Download Training Directory (2021) Download Now

Jan 2021

1-Day Neural Network Using Keras for Deep Learning Workshop (LIVE Stream)

Course Information

Start Date31 Jan 2021, Sunday
End Date31 Jan 2021, Sunday
Time09:30 am - 04:30 pm
VenueOnline LIVE Stream via Zoom
Fee$680 (Excluding GST) Inclusive of e-materials and e-certificate
Contact6720 3333 (Ms Rina Lim) training.aventis@gmail.com
Register Now
Get Group Quote
LIVE Stream

Keras: A User Friendly TensorFlow API

Tensorflow, developed by Google, is a popular open source Deep Learning framework. Keras is a user friendly Tensorflow API that simplifies the coding for neural networks and deep learning. Kera allows one to build neural network deep learning models readily for predictive data analytics, regression, classification and data modelling. Learners can apply Keras for supervised and unsupervised machine learning tasks.

Learn & Add this In-Demand Skills to Your Skills Set

Cut through the noise and Learn Real Skills with a step-by-step approach to understanding deep learning with Keras programming. In this comprehensive, yet straight-forward Deep Learning with Keras course, you will become familiar with the language and fundamental concepts of artificial neural networks, PyTorch, autoencoders, and more. Upon completion, you will be able to build deep learning models, interpret results and learn how to start your own deep learning project.

Key Takeaways
  • Gain insight into the fundamental concepts of neural networks
  • Learn to think like a data scientist and understand the difference between machine learning and deep learning
  • Discover various techniques to evaluate, tweak, and improve your models
  • Explore different techniques to manipulate your data
  • Explore alternative techniques to verify the accuracy of your model
Who Should Attend?

Deep learning is one of the latest technologies in AI and machine learning, used in smartphone apps, power grids, helping us find solutions to climate change, and more.  This course can lead to rewarding roles in IT, Marketing healthcare, FinTech, e-commerce, and other industries.

Highly suitable for:

  • Suited for Finance, Sales, Service, Business Development and Marketing Analysis seeking to gain deeper understanding of deep learning and neural network
  • Existing programmers, data scientists , data analyst and engineers with development experience in another high-level programming or scripting language such as C/C++, Perl, PHP etc. will also benefit by learning the Python Programming Language.


Note: Participant is required to bring their own laptop with access to internet (WiFi network will be provided)

Data Scientist: Dr Sudipta Samanta, Senior Research Engineer at Temasek Life Sciences Lab

Dr. Sudipta Samanta has over 15 years of research and teaching experience. Dr Sudipta received his Ph.D. for his research in the field of computational Biophysics and has over 10 years of research experience as a Research Scientist. His current interests include Machine Learning, Deep Learning, Health Care Data analysis and computer simulation. Prior to joining Temasek, Dr Sudipta was Visiting Scientist of Internationally renowned MIT Computer Science and Artificial Intelligence Laboratory (CSAIL), USA from 2011 to 2018 and a Research Scientist at Singapore-MIT Alliance for Research & Technology.

Workshop Outline

Topic 1 Introduction
• What is Keras?
• Keras vs TensorFlow
• Install and Run Keras
• Machine Learning vs Deep Learning

Topic 2 Fully Connected Neural Network (FcNN)
• What is Neural Network (NN)?
• Details understanding of FcNN.
• Activation Functions
• Loss Function and Optimizer
• Regression Predictive Model with NN
• MNIST and CIFAR10 Datasets
• Training and Validation
• One Hot Encoding
• SoftMax activation function for classification
• Build FcNN based classification model with Image datasets.

Topic 3 Convolutional Neural Network (CNN)
• What is CNN?
• Convolution, Padding, Stride, Activation and Pooling Operations.
• Image Classification Model with CNN
• Overfitting and Underfitting
• Data Augmentation and Dropout

Topic 4 Transfer Learning
• What is Transfer Learning
• Keras Pre-Trained Models (Vgg16, ResNet, DenseNet, etc.)
• Fine Tuning Pre-Trained Models

Topic 5 Recurrent Neural Network (RNN)
• Sequential Data
• What is RNN?
• Different types of RNN?
• Long Term Dependencies with LSTM cells
• Model Building with RNN