Keras Training Prerequisites
To apply for the Keras Training, you need to either:
- You need to have a good foundation in mathematical concepts like linear algebra, calculus, probability and statistic
- You need to know at least one programming language like Python or R. You need to have a good understanding of OOPs concepts, algorithms and data structures.
- You need to have some basic data analysis and data visualisation skills.
Course Curriculum
Module 1: Basic overview of Keras
- Understanding the features of Keras
- Advantages of Keras
- Keras Limitations
Module 2: Keras installation and API
- Installing Keras
- Installation of dependencies
- Installation of Theano
- Installation of TensorFlow
- Installation of Keras
- Testing each Installation
- Configuring Keras
- Installation of Keras on Docker
- Installation of Keras on Google Cloud ML
- Installation of Keras on Amazon AWS
- Installation of Keras on Microsoft Azure
- Keras API
- Basic Architecture of Keras
- Overview of predefined neural network layers
- Overview of predefined activation functions
- Understanding Losses Functions
- Understanding Metrics
- Some Useful Operations
Module 3: Overview of Deep Learning with ConvNets
- Deep Convolutional Neural Network
- Understanding Deep Convolutional Neural Network (DCNN)
- Simple Example of DCNN
- Recognizing CIFAR-10 images with DL
Module 4: Word Embeddings
-
- Distributed Representations
- Understanding word2vec
- GloVe Exploring functionalities
- Using pre-trained embeddings
Module 5: Concept of Generative Adversarial Networks and Wavenet
- Overview of GAN
- Keras adversarial GANs for forging MNIST
- Keras adversarial GANs for forging CIFAR
- Understanding WaveNet
Module 6: Overview of Recurrent Neural Networks RNN
- Basics of SimpleRNN cells
- Understanding RNN Topologies
- Using Gated Recurrent Unit -GRU
- Concept of Bidirectional RNNs
- Vanishing and exploding gradients
- Using Long Short Term Memory -LSTM
- Understanding Stateful RNNs
Module 7: Additional Deep Learning Models
- Dealing with Keras Functional API
- Understanding Regression Networks
- Concept of Unsupervised Learning
- Keras Customization scenario
- Using Generative Models
