fixed :fixed the testing error

features/nepali_text_classifier/model_loader.py

@@ -8,6 +8,7 @@ from pathlib import Path
 import numpy as np
 import pandas as pd
 from huggingface_hub import snapshot_download
+from sklearn.linear_model import LogisticRegression, LogisticRegressionCV
 
 from config import Config
 
@@ -45,6 +46,13 @@ DEFAULT_MODEL_RANKING = [
 ]
 
 
+def _patch_legacy_logistic_model(model):
+    """Backfill attributes expected by newer sklearn versions."""
+    if isinstance(model, (LogisticRegression, LogisticRegressionCV)) and not hasattr(model, "multi_class"):
+        model.multi_class = "auto"
+    return model
+
+
 class NepaliRichFeatures:
     """Burstiness + stylometry feature extractor used during model training."""
 
@@ -176,7 +184,7 @@ def load_artifacts():
             unavailable[model_name] = "Missing model file"
             continue
         with open(file_path, "rb") as fp:
-            models[model_name] = pickle.load(fp)
+            models[model_name] = _patch_legacy_logistic_model(pickle.load(fp))
 
     with open(model_dir / "word_vectorizer.pkl", "rb") as fp:
         word_vectorizer = pickle.load(fp)

features/text_classifier/model_loader.py

@@ -6,6 +6,7 @@ from pathlib import Path
 
 import torch
 from huggingface_hub import snapshot_download
+from sklearn.linear_model import LogisticRegression, LogisticRegressionCV
 
 from config import Config
 
@@ -26,6 +27,13 @@ REQUIRED_FILES = (
 )
 
 
+def _patch_legacy_logistic_model(model):
+    """Backfill attributes expected by newer sklearn versions."""
+    if isinstance(model, (LogisticRegression, LogisticRegressionCV)) and not hasattr(model, "multi_class"):
+        model.multi_class = "auto"
+    return model
+
+
 def _has_required_artifacts(model_dir: Path) -> bool:
     if not model_dir.exists() or not model_dir.is_dir():
         return False
@@ -79,6 +87,7 @@ def load_model():
 
     with open(artifact_dir / "classifier.pkl", "rb") as f:
         loaded_classifier = pickle.load(f)
+    loaded_classifier = _patch_legacy_logistic_model(loaded_classifier)
 
     with open(artifact_dir / "scaler.pkl", "rb") as f:
         loaded_scaler = pickle.load(f)