Fix univariate plots

README.md

@@ -148,7 +148,6 @@ train_data = feat_type_detector.fit_transform_feature_types(train_data)
 # show univariate plots
 univariate_plots(
         train_data.loc[:, feat_type_detector.num_columns],  # here the target column EC1 is already included
-        "EC1",
     )
 
 # show bi-variate plots

bluecast/eda/analyse.py

@@ -12,20 +12,14 @@
 from sklearn.manifold import TSNE
 
 
-def univariate_plots(df: pd.DataFrame, target: str) -> None:
+def univariate_plots(df: pd.DataFrame) -> None:
     """
-    Plots univariate plots for all the columns in the dataframe.
-    The target column must be part of the provided DataFrame.
+    Plots univariate plots for all the columns in the dataframe. Only numerical columns are expected.
+    The target column does not need to be part of the provided DataFrame.
 
     Expects numeric columns only.
     """
-    if target not in df.columns.to_list():
-        raise ValueError("Target column must be part of the provided DataFrame")
     for col in df.columns:
-        # Check if the col is the target column (EC1 or EC2)
-        if col == target:
-            continue  # Skip target columns in univariate analysis
-
         plt.figure(figsize=(8, 4))
 
         # Histogram

bluecast/tests/test_analyse.py

@@ -80,8 +80,7 @@ def test_univariate_plots(synthetic_train_test_data):
     univariate_plots(
         synthetic_train_test_data[0].loc[
             :, ["numerical_feature_1", "numerical_feature_2", "numerical_feature_3"]
-        ],
-        "target",
+        ]
     )
     assert True
 

docs/source/index.md

@@ -148,7 +148,6 @@ train_data = feat_type_detector.fit_transform_feature_types(train_data)
 # show univariate plots
 univariate_plots(
         train_data.loc[:, feat_type_detector.num_columns],  # here the target column EC1 is already included
-        "EC1",
     )
 
 # show bi-variate plots