Email:info@araniconsulting.com
Module 1
Unit 1: Deep Learning Introduction
- Introduction to DL problems
- DL terminologies
- DL project workflow
- DL real life examples
Unit 2: Jupyter Notebook introduction
- Working with Jupyter notebooks
- Markdown and Code blocks
- Keyboard shortcuts
Unit 3: Python Basics
- Python syntax
- Basic data types
- Basic data structures
Unit 4: Python advanced
- Numpy Arrays
- Plotting using Matplotlib
- Pandas Dataframes
- Introduction to Keras
Module 2
Unit 1: Artificial Neural Networks (ANN)
- What is a Neuron
- What are Activation Functions
- How does a neural network learn?
- Gradient Descent
- Stochastic Gradient Descent
- Back Propagation
- Artificial Neural Networks in Keras
- Linear model (No Hidden Layers)
- Neural network with a single hidden layer
Module 3
Unit 1: Convolutional Neural Networks (CNN)
- Image representation
- ConvNets or CNN
- Convolution Layer
- Padding
- How do we learn these kernels?
- Can we force a particular kernel to learn to recognize a specific feature?
- Non-Linear Activations
- Downsampling or Pooling
- Full Connection
- Loading MNIST data
- Implementation of CNN in Keras
Module 4
Unit 1: Auto Encoders (AE)
- Introduction to Auto Encoders
- Why learn identity function?
- Properties of learned function
- Real world applications of Autoencoders
- MNIST Dimensionality Reduction
- A Simple Autoencoder
- Functional API
- Deep Autoencoder
- Convolutional Autoencoder
- Image Denoising
- Data Specific Encoding and Decoding
Module 5
Unit 1: Recurrent Neural Networks (RNN)
- Introduction to Recurrent Neural Networks
- Sequence Learning
- Regular Neural Network
- Simple RNN
- Problems with RNN
- Long Short Term Memory (LSTM)
- Stacked (Deep) LSTM Model
- Deep Stacked LSTM with Stateful Cells
- Gated Recurrent Unit