Deep learning using Tensorflow Lite on Raspberry Pi

Dive into the world of Embedded Deep Learning with Python! This hands-on course leverages the power of the Raspberry Pi 4 to build practical AI projects using custom datasets. You'll gain expertise in deploying deep learning models on edge devices and optimizing them for performance.

Here's a glimpse of what you'll learn:

• Trigonometric Function Approximation: Start with the fundamentals by approximating trigonometric functions like the Sin function using generated random data and model building.
• Visual Calculator: Develop a computer vision project that takes images as input, interprets them as equations, and calculates the result. This utilizes convolutional neural networks (CNNs) for categorical classification.
• Custom Voice Controlled LEDs: Explore audio processing and CNNs by building a system that controls multiple LEDs with your voice, incorporating basic electronics.
• Post-Quantization for Model Optimization: A unique learning experience! Learn to reduce the size of TensorFlow models (trained on Google Colab) by up to 3x and significantly increase inferencing speed (up to 0.03 seconds per input) using post-quantization.

Course Sections:

• Non-Linear Function Approximation
• Visual Calculator
• Custom Voice Controlled LED

What You'll Achieve:

• Create Deep Learning Projects on Embedded Hardware: Gain the skills to deploy AI models on the Raspberry Pi.
• Convert Models to TensorFlow Lite: Optimize your models for efficient execution on resource-constrained devices.
• Speed Up Inferencing: Reduce latency and improve responsiveness of your embedded AI applications.
• Post-Quantization Techniques: Master techniques to compress and accelerate deep learning models.
• Work with Custom Data for AI Projects: Learn to train models on your own unique datasets.
• Hardware Optimized Neural Networks: Understand how to tailor models for specific hardware platforms.
• Computer Vision Projects with OpenCV: Apply computer vision techniques using the popular OpenCV library.
• Fast Inferencing with Deep Neural Networks: Optimize deep learning models for real-time performance.

Hardware Requirements:

• Raspberry Pi 4
• 12V Power Bank
• 2 LEDs (Red and Green)
• Jumper Wires
• Bread Board
• Raspberry Pi Camera V2
• RPI 4 Fan
• 3D Printed Parts (optional)

Software Requirements:

• Python 3
• A motivated mindset for a challenging programming project!

Ready to get started? Explore the course GitHub repository before you enroll!