Update config.py

config.py

@@ -2,14 +2,20 @@ import os
 
 # --- Paths ---
 BASE_DIR = os.path.abspath(os.path.dirname(__file__))
+
+# Path to dataset
 DATA_PATH = os.path.join(BASE_DIR, "data", "synthetic_transactions_samples_5000.csv")
 
+# Directory to save models
 MODEL_SAVE_DIR = os.path.join(BASE_DIR, "models")
-LABEL_ENCODERS_PATH = os.path.join(MODEL_SAVE_DIR, "label_encoders.pkl")
+os.makedirs(MODEL_SAVE_DIR, exist_ok=True)
+
+# Save paths for Logistic Regression model + artifacts
+MODEL_PATH = os.path.join(MODEL_SAVE_DIR, "logreg_model.pkl")              # ✅ Logistic Regression model
 TFIDF_VECTORIZER_PATH = os.path.join(MODEL_SAVE_DIR, "tfidf_vectorizer.pkl")
-MODEL_PATH = os.path.join(MODEL_SAVE_DIR, "lgbm_models.pkl")  # <-- LGBM model file
+LABEL_ENCODERS_PATH = os.path.join(MODEL_SAVE_DIR, "label_encoders.pkl")
 
-# --- Columns ---
+# --- Text & Label Columns ---
 TEXT_COLUMN = "Sanction_Context"
 LABEL_COLUMNS = [
     "Red_Flag_Reason",
@@ -23,8 +29,8 @@ LABEL_COLUMNS = [
 # --- TF-IDF Settings ---
 TFIDF_MAX_FEATURES = 5000
 NGRAM_RANGE = (1, 2)
-USE_STOPWORDS = True
+USE_STOPWORDS = True  # English stopwords will be removed if True
 
 # --- Train/Test Split ---
 RANDOM_STATE = 42
-TEST_SIZE = 0.2
+TEST_SIZE = 0.2