Deep Learning Specialization: Convolutional Neural Networks (CNNs) Course Projects

Overview

This repository contains my projects and assignments completed as part of the Convolutional Neural Networks course in the Deep Learning Specialization on Coursera. This course provided an in-depth understanding of how computer vision has evolved, covering exciting applications such as autonomous driving, face recognition, and reading radiology images.

Key Learning Outcomes:

• Building and training convolutional neural networks, including variations such as residual networks.
• Applying CNNs to visual detection and recognition tasks.
• Mastering neural style transfer for art generation.
• Utilizing deep CNNs in various image, video, and other 2D or 3D data applications.

Course Structure and Content

The course comprised four modules with a blend of theory and practical assignments.

Module 1: CNN Basics and Case Studies

• Videos: Covering classic networks like ResNet, Inception Network, MobileNet, and EfficientNet.
• Programming Assignments:
  • Implementing Residual Networks
  • Transfer Learning with MobileNet

Module 2: Object Detection

• Videos: Focusing on object localization, YOLO algorithm, and semantic segmentation with U-Net.
• Programming Assignments:
  • Car Detection using YOLO
  • Image Segmentation with U-Net

Module 3: Specialized CNN Applications

• Videos: Exploring face recognition and neural style transfer.
• Programming Assignments:
  • Face Recognition
  • Art Generation with Neural Style Transfer

Assessments

• Quizzes and readings complementing each module to solidify the understanding of concepts.

Repository Contents

Each folder in this repository corresponds to a module in the course and includes:

• Jupyter notebooks with code and detailed explanations.
• Datasets used in the projects (or links to access them).
• Supplementary resources and notes taken during the course.

Skills Acquired

• Deep understanding of CNN architectures and their applications.
• Proficiency in Python, TensorFlow, and PyTorch for implementing deep learning models.
• Ability to evaluate model performance using metrics like BLEU and ROUGE.
• Experience in real-world applications of CNNs in various domains.

How to Use This Repository

1. Explore Individual Projects: Each project folder contains all the necessary files to understand and run the project.
2. Dataset Access: Follow the links/instructions in each project folder to access the datasets used.
3. Running the Notebooks: You can run the Jupyter notebooks in your environment to see the models in action.

Acknowledgments

A big thank you to the course instructors and Coursera for providing such a comprehensive and practical learning experience in Deep Learning.