TWESD Workshop - TensorFlow Lite Model Optimization for On-Device Machine Learning

This repository contains two folders:

1. CNN Model with TFLite Conversion

   • This folder contains a Jupyter Notebook (TensorFlow_Lite_Model_Optimization_for_On_Device_Machine_Learning.ipynb) for creating a Convolutional Neural Network (CNN) model.
   • The notebook demonstrates how to convert the trained model using TFLite with various quantization techniques.
   • The trained model files with the .tflite extension are included in this folder.

2. EfficientNet Model with Transfer Learning and TFLite Export

   • This folder includes a Jupyter Notebook (TensorFlow_Lite_Model_Optimization_with_transfer_learning.ipynb) for utilizing the EfficientNet model with transfer learning.
   • The notebook showcases the process of creating a model using the TFLite Creator and exporting it to different quantization techniques with the .tflite extension.
   • The exported model files are included in this folder.

Feel free to explore the folders and notebooks for detailed explanations, code samples, and instructions on using and converting deep learning models.

Prerequisites

Make sure you have the following dependencies installed:

• Python 3.x
• TensorFlow
• TensorFlow Lite
• EfficientNet

Usage

1. Clone the repository to your local machine:

git clone https://github.com/Mohamed-Dhouioui/TWESD.git

You can also download the repository as a zip file and extract it.

As for the presentation slides, you can find them in the following link: https://my.visme.co/view/y4zqj77x-twesd