{"id":13219,"date":"2026-08-10T10:56:20","date_gmt":"2026-08-10T10:56:20","guid":{"rendered":"https:\/\/nokobox.com\/index.php\/item\/udemy-deep-learning-using-tensorflow-lite-on-raspberry-pi-2024-7\/"},"modified":"2026-08-10T10:56:20","modified_gmt":"2026-08-10T10:56:20","slug":"udemy-deep-learning-using-tensorflow-lite-on-raspberry-pi-2024-7","status":"publish","type":"digital_item","link":"https:\/\/nokobox.com\/index.php\/item\/udemy-deep-learning-using-tensorflow-lite-on-raspberry-pi-2024-7\/","title":{"rendered":"Udemy \u2013 Deep learning using Tensorflow Lite on Raspberry Pi 2024-7"},"content":{"rendered":"
\n

Descriptions<\/h2>\n

Deep learning using Tensorflow Lite on Raspberry Pi, This course is focused on Embedded Deep learning in Python. Raspberry Pi 4 is used as the main hardware, and you will build practical projects with custom data. You start by approximating trigonometric functions, generating random data and producing a model for Sin function approximation. Next, you create a calculator that takes images as input, builds equations, and produces results using convolutional network architecture for categorical classification. Another project uses custom voice recordings and convolutional networks, with electronics to control multiple LEDs using your own voice. A unique learning point is post quantization applied to TensorFlow models trained on Google Colab, reducing model size by three times and increasing inference speed up to 0.03 seconds per input. You will work on non-linear function approximation, a visual calculator, and a custom voice-controlled LED project. By the end, you can create deep learning projects on embedded hardware, convert models to TensorFlow Lite, speed up inference on embedded devices, apply post quantization, use custom data for AI projects, optimize neural networks for hardware, build computer vision projects with OpenCV, and deploy deep neural networks with fast inference speed. Hardware requirements include Raspberry Pi 4, 12V power bank, two LEDs (red and green), jumper wires, breadboard, Raspberry Pi Camera V2, RPI 4 fan, and 3D printed parts. Software requirements include Python 3 and a motivated mind for a large programming project.<\/p>\n

What you\u2019ll learn<\/h3>\n