Fix all HTML entities and onnxruntime issues

Dockerfile

@@ -2,7 +2,7 @@ FROM python:3.11-slim
 
 WORKDIR /app
 
-# Install system dependencies (FIXED VERSION)
+# Install system dependencies (FIXED)
 RUN apt-get update && apt-get install -y \
     libglib2.0-0 \
     libsm6 \

app.py

@@ -13,6 +13,7 @@ import cv2
 import requests
 from typing import Optional, Dict, Tuple, Any
 from deepface import DeepFace
+from sklearn.metrics.pairwise import cosine_similarity
 import tensorflow as tf
 
 # Optimize TensorFlow for HuggingFace Spaces
@@ -60,14 +61,14 @@ try:
 except Exception as e:
     print(f"MongoDB connection error: {e}")
 
-# ---------------- Model Download Functions (Fixed) ----------------
+# ---------------- Model Download Functions (FIXED) ----------------
 
 def download_file_from_google_drive(file_id, destination):
     """Download file from Google Drive - Fixed version"""
     try:
         if not os.path.exists(destination):
             print(f"Downloading {destination}...")
-            # Fixed URL format without HTML entities
+            # FIXED URL format without HTML entities
             url = f"https://drive.google.com/uc?export=download&id={file_id}"
             
             os.makedirs(os.path.dirname(destination), exist_ok=True)
@@ -147,7 +148,7 @@ def setup_models():
 print("Setting up models...")
 yolo_available, antispoof_available = setup_models()
 
-# ---------------- YOLO Face Detection (Fixed HTML entities) ----------------
+# ---------------- YOLO Face Detection (FIXED) ----------------
 
 def _get_providers():
     available = ort.get_available_providers()
@@ -189,7 +190,7 @@ def _nms(boxes: np.ndarray, scores: np.ndarray, iou_threshold: float):
     areas = (x2 - x1) * (y2 - y1)
     order = scores.argsort()[::-1]
     keep = []
-    while order.size > 0:  # Fixed: removed HTML entity
+    while order.size > 0:  # FIXED: removed HTML entity
         i = int(order[0])
         keep.append(i)
         if order.size == 1:
@@ -202,7 +203,7 @@ def _nms(boxes: np.ndarray, scores: np.ndarray, iou_threshold: float):
         h = np.maximum(0.0, yy2 - yy1)
         inter = w * h
         iou = inter / (areas[i] + areas[order[1:]] - inter + 1e-6)
-        inds = np.where(iou <= iou_threshold)[0]  # Fixed: removed HTML entity
+        inds = np.where(iou <= iou_threshold)[0]  # FIXED: removed HTML entity
         order = order[inds + 1]
     return keep
 
@@ -222,7 +223,7 @@ class YoloV5FaceDetector:
             self.input_size = int(shape[2])
 
     @staticmethod
-    def _xywh2xyxy(x: np.ndarray) -> np.ndarray:  # Fixed: removed HTML entity
+    def _xywh2xyxy(x: np.ndarray) -> np.ndarray:  # FIXED: removed HTML entity
         y = np.zeros_like(x)
         y[:, 0] = x[:, 0] - x[:, 2] / 2
         y[:, 1] = x[:, 1] - x[:, 3] / 2
@@ -243,7 +244,7 @@ class YoloV5FaceDetector:
         if preds.ndim != 2:
             raise RuntimeError(f"Unexpected YOLO output shape: {preds.shape}")
         num_attrs = preds.shape[1]
-        has_landmarks = num_attrs >= 15  # Fixed: removed HTML entity
+        has_landmarks = num_attrs >= 15  # FIXED: removed HTML entity
         boxes_xywh = preds[:, 0:4]
         if has_landmarks:
             scores = preds[:, 4]
@@ -255,7 +256,7 @@ class YoloV5FaceDetector:
             else:
                 class_conf = cls_scores.max(axis=1, keepdims=True)
                 scores = (obj * class_conf).squeeze(-1)
-        keep = scores > self.conf_threshold  # Fixed: removed HTML entity
+        keep = scores > self.conf_threshold  # FIXED: removed HTML entity
         boxes_xywh = boxes_xywh[keep]
         scores = scores[keep]
         if boxes_xywh.shape[0] == 0:
@@ -269,16 +270,16 @@ class YoloV5FaceDetector:
         boxes_xyxy[:, 2] = np.clip(boxes_xyxy[:, 2], 0, w0 - 1)
         boxes_xyxy[:, 3] = np.clip(boxes_xyxy[:, 3], 0, h0 - 1)
         keep_inds = _nms(boxes_xyxy, scores, self.iou_threshold)
-        if len(keep_inds) > max_det:  # Fixed: removed HTML entity
+        if len(keep_inds) > max_det:  # FIXED: removed HTML entity
             keep_inds = keep_inds[:max_det]
         dets = []
         for i in keep_inds:
             dets.append({"bbox": boxes_xyxy[i].tolist(), "score": float(scores[i])})
         return dets
 
-# ---------------- Anti-Spoof Model (Fixed) ----------------
+# ---------------- Anti-Spoof Model (FIXED) ----------------
 
-def _sigmoid(x: np.ndarray) -> np.ndarray:  # Fixed: removed HTML entity
+def _sigmoid(x: np.ndarray) -> np.ndarray:  # FIXED: removed HTML entity
     return 1.0 / (1.0 + np.exp(-x))
 
 class AntiSpoofBinary:
@@ -297,7 +298,7 @@ class AntiSpoofBinary:
         self.input_name = self.session.get_inputs()[0].name
         self.output_names = [o.name for o in self.session.get_outputs()]
 
-    def _preprocess(self, face_bgr: np.ndarray) -> np.ndarray:  # Fixed: removed HTML entity
+    def _preprocess(self, face_bgr: np.ndarray) -> np.ndarray:  # FIXED: removed HTML entity
         img = cv2.resize(face_bgr, (self.input_size, self.input_size), interpolation=cv2.INTER_LINEAR)
         if self.rgb:
             img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
@@ -308,11 +309,11 @@ class AntiSpoofBinary:
         img = np.expand_dims(img, 0).astype(np.float32)
         return img
 
-    def predict_live_prob(self, face_bgr: np.ndarray) -> float:  # Fixed: removed HTML entity
+    def predict_live_prob(self, face_bgr: np.ndarray) -> float:  # FIXED: removed HTML entity
         inp = self._preprocess(face_bgr)
         outs = self.session.run(self.output_names, {self.input_name: inp})
         out = outs[0]
-        if out.ndim > 1:  # Fixed: removed HTML entity
+        if out.ndim > 1:  # FIXED: removed HTML entity
             out = np.squeeze(out, axis=0)
         if out.size == 2:
             vec = out.astype(np.float32)
@@ -348,7 +349,7 @@ def draw_live_overlay(img_bgr: np.ndarray, bbox, label: str, prob: float, color)
     cv2.rectangle(img_bgr, (x1, y_top), (x1 + tw + 6, y_top + th + 6), color, -1)
     cv2.putText(img_bgr, text, (x1 + 3, y_top + th), cv2.FONT_HERSHEY_SIMPLEX, 0.7, (0, 0, 0), 2, cv2.LINE_AA)
 
-def image_to_data_uri(img_bgr: np.ndarray) -> Optional[str]:  # Fixed: removed HTML entity
+def image_to_data_uri(img_bgr: np.ndarray) -> Optional[str]:  # FIXED: removed HTML entity
     success, buf = cv2.imencode(".jpg", img_bgr, [int(cv2.IMWRITE_JPEG_QUALITY), 85])
     if not success:
         return None
@@ -389,7 +390,7 @@ try:
 except Exception as e:
     print(f"Error loading anti-spoofing model: {e}")
 
-# ----------------------------- DeepFace Recognition (with timeout) -----------------------------
+# ----------------------------- DeepFace Recognition -----------------------------
 
 def get_face_features_deepface(image):
     """Extract face features using DeepFace with timeout protection"""
@@ -402,7 +403,7 @@ def get_face_features_deepface(image):
             enforce_detection=False
         )
         
-        if isinstance(embedding, list) and len(embedding) > 0:  # Fixed: removed HTML entity
+        if isinstance(embedding, list) and len(embedding) > 0:  # FIXED: removed HTML entity
             return np.array(embedding[0]['embedding'])
         else:
             return np.array(embedding['embedding']) if 'embedding' in embedding else None
@@ -440,8 +441,6 @@ def recognize_face_deepface(image, user_id, user_type='student'):
             return False, "No face detected in reference image"
 
         # Calculate cosine similarity
-        from sklearn.metrics.pairwise import cosine_similarity
-        
         similarity = cosine_similarity([features], [ref_features])[0][0]
         distance = 1 - similarity
         threshold = 0.4
@@ -450,7 +449,7 @@ def recognize_face_deepface(image, user_id, user_type='student'):
         inference_times.append(inference_time)
         total_attempts += 1
 
-        if distance < threshold:  # Fixed: removed HTML entity
+        if distance < threshold:  # FIXED: removed HTML entity
             correct_recognitions += 1
             return True, f"Face recognized (distance={distance:.3f}, similarity={similarity:.3f}, time={inference_time:.2f}s)"
         else:
@@ -463,7 +462,7 @@ def recognize_face_deepface(image, user_id, user_type='student'):
 def recognize_face(image, user_id, user_type='student'):
     return recognize_face_deepface(image, user_id, user_type)
 
-# ---------------------- Metrics helpers (same as before but fixed HTML entities) ----------------------
+# ---------------------- Metrics helpers ----------------------
 
 def log_metrics_event(event: dict):
     try:
@@ -505,7 +504,7 @@ def log_metrics_event_normalized(
     }
     log_metrics_event(doc)
 
-def classify_event(ev: Dict[str, Any]) -> Tuple[Optional[str], Optional[str]]:  # Fixed: removed HTML entity
+def classify_event(ev: Dict[str, Any]) -> Tuple[Optional[str], Optional[str]]:  # FIXED: removed HTML entity
     if ev.get("event"):
         e = ev.get("event")
         at = ev.get("attempt_type")
@@ -590,14 +589,14 @@ def compute_metrics(limit: int = 10000):
         }
     }
 
-def compute_latency_avg(limit: int = 300) -> Optional[float]:  # Fixed: removed HTML entity
+def compute_latency_avg(limit: int = 300) -> Optional[float]:  # FIXED: removed HTML entity
     cursor = metrics_events.find({"latency_ms": {"$exists": True}}, {"latency_ms": 1, "_id": 0}).sort("ts", -1).limit(limit)
     vals = [float(d["latency_ms"]) for d in cursor if isinstance(d.get("latency_ms"), (int, float))]
     if not vals:
         return None
     return sum(vals) / len(vals)
 
-# --------- ALL ROUTES (keeping all your existing routes) ---------
+# --------- ALL ROUTES ---------
 
 @app.route('/')
 def home():
@@ -699,7 +698,7 @@ def face_login():
             flash('Invalid role selected for face login.', 'danger')
             return redirect(url_for('login_page'))
 
-        users = collection.find({'face_image': {'$exists': True, '$ne': None}}).limit(20)  # Limit for performance
+        users = collection.find({'face_image': {'$exists': True, '$ne': None}}).limit(20)
         test_features = get_face_features_deepface(image)
         
         if test_features is None:
@@ -716,11 +715,10 @@ def face_login():
                 if ref_features is None:
                     continue
                 
-                from sklearn.metrics.pairwise import cosine_similarity
                 similarity = cosine_similarity([test_features], [ref_features])[0][0]
                 distance = 1 - similarity
                 
-                if distance < 0.4:  # Fixed: removed HTML entity
+                if distance < 0.4:  # FIXED: removed HTML entity
                     session['logged_in'] = True
                     session['user_type'] = face_role
                     session[id_field] = user[id_field]
@@ -762,7 +760,7 @@ def auto_face_login():
             id_field = 'student_id'
             dashboard_route = '/dashboard'
 
-        users = collection.find({'face_image': {'$exists': True, '$ne': None}}).limit(20)  # Limit for performance
+        users = collection.find({'face_image': {'$exists': True, '$ne': None}}).limit(20)
         for user in users:
             try:
                 ref_image_array = np.frombuffer(user['face_image'], np.uint8)
@@ -771,11 +769,10 @@ def auto_face_login():
                 if ref_features is None:
                     continue
                 
-                from sklearn.metrics.pairwise import cosine_similarity
                 similarity = cosine_similarity([test_features], [ref_features])[0][0]
                 distance = 1 - similarity
                 
-                if distance < 0.4:  # Fixed: removed HTML entity
+                if distance < 0.4:  # FIXED: removed HTML entity
                     session['logged_in'] = True
                     session['user_type'] = face_role
                     session[id_field] = user[id_field]
@@ -887,7 +884,7 @@ def mark_attendance():
     
     if anti_spoof_bin:
         live_prob = anti_spoof_bin.predict_live_prob(face_crop)
-        is_live = live_prob >= 0.7  # Fixed: removed HTML entity
+        is_live = live_prob >= 0.7  # FIXED: removed HTML entity
     
     label = "LIVE" if is_live else "SPOOF"
     color = (0, 200, 0) if is_live else (0, 0, 255)
@@ -1044,13 +1041,13 @@ def liveness_preview():
             live_prob = anti_spoof_bin.predict_live_prob(face_crop)
         
         threshold = 0.7
-        label = "LIVE" if live_prob >= threshold else "SPOOF"  # Fixed: removed HTML entity
+        label = "LIVE" if live_prob >= threshold else "SPOOF"  # FIXED: removed HTML entity
         color = (0, 200, 0) if label == "LIVE" else (0, 0, 255)
         draw_live_overlay(vis, (x1e, y1e, x2e, y2e), label, live_prob, color)
         overlay_data = image_to_data_uri(vis)
         return jsonify({
             'success': True,
-            'live': bool(live_prob >= threshold),  # Fixed: removed HTML entity
+            'live': bool(live_prob >= threshold),  # FIXED: removed HTML entity
             'live_prob': float(live_prob),
             'overlay': overlay_data
         })
@@ -1058,7 +1055,7 @@ def liveness_preview():
         print("liveness_preview error:", e)
         return jsonify({'success': False, 'message': 'Server error during preview'})
 
-# --------- TEACHER ROUTES (same as before) ---------
+# --------- TEACHER ROUTES ---------
 @app.route('/teacher_register.html')
 def teacher_register_page():
     return render_template('teacher_register.html')
@@ -1162,7 +1159,7 @@ def metrics_data():
             if r.get("decision") == "spoof_blocked":
                 r["liveness_pass"] = False
             elif isinstance(r.get("live_prob"), (int, float)):
-                r["liveness_pass"] = bool(r["live_prob"] >= 0.7)  # Fixed: removed HTML entity
+                r["liveness_pass"] = bool(r["live_prob"] >= 0.7)  # FIXED: removed HTML entity
             else:
                 r["liveness_pass"] = None
         normalized_recent.append(r)
@@ -1183,9 +1180,9 @@ def metrics_json():
     frr_pct = rates["FRR"] * 100.0
 
     return jsonify({
-        'Accuracy': f"{accuracy_pct:.2f}%" if totals["totalAttempts"] > 0 else "N/A",  # Fixed: removed HTML entity
-        'False Accepts (FAR)': f"{far_pct:.2f}%" if counts["impostorAttempts"] > 0 else "N/A",  # Fixed: removed HTML entity
-        'False Rejects (FRR)': f"{frr_pct:.2f}%" if counts["genuineAttempts"] > 0 else "N/A",  # Fixed: removed HTML entity
+        'Accuracy': f"{accuracy_pct:.2f}%" if totals["totalAttempts"] > 0 else "N/A",  # FIXED: removed HTML entity
+        'False Accepts (FAR)': f"{far_pct:.2f}%" if counts["impostorAttempts"] > 0 else "N/A",  # FIXED: removed HTML entity
+        'False Rejects (FRR)': f"{frr_pct:.2f}%" if counts["genuineAttempts"] > 0 else "N/A",  # FIXED: removed HTML entity
         'Average Inference Time (s)': f"{(avg_latency/1000.0):.2f}" if isinstance(avg_latency, (int, float)) else "N/A",
         'Correct Recognitions': counts["trueAccepts"],
         'Total Attempts': totals["totalAttempts"],

requirements.txt

@@ -3,7 +3,7 @@ pymongo==4.5.0
 python-dotenv==1.0.0
 opencv-python-headless==4.8.1.78
 numpy==1.24.3
-onnxruntime==1.15.1
+onnxruntime-cpu==1.15.1
 requests==2.31.0
 deepface==0.0.79
 tensorflow==2.13.0