ganeshkonapalli commited on
Commit
4fcdb47
·
verified ·
1 Parent(s): 84af4eb

Create config.py

Browse files
Files changed (1) hide show
  1. config.py +36 -0
config.py ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ import os
2
+
3
+ # --- Paths ---
4
+ BASE_DIR = os.path.abspath(os.path.dirname(__file__))
5
+
6
+ # Path to dataset
7
+ DATA_PATH = os.path.join(BASE_DIR, "data", "synthetic_transactions_samples_5000.csv")
8
+
9
+ # Directory to save models
10
+ MODEL_SAVE_DIR = os.path.join(BASE_DIR, "models")
11
+ os.makedirs(MODEL_SAVE_DIR, exist_ok=True)
12
+
13
+ # Save paths for Logistic Regression model + artifacts
14
+ MODEL_PATH = os.path.join(MODEL_SAVE_DIR, "logreg_model.pkl") # ✅ Logistic Regression model
15
+ TFIDF_VECTORIZER_PATH = os.path.join(MODEL_SAVE_DIR, "tfidf_vectorizer.pkl")
16
+ LABEL_ENCODERS_PATH = os.path.join(MODEL_SAVE_DIR, "label_encoders.pkl")
17
+
18
+ # --- Text & Label Columns ---
19
+ TEXT_COLUMN = "Sanction_Context"
20
+ LABEL_COLUMNS = [
21
+ "Red_Flag_Reason",
22
+ "Maker_Action",
23
+ "Escalation_Level",
24
+ "Risk_Category",
25
+ "Risk_Drivers",
26
+ "Investigation_Outcome"
27
+ ]
28
+
29
+ # --- TF-IDF Settings ---
30
+ TFIDF_MAX_FEATURES = 5000
31
+ NGRAM_RANGE = (1, 2)
32
+ USE_STOPWORDS = True # English stopwords will be removed if True
33
+
34
+ # --- Train/Test Split ---
35
+ RANDOM_STATE = 42
36
+ TEST_SIZE = 0.2