Submit project for Team T039_BlueScreeners

submissions/T039_BlueScreeners/ai_services.py

@@ -0,0 +1,61 @@
+# ai_services.py
+import joblib
+
+# --- Placeholder functions for your ML models ---
+# You will replace these with the actual code from your model files.
+
+def load_prediction_model(path='ml_models/trained_model.pkl'):
+    """Loads the trained risk prediction model."""
+    try:
+        model = joblib.load(path)
+        print("Prediction model loaded successfully.")
+        return model
+    except FileNotFoundError:
+        print("Prediction model file not found. Please check the path.")
+        return None
+
+def get_risk_prediction(model, student_data):
+    """
+    Takes the model and student data to return a risk prediction.
+    - student_data should be a list or array in the format your model expects,
+      e.g., [attendance_percentage, avg_assignment_score, avg_quiz_score]
+    """
+    if model is None:
+        return "Model not loaded", 0.0
+
+    # This is an example; you'll need to format the data exactly as your model was trained
+    # prediction = model.predict([student_data])
+    # probability = model.predict_proba([student_data])
+
+    # Returning mock data for now
+    risk_level = "High" if student_data[0] < 75 else "Low"
+    risk_score = 0.85 if risk_level == "High" else 0.25
+
+    return risk_level, risk_score
+
+def generate_quiz_questions(topic, level='hard'):
+    """
+    Calls your quiz_generator_v3.py logic.
+    """
+    # Import or call your quiz generator function here
+    print(f"Generating {level} quiz for topic: {topic}")
+    # Returning mock data
+    return [
+        {
+            "question": f"What is the capital of {topic}?",
+            "options": ["Option A", "Option B", "Option C", "Correct Answer"],
+            "correct_answer": "Correct Answer"
+        }
+    ]
+
+def get_chatbot_response(user_query, history=None):
+    """
+    Calls your chatbot_v2.py logic.
+    """
+    # Import or call your chatbot function here
+    print(f"Chatbot processing query: {user_query}")
+    # Returning a simple response
+    if "schedule" in user_query.lower():
+        return "Your class schedule can be found on the student portal under 'My Schedule'."
+    else:
+        return "I am a helpful assistant. How can I help you with your academic questions today?"

submissions/T039_BlueScreeners/chatbot_v2.py

@@ -0,0 +1,46 @@
+from transformers import pipeline
+
+def start_chat_v2():
+    """
+    Initializes and runs a conversational chatbot.
+    This version manually manages the chat history to avoid import issues.
+    """
+    print("🤖 Initializing chatbot... This may take a moment.")
+    chatbot = pipeline("conversational", model="microsoft/DialoGPT-medium")
+    print("✅ Chatbot is ready! Type 'exit' to end the conversation.")
+    print("-" * 30)
+
+    # We will use simple lists to keep track of the conversation history
+    chat_history = {
+        "past_user_inputs": [],
+        "generated_responses": []
+    }
+
+    while True:
+        user_input = input(">> You: ")
+
+        if user_input.lower() in ["exit", "quit"]:
+            print("👋 Goodbye!")
+            break
+
+        # Manually build the conversation object the pipeline expects
+        conversation = {
+            "past_user_inputs": chat_history["past_user_inputs"],
+            "generated_responses": chat_history["generated_responses"],
+            "text": user_input
+        }
+
+        # Get the response
+        result = chatbot(conversation)
+
+        # Print the latest response from the bot
+        bot_response = result["generated_responses"][-1]
+        print(f">> Bot: {bot_response}")
+
+        # Update our history with the latest turn
+        chat_history["past_user_inputs"].append(user_input)
+        chat_history["generated_responses"].append(bot_response)
+
+# --- Start the Chatbot ---
+if __name__ == "__main__":
+    start_chat_v2()

submissions/T039_BlueScreeners/encode_faces.py

@@ -0,0 +1,45 @@
+# encode_faces.py (Bypassing OpenCV)
+import face_recognition
+import pickle
+import os
+
+# Path to the directory where student images are stored
+folderPath = 'student_images'
+pathList = os.listdir(folderPath)
+print(f"Found files: {pathList}")
+
+encodeListKnown = []
+studentUsernames = []
+
+print("\nStarting encoding process using the direct loader...")
+for path in pathList:
+    username = os.path.splitext(path)[0]
+    try:
+        # Load the image file directly using face_recognition's built-in loader
+        # This function handles all the necessary conversions automatically
+        print(f"Processing image for '{username}'...")
+        image = face_recognition.load_image_file(os.path.join(folderPath, path))
+
+        # Get face encodings from the image
+        encodings = face_recognition.face_encodings(image)
+
+        if encodings:
+            # Use the first face found in the image
+            encodeListKnown.append(encodings[0])
+            studentUsernames.append(username)
+            print(f"--> Successfully encoded face for: '{username}'")
+        else:
+            print(f"Warning: No face was found in the image for '{username}'. Skipping.")
+
+    except Exception as e:
+        print(f"!!! An unexpected error occurred while processing '{username}': {e}")
+
+# Save the data only if at least one face was successfully encoded
+if encodeListKnown:
+    encodeData = [studentUsernames, encodeListKnown]
+
+    with open('EncodeFile.p', 'wb') as file:
+        pickle.dump(encodeData, file)
+    print("\nEncoding Complete. Face data saved to EncodeFile.p")
+else:
+    print("\nEncoding failed. No faces were successfully encoded.")

submissions/T039_BlueScreeners/import_data.py

@@ -0,0 +1,39 @@
+# import_data.py
+import csv
+from server import app, db, User, Student, Teacher # Imports from the configured server file
+from werkzeug.security import generate_password_hash
+
+def import_csv_data():
+    with app.app_context():
+        print("Dropping all tables...")
+        db.drop_all()
+        print("Creating all tables...")
+        db.create_all()
+
+        with open('student_data.csv', newline='') as csvfile:
+            reader = csv.DictReader(csvfile)
+            print("Starting data import...")
+            for row in reader:
+                if User.query.filter_by(username=row['username']).first():
+                    print(f"User '{row['username']}' already exists. Skipping.")
+                    continue
+
+                hashed_password = generate_password_hash(row['password'])
+                new_user = User(username=row['username'], password_hash=hashed_password, role=row['role'])
+                db.session.add(new_user)
+                db.session.commit()
+
+                if new_user.role == 'student':
+                    new_profile = Student(full_name=row['full_name'], user_id=new_user.id)
+                    db.session.add(new_profile)
+                elif new_user.role == 'teacher':
+                    new_profile = Teacher(full_name=row['full_name'], user_id=new_user.id)
+                    db.session.add(new_profile)
+
+                print(f"Added user: {new_user.username} (Role: {new_user.role})")
+
+        db.session.commit()
+        print("\nData import complete!")
+
+if __name__ == '__main__':
+    import_csv_data()

submissions/T039_BlueScreeners/ml_models/chatbot_v2.py

@@ -0,0 +1,46 @@
+from transformers import pipeline
+
+def start_chat_v2():
+    """
+    Initializes and runs a conversational chatbot.
+    This version manually manages the chat history to avoid import issues.
+    """
+    print("🤖 Initializing chatbot... This may take a moment.")
+    chatbot = pipeline("conversational", model="microsoft/DialoGPT-medium")
+    print("✅ Chatbot is ready! Type 'exit' to end the conversation.")
+    print("-" * 30)
+
+    # We will use simple lists to keep track of the conversation history
+    chat_history = {
+        "past_user_inputs": [],
+        "generated_responses": []
+    }
+
+    while True:
+        user_input = input(">> You: ")
+
+        if user_input.lower() in ["exit", "quit"]:
+            print("👋 Goodbye!")
+            break
+
+        # Manually build the conversation object the pipeline expects
+        conversation = {
+            "past_user_inputs": chat_history["past_user_inputs"],
+            "generated_responses": chat_history["generated_responses"],
+            "text": user_input
+        }
+
+        # Get the response
+        result = chatbot(conversation)
+
+        # Print the latest response from the bot
+        bot_response = result["generated_responses"][-1]
+        print(f">> Bot: {bot_response}")
+
+        # Update our history with the latest turn
+        chat_history["past_user_inputs"].append(user_input)
+        chat_history["generated_responses"].append(bot_response)
+
+# --- Start the Chatbot ---
+if __name__ == "__main__":
+    start_chat_v2()

submissions/T039_BlueScreeners/ml_models/chatbot_v2_web.py

@@ -0,0 +1,27 @@
+from transformers import pipeline
+
+class WebChatbot:
+    def __init__(self):
+        """
+        Initializes the text-generation chatbot pipeline and chat history.
+        """
+        self.chatbot = pipeline("text-generation", model="microsoft/DialoGPT-medium")
+        self.chat_history = []
+
+    def get_response(self, user_input):
+        """
+        Accepts user input string, returns bot response string and updates chat history.
+        """
+        # Append user input to chat history
+        self.chat_history.append(user_input)
+
+        # Generate response using text-generation pipeline
+        # Join chat history as context
+        context = " ".join(self.chat_history[-5:])  # last 5 messages as context
+        outputs = self.chatbot(context, max_length=100, num_return_sequences=1)
+        bot_response = outputs[0]['generated_text'][len(context):].strip()
+
+        # Append bot response to chat history
+        self.chat_history.append(bot_response)
+
+        return bot_response, self.chat_history