Creating Linear Models Maths Presentation in Colourful Illustrative Style

LINEAR ALGEBRA A1 TUTOR POWERED BY GROQ API, GRADIO Ul & HUGGING FACE SPACES.

INTRODUCTION • A1-powered educational tool for learning Linear Algebra Based on Elementary Linear Algebra Applications Version (Anton, Rorres, Kaul, 2019) Fast, high-quality responses using Groq LLMs Clean, interactive frontend built with Gradio Deployed on Hugging Face Spaces.

GOAL OF THE APP Explains linear algebra concepts using Llama 3.1 (Groq APD • Solves exercises with step-by-step reasoning Extracts text from PDF or images (OCR) • Visualizes vectors, matrix transformations, and eigenvectors Exports explanations as beautifully formatted PDF files Maintains conversation memory for follow-up questions.

ARCHITECTURE OVERVIEW Frontend: Gradio - with interactive tabs, custom layouts, and enhanced styling Backend: Python integrated with Groq API for high-performance computations LLM Model: Mixtral-8x7b-32768, optimized for text understanding and generation File Processing: PyPDF2 for PDF handling, pytesseract for OCR of images Deployment: Hosted on Hugging Face Spaces for easy sharing and access API Key Management: Secure storage using environment variables.

SYSTEM PROMPT DESIGN The LLM is guided bu a strict system prompt to ensure: • Textbook-aligned explanations for academic rigor • Geometric interpretations for visual understanding • Clear, intuitive examples for easier comprehension • Step-by-step solutions for methodical learning • Multidisciplinary applications connecting concepts across fields Outcome: Ensures consistent, accurate, and academically reliable responses..

TOPICS User Input: A Linear Algebra concept (or any mathematical topic) LLM Output Includes: Definition — Clear, textbook-aligned explanation Geometric Interpretation - Visual or spatial understanding 3-5 Multidisciplinary Applications — Real-world or cross-field relevance Step-by-step Example - Easy to follow, illustrative problem Known Facts / Properties — Important points and commonly used results Formatting: Neatly organized with Markdown headings for readability.

SOLUTIONS TO EXERCISES • User Inputs: • Typed exercise in the text box • OR image/ PDF upload (processed via PyPDF2 for PDFs, putesser act OCR for images) • LLM OUtPUt Includes: • Complete step-by-step solution • Interpretation of any referenced diagrams or figures • Methods and approaches aligned with Anton & Rorres textbook • Output formatted neatly for readability.

APPLICATIONS User Input: • A real-world domain (typed or selected) LLM Output Includes: Relevant linear algebra concepts used in that domain Clear explanations of each concept Concrete examples drawn from the real-world field • Short worked example demonstrating application Output is formatted neatly for easy understanding.

A pp INTERFACE utor near ge ra Corr•pe X to a • mth is m Wutims Giwn it Let 2 2 . • Antm WRh up is • the R by v. is a of v.

ERROR HANDLING & ROBUSTNESS • Try/Except for file extraction ensures smooth processing Provides clear, user-friendly messages for errors Handles multiple input types: PDFs, images, empty inputs Catches Groq API communication errors Prevents app crashes in production for a reliable user experience.

Ul DESIGN Frontend Design & LatJOUt • Built using gr.BIocks() for flexible Ul structure • Modern theme: dark blue + light grey for professional look • Clean card-based layout for clarity and fOCUS • Markdown used to structure content neatly • Tabs provide simple and intuitive navigation.

HOW TO GET YOUR GROQ API KEY • Go to Groq Console -+ console.groq.com • Sign in with your account • Navigate to API Keys • Click Create API Key • Copu the keu (starts with gsk_...) • Do not share your keg publicly..

STORING THE GROQ API KEY AS A SECRET Environment variables (safe for Linux/CoIab) • Secrets Manager (Hugging Face Spaces) • Hidden form field in Gradio Ul (for private apps) Prevent unauthorized API usage Protect your building and account Never put API keys directly in code..

USING GROQ API KEY IN GOOGLE COLAB Step 1 — Set the key securely Python import os " your_key_here" KEY ] = " Step 2— Use it in your PutTnon script Python from groq import Groq client = Groq(api_key=os.environ["GROQ_APl_KEY"]) Step 3— Instead of hardcoding the key, Use: Python google.colab import userdata os.environ["GROQ_APl_KEY"]=userdata.get('GROQ_APl KEY').

HOW TO STORE SECRETS IN GOOGLE COLAB Option A — Using Colab "secrets" sidebar Open Secrets • Add new secret Retrieve in code: Python from google.colab import userdata key = userdata.get('GROQ API KEY') Option B — Manual Input (safe Tor sfiaring notebooks) Python import getpass os.environ["GROQ_APl_KEY"] =getpass.getpass("Enter Groq API Key:").

USING GROQ API KEY IN HUGGING FACE SPACES Storing the key: • Go to your Space • Click Settings + Variables and Secrets • Add: Name: Value: your key Using the secret inside app script: Python i import os from groq import Groq client = Groq(api_key=os.environ["GROQ_API_KEY"]) • Secrets remain hidden and secure..

SUMMARY(API KEY) • Groq API Key -+ Get from Groq Console • Keep it secret -+ Use environment variables • Colab Store via userdata or input prompt • Hugging Face -+ Store using "Variables & Secrets" • Usage is simple with.

DEPLOYMENT ON HUGGING FACE SPACES • Deployment on Hugging Face Spaces • Create a new Space -+ select Gradio as the SDK • Upload project files: app.py and requirements.txt • Set environment variable: Add GROQ API KEY in Space —5 Settings Secrets • Automatic build: Hugging Face builds and launches the app • Access & share: Use the public link to share the app.

FUTURE ENHANCEMENTS • LaTeX Math Rendering Display formulas and equations clearly • 2D/3D Visualization Tools -+ Graphs, vectors, matrices, and geometric interpretations • User Progress Tracking * Monitor exercises completed, scores, and improvements • Model Fine-Tuning -+ Adapt LLM to textbook- aligned solutions for accuracy • Quiz / Exam Mode * Test understanding with automated exercises and solutions.

CONCLUSION A1 Tutor Powered by Reliable Textbook Logic Outputs follow structured, academically aligned reasoning. High-Speed Groq LLM Responses • Mixtral model provides fast, accurate, and well-formatted explanations. Clean, Intuitive Gradio Interface Modern layout using gr.Blocks(), tabs, cards, and Markdown. Ideal for Students, Educators & Researchers Supports problem solving, conceptual explanations, domain applications, and OCR-based input..

PRESENTED BY: Team Leader: Ghulam Murtaza Team Member 1: Mujahid Abbas Team Member 2: Shagufta Abbas Team Member 3: Laiba Saqlain Team Member 4: Saima Noureen.

01b.