Course Overview
This course gives learners an overview of working of neural network for predictive analytics and its use in performing advanced machine learning and building artificial intelligent systems. The learners get to work on advanced libraries such as TensorFlow and Keras developed by Google.
Course Objectives
- Learn the statistical algorithms behind deep learning and understand how to test hypotheses
- Understand the platforms and tools available for deep learning and the various Python libraries that can be used
- Understand and learn how to code deep learning algorithms and develop models using Python Programming
- Use statistical modelling and deep learning algorithms to train data to perform various actions and derive patterns
- Evaluate deep learning models to identify intended and unintended outcomes
- Evaluate and optimize effectiveness of deep learning models
- Optimize model using Python programming and statistical methods
- Use deep learning models for predictive and diagnostic analysis and draw relevant insights required to support decision making
Pre-requisites
This course requires working knowledge of Python and basic understanding of machine learning. Participants who do not meet these requirements are encouraged to take up Fundamentals of Python Programming and Advanced Analytics and Machine Learning using Python prior to this course.
- Hardware & Software
This course will be conducted as a Virtual Live Class (VLC) via Zoom platform. Participants must own a zoom account and have a laptop or a desktop with “Zoom Client for Meetings” installed. This can be downloaded from https://zoom.us/download
System Requirement |
Must Have:
Please ensure that your computer or laptop meets the following requirements.
|
Course Outline
MODULE 1: Introduction to AI and Basics of Neural Networks
This module introduces the fundamentals of Artificial Intelligence and Deep Learning. You will learn about the role of AI and Deep Learning in businesses today. The module also covers the basics of Neural Networks and how you can use Neural Networks to solve simple problems.
- Introduction to AI and Deep Learning
- What is AI and Deep Learning?
- Role of AI and Deep Learning in businesses today
- What can you do with deep learning?
- Neural Networks
- Building Blocks of neural networks
- Implementing Neural Networks using Python
- Multilayer perceptron for deeper networks
- Activity 1: Creating a simple NN
- Activity 2: Creating NN for multiple outputs
- Activation Functions and Cost Functions
- Gradient Descent Backpropagation
- Hyperparameters of an NN architecture
- Activity 3: Manual neural network classification task
MODULE 2: Introduction to TensorFlow
This module covers Python libraries for Deep Learning and provides an overview of TensorFlow basics. You will learn about TensorFlow graphs, variables, and placeholders, and develop NN models using TensorFlow.
- Python libraries for Deep Learning
- TensorFlow basics
- TensorFlow graphs, variables and placeholders
- Creating NN with TensorFlow
- Regression using TensorFlow
- Classification using TensorFlow
- Activity 1: Developing a regression model using TensorFlow
- Activity 2: Developing a classification model using TensorFlow
- Saving and restoring models
- Deployment of inference ft. Gradio
MODULE 3: Convolutional Neural Networks
This module provides an in-depth overview of the architecture of Convolutional Neural Networks. You will explore the MNIST dataset and learn how to classify images using CNNs.
- Understanding CNNs and Architecture of a CNN
- MNIST data – Overview
- Image classification using CNN
- Activity: Developing CNN model to classify MNIST CNN dataset
- Real world industry examples of CNNs in action
MODULE 4: Recurrent Neural Networks
This module introduces the RNN layer and teaches you how to create a RNN. You will also learn about the Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU).
- Understanding RNNs
- Architecture of an RNN and Implementing RNN using Python
- Introduction to LSTM and GRU
- RNN with TensorFlow API
- Activity: Time series forecasting using RNN
- Real world industry examples of RNNs in action
MODULE 5: Object Detection and Deep Fakes
This module introduces Autoencoders and Generative Adversarial Networks (GAN). You will learn how deep fakes are created and explore object detection using GAN.
- Introduction to AutoEncoders
- Introduction to Generative Adversarial Networks (GAN)
- How deep fakes are created?
- Activity: Object detection using GANs
- Real world industry examples of GANs in action
Certificate Obtained and Conferred by
Certificate of Completion from NTUC LearningHub
Upon meeting at least 75% attendance and passing the assessment(s), participants will receive a Certificate of Completion from NTUC LearningHub.
Categories
Subjects
More Information
- NTUC LearningHub
Add a review