Deep-Learning-Specialization

Table of contents

• General info
• Code setup Information
• Introduction to Deep Learning and Neural Networks with Keras
• Deep Neural Networks with PyTorch
• Building Deep Learning Models with TensorFlow

General info

This repository is on Deep Learning Specialization using Python 3.8.3 on Visual Studio Code. Most of the programs are from IBM AI Engineering are presenterd only for learning purpose.

Code setup Information

• The code is writin in Visual Studio code using Python extension with Python 3.8.3 installed on the system.
• Output results are written in text file (.txt)
• The illustrations are saved as eps and using latex, each documentation is created as .pdf

Introduction to Deep Learning and Neural Networks with Keras

This course focuses on the concept of Deep Learning using the idea of Artificial Neural Networks (ANN). Along with fundamental concepts, ANN techniques such as Forward Propagation, Gradient Descent, Backpropagation, Vanishing Gradient, Activation Functions are explained in this course. There are 5 modules in this course.

• Module 1 - Introduction to Deep Learning
• Module 2 - Artificial Neural Networks
• Module 3 - Deep Learning Libraries
• Module 4 - Deep Learning Models
• Module 5 - Course Assignment

Deep Neural Networks with PyTorch

This course focuses on the general coding concepts for PyTorch library using Python and covers fundamental concepts such as Tensors, Dataset Transformation, Linear Regrression including Multiple Input Output Linear Regression. For classification, Logistic Regression using PyTorch is covered along with Softmax Regression. Fundamental concepts of Shallow Neural Networks and Deep Networks is explained in this course. Finally, an important and popular deep learning technique Convolution Neural Network for image classification and object detection is covered in the end.

Honors Peer Review Assignment

   Fashion MNIST Classification Assignment
   Objective - to apply and train CNN
   Not Graded but to receive honors badge

There are 7 modules in this course.

• Module 1 - Tensors and Datset
• Module 2 - Linear Regression / Linear Regression PyTorch Way
• Module 3 - Multiple Input Output Linear Regression / Logistic Regression for Classification
• Module 4 - Softmax Rergresstion / Shallow Neural Networks
• Module 5 - Deep Networks
• Module 6 - Convolutional Neural Network
• Module 7 - Honors Peer Review Assignment

Building Deep Learning Models with TensorFlow

This course focuses on the general machine learning and deep learning algorithms using TensorFlow. It consists of general mathematical concepts, supervised learning like Convolution Neural Networks (CNN), Recurrent Neural Networks (RNNs) and unsupervised learning concepts like Restricted Boltzmann Machines (RBMs) and Autoencoders.

Applications of above concepts are:

   Object Recognition - CNN
   Handwritten Notes - CNN
   Natural Language Processing (NLP) - Sequential data and RNN
   Pattern Detection - RBMs
   Movie Recommendation System - Autoencoder

There are 5 modules in this course.

• Module 1 - Introduction to TensorFlow
• Module 2 - Convolutional Neural Networks (CNNs)
• Module 3 - Recurrent Neural Networks (RNNs)
• Module 4 - Restricted Boltzmann Machines (RBMs)
• Module 5 - Autoencoders

---

Note: Each program is compiled along with output log file and results in pdf using latex.