [2, 2, 1, 1],
         "date_obj2_year": [2012, 2009, 1995, 2004],
-        "date_obj2_woty": [41, 53, 26, 12],
-        "date_obj2_dotw": [3, 3, 4, 2],
-        "date_obj2_dotm": [11, 31, 30, 17],
-        "date_obj2_doty": [285, 365, 181, 77],
+        "date_obj2_week": [41, 53, 26, 12],
+        "date_obj2_day_of_week": [3, 3, 4, 2],
+        "date_obj2_day_of_month": [11, 31, 30, 17],
+        "date_obj2_day_of_year": [285, 365, 181, 77],
         "date_obj2_weekend": [0, 0, 0, 0],
         "date_obj2_month_start": [0, 0, 0, 0],
         "date_obj2_month_end": [0, 1, 1, 0],
@@ -70,10 +70,10 @@ def df_datetime_transformed(df_datetime):
         "time_obj_quarter": [today.quarter] * 4,
         "time_obj_semester": [1 if today.month <= 6 else 2] * 4,
         "time_obj_year": [today.year] * 4,
-        "time_obj_woty": [today.week] * 4,
-        "time_obj_dotw": [today.dayofweek] * 4,
-        "time_obj_dotm": [today.day] * 4,
-        "time_obj_doty": [today.dayofyear] * 4,
+        "time_obj_week": [today.week] * 4,
+        "time_obj_day_of_week": [today.dayofweek] * 4,
+        "time_obj_day_of_month": [today.day] * 4,
+        "time_obj_day_of_year": [today.dayofyear] * 4,
         "time_obj_weekend": [1 if today.dayofweek > 4 else 0] * 4,
         "time_obj_month_start": [int(today.is_month_start)] * 4,
         "time_obj_month_end": [int(today.is_month_end)] * 4,

tests/test_datetime/test_datetime_features.py

@@ -226,25 +226,25 @@ def test_extract_all_datetime_features(df_datetime, df_datetime_transformed):
 
 def test_extract_specified_datetime_features(df_datetime, df_datetime_transformed):
     X = DatetimeFeatures(
-        features_to_extract=["semester", "week_of_the_year"]
+        features_to_extract=["semester", "week"]
     ).fit_transform(df_datetime)
     pd.testing.assert_frame_equal(
         X,
         df_datetime_transformed[
             vars_non_dt
-            + [var + "_" + feat for var in vars_dt for feat in ["semester", "woty"]]
+            + [var + "_" + feat for var in vars_dt for feat in ["semester", "week"]]
         ],
     )
 
     # different order than they appear in the glossary
-    X = DatetimeFeatures(features_to_extract=["hour", "day_of_the_week"]).fit_transform(
+    X = DatetimeFeatures(features_to_extract=["hour", "day_of_week"]).fit_transform(
         df_datetime
     )
     pd.testing.assert_frame_equal(
         X,
         df_datetime_transformed[
             vars_non_dt
-            + [var + "_" + feat for var in vars_dt for feat in ["hour", "dotw"]]
+            + [var + "_" + feat for var in vars_dt for feat in ["hour", "day_of_week"]]
         ],
     )
 
@@ -343,7 +343,7 @@ def test_extract_features_without_dropping_original_variables(
 ):
     X = DatetimeFeatures(
         variables=["datetime_range", "date_obj2"],
-        features_to_extract=["week_of_the_year", "quarter"],
+        features_to_extract=["week", "quarter"],
         drop_original=False,
     ).fit_transform(df_datetime)
 
@@ -357,7 +357,7 @@ def test_extract_features_without_dropping_original_variables(
                 for feat in [
                     var + "_" + feat
                     for var in ["datetime_range", "date_obj2"]
-                    for feat in ["woty", "quarter"]
+                    for feat in ["week", "quarter"]
                 ]
             ],
             axis=1,
@@ -377,11 +377,11 @@ def test_extract_features_from_variables_containing_nans():
 
 def test_extract_features_with_different_datetime_parsing_options(df_datetime):
     X = DatetimeFeatures(
-        features_to_extract=["day_of_the_month"], dayfirst=True
+        features_to_extract=["day_of_month"], dayfirst=True
     ).fit_transform(df_datetime[["date_obj2"]])
     pd.testing.assert_frame_equal(
         X,
-        pd.DataFrame({"date_obj2_dotm": [10, 31, 30, 17]}),
+        pd.DataFrame({"date_obj2_day_of_month": [10, 31, 30, 17]}),
     )
 
     X = DatetimeFeatures(features_to_extract=["year"], yearfirst=True).fit_transform(