pleaes

app.py

@@ -19,6 +19,7 @@ CORS(app)
 
 # ============================================================================
 # FIX: Use raw strings or forward slashes for Windows paths
+<<<<<<< HEAD
 # ================================================
 # Base directory = where app.py lives
 BASE_DIR = os.path.dirname(os.path.abspath(__file__))
@@ -142,6 +143,133 @@ load_models()
 LAT_MIN, LAT_MAX = 12.70, 13.30
 LON_MIN, LON_MAX = 77.30, 78.00
 
+=======
+# ============================================================================
+
+BASE_DIR = os.path.dirname(os.path.abspath(__file__))
+DATASET_PATH = os.path.join(BASE_DIR, '..', 'model', 'dataset_cleaned.csv')
+MODEL_DIR = os.path.join(BASE_DIR, '..', 'model')
+
+# Convert to absolute paths and normalize
+DATASET_PATH = os.path.abspath(DATASET_PATH)
+MODEL_DIR = os.path.abspath(MODEL_DIR)
+
+print(f"\n📂 BASE_DIR: {BASE_DIR}")
+print(f"📊 DATASET_PATH: {DATASET_PATH}")
+print(f"📂 MODEL_DIR: {MODEL_DIR}")
+print(f"✓ Dataset exists: {os.path.exists(DATASET_PATH)}")
+print(f"✓ Model dir exists: {os.path.exists(MODEL_DIR)}\n")
+
+df = None
+model1 = None
+model2 = None
+le = None
+
+def load_dataset():
+    """Load the crime dataset"""
+    global df
+    try:
+        if not os.path.exists(DATASET_PATH):
+            print(f"❌ Dataset not found at: {DATASET_PATH}")
+            df = pd.DataFrame()
+            return False
+        
+        df = pd.read_csv(DATASET_PATH)
+        print(f"✅ Dataset loaded: {len(df)} records")
+        print(f"   Columns: {list(df.columns)}")
+        return True
+    except Exception as e:
+        print(f"❌ Error loading dataset: {e}")
+        traceback.print_exc()
+        df = pd.DataFrame()
+        return False
+
+def load_models():
+    """Load trained models with fallback options"""
+    global model1, model2, le
+    
+    print("\n📦 Attempting to load models...")
+    print(f"   Looking in: {MODEL_DIR}\n")
+    
+    model1_path = os.path.join(MODEL_DIR, 'model1.pkl')
+    model2_path = os.path.join(MODEL_DIR, 'model2.pkl')
+    le_path = os.path.join(MODEL_DIR, 'label_encoder.pkl')
+    
+    print(f"Model1 path: {model1_path} (exists: {os.path.exists(model1_path)})")
+    print(f"Model2 path: {model2_path} (exists: {os.path.exists(model2_path)})")
+    print(f"LE path: {le_path} (exists: {os.path.exists(le_path)})\n")
+    
+    # Try Model 1
+    if os.path.exists(model1_path):
+        try:
+            model1 = joblib.load(model1_path)
+            print(f"✅ Model1 loaded successfully")
+        except Exception as e:
+            print(f"⚠️  Joblib failed for model1: {e}")
+            try:
+                with open(model1_path, 'rb') as f:
+                    model1 = pickle.load(f)
+                print(f"✅ Model1 loaded with pickle")
+            except Exception as e2:
+                print(f"❌ Failed to load model1: {e2}")
+                model1 = None
+    else:
+        print(f"⚠️  Model1 not found")
+    
+    # Try Model 2
+    if os.path.exists(model2_path):
+        try:
+            model2 = joblib.load(model2_path)
+            print(f"✅ Model2 loaded successfully")
+        except Exception as e:
+            print(f"⚠️  Joblib failed for model2: {e}")
+            try:
+                with open(model2_path, 'rb') as f:
+                    model2 = pickle.load(f)
+                print(f"✅ Model2 loaded with pickle")
+            except Exception as e2:
+                print(f"❌ Failed to load model2: {e2}")
+                model2 = None
+    else:
+        print(f"⚠️  Model2 not found")
+    
+    # Try LabelEncoder
+    if os.path.exists(le_path):
+        try:
+            le = joblib.load(le_path)
+            print(f"✅ LabelEncoder loaded successfully")
+        except Exception as e:
+            print(f"⚠️  Joblib failed for LabelEncoder: {e}")
+            try:
+                with open(le_path, 'rb') as f:
+                    le = pickle.load(f)
+                print(f"✅ LabelEncoder loaded with pickle")
+            except Exception as e2:
+                print(f"❌ Failed to load LabelEncoder: {e2}")
+                le = None
+    else:
+        print(f"⚠️  LabelEncoder not found")
+    
+    if not model1 and not model2:
+        print("\n⚠️  Using MOCK predictions (models not available)")
+    else:
+        print("\n✅ Models ready for predictions")
+
+# Load on startup
+print("\n" + "="*60)
+print("🚀 OPENSIGHT API INITIALIZATION")
+print("="*60)
+load_dataset()
+load_models()
+
+# ============================================================================
+# BENGALURU CONFIGURATION
+# ============================================================================
+
+LAT_MIN, LAT_MAX = 12.70, 13.30
+LON_MIN, LON_MAX = 77.30, 78.00
+
+>>>>>>> f23833a (PUSHED EVRYTHING)
 def is_in_bengaluru(lat, lon):
     """Check if coordinates are within Bengaluru bounds"""
     return LAT_MIN <= lat <= LAT_MAX and LON_MIN <= lon <= LON_MAX
@@ -337,6 +465,10 @@ def predict_crime():
         
         print(f"🔮 Prediction request: location={location_name}, date={date_str}")
         
+<<<<<<< HEAD
+=======
+        # Resolve location name to coordinates
+>>>>>>> f23833a (PUSHED EVRYTHING)
         if location_name and not latitude:
             location_coords = {
                 'koramangala': (12.9352, 77.6245),
@@ -383,6 +515,7 @@ def predict_crime():
         crime_count = len(nearby_crimes)
         risk_level = calculate_risk_level(crime_count)
         
+<<<<<<< HEAD
         # Get crime types - check for CrimeHead_Name or CrimeType column
         crime_types = {}
         if len(nearby_crimes) > 0:
@@ -395,6 +528,12 @@ def predict_crime():
             elif 'CrimeHead_Name' in nearby_crimes.columns:
                 crime_types = nearby_crimes['CrimeHead_Name'].value_counts().head(10).to_dict()
                 print(f"   Found {len(crime_types)} crime types from CrimeHead_Name")
+=======
+        # Get crime types
+        crime_types = {}
+        if len(nearby_crimes) > 0 and 'CrimeType' in nearby_crimes.columns:
+            crime_types = nearby_crimes['CrimeType'].value_counts().head(5).to_dict()
+>>>>>>> f23833a (PUSHED EVRYTHING)
         
         # Calculate confidence
         confidence = min(95, 60 + (crime_count / max(len(df), 1) * 100))
@@ -459,10 +598,21 @@ def search_location():
         traceback.print_exc()
         return jsonify({'results': []})
 
+<<<<<<< HEAD
+@app.errorhandler(404)
+def not_found(e):
+    return jsonify({'error': 'Endpoint not found'}), 404
+
+=======
+# ============================================================================
+# ERROR HANDLERS
+# ============================================================================
+
 @app.errorhandler(404)
 def not_found(e):
     return jsonify({'error': 'Endpoint not found'}), 404
 
+>>>>>>> f23833a (PUSHED EVRYTHING)
 @app.errorhandler(500)
 def server_error(e):
     return jsonify({'error': 'Internal server error'}), 500

model/dataset_cleaned.csv

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:91ba2ca747fb549ef39c2e0d9eae2bf906491e953d7db0f72731ee5a8c953854
-size 42073036
+oid sha256:efbc3e8d31e67cc706d95f596e568ac2aa9ee1a0f8f623be43d0308ad3566630
+size 41960106

requirements.txt

@@ -1,3 +1,4 @@
+<<<<<<< HEAD
 Flask==2.3.0
 Flask-CORS==4.0.0
 pandas==2.0.0
@@ -6,3 +7,12 @@ scikit-learn==1.2.0
 joblib==1.2.0
 python-dotenv==1.0.0
 xgboost==1.7.6
+=======
+Flask==2.3.0
+Flask-CORS==4.0.0
+pandas==2.0.0
+numpy==1.23.0
+scikit-learn==1.2.0
+joblib==1.2.0
+python-dotenv==1.0.0
+>>>>>>> f23833a (PUSHED EVRYTHING)