Overview

Save Big on Coursera Plus. 7,000+ courses at $160 off. Limited Time Only!

Grab it

Embark on a journey through the intricate workings of advanced Convolutional Neural Networks (CNNs), Transfer Learning, and Recurrent Neural Networks (RNNs). This course begins with a thorough exploration of CNNs, delving into sophisticated architectures like VGG16 and practical applications through multi-part case studies. Each segment is designed to build your foundational knowledge and practical skills incrementally. Transitioning into Transfer Learning, the course explores pivotal models such as AlexNet, GoogleNet, and ResNet. You will engage with numerous hands-on sessions, applying transfer learning techniques to real-world datasets. These sessions are meticulously crafted to ensure a robust understanding of how pre-trained models can accelerate your projects and improve outcomes. The course culminates with an in-depth study of Recurrent Neural Networks, including Long Short-Term Memory (LSTM) networks and Gated Recurrent Units (GRUs). By working through comprehensive case studies, you'll gain practical experience in applying RNNs to sequential data tasks such as part-of-speech tagging and text generation. Each module is designed to provide a seamless learning experience, combining theoretical insights with practical implementation. This course is tailored for data scientists, machine learning engineers, and AI enthusiasts with a solid understanding of basic neural networks and Python programming. Prerequisites include prior experience with deep learning frameworks such as TensorFlow or Keras, and familiarity with fundamental machine learning concepts.

Syllabus

CNN-Keras

In this module, we will delve into the basics of CNNs, examining the VGG16 architecture, and engage in a comprehensive case study spread across multiple practical sessions. These hands-on exercises will reinforce the theoretical concepts covered.

CNN-Transfer Learning

In this module, we will explore various pre-trained models, their architectures, and the principles of transfer learning. Through a series of detailed sessions, we will apply these concepts in practical settings, culminating in case studies and analytical discussions.

CNN-Industry Live Project: Playing with Real-World Natural Images

In this module, we will apply CNN techniques to real-world natural images, specifically focusing on flower images. Through an extensive case study spread over multiple sessions, we will learn to implement, evaluate, and refine models in a practical, industry-relevant context.

CNN-Industry Live Project: Find Medical Abnormalities and Save a Life

In this module, we will tackle the challenge of identifying medical abnormalities using CNNs. Focusing on X-Ray images, we will conduct a detailed case study over several sessions, learning to interpret medical data and develop effective diagnostic models.

Recurrent Neural Networks: Introduction

In this module, we will introduce Recurrent Neural Networks, covering their basic concepts, architecture, and types. We will delve into training methods and address common challenges like the vanishing gradient problem through a series of detailed sessions.

Recurrent Neural Networks: LSTM

In this module, we will focus on Long Short-Term Memory (LSTM) networks, covering their architecture and functionality. We will compare LSTM with other RNN variants like GRU and implement these networks in practical scenarios through a series of detailed sessions.

Recurrent Neutral Networks: Part-Of-Speech Tagger

In this module, we will apply RNN techniques to develop a Part-Of-Speech tagger for natural language processing tasks. Through an extended case study spread across multiple sessions, we will develop, evaluate, and refine the performance of the Part-Of-Speech tagger.

Text Generation Using RNN

In this module, we will delve into the practical application of RNNs for text generation by exploring a comprehensive code generator case study divided into four parts. Each part builds on the previous one, enhancing our understanding and skills in using RNNs for generating coherent text.