
Explained the Primary Focus of this course.
Described a detailed course layout.
You will develop a broad level of understanding of Machine Learning.
You will learn about Linear Regression and steps to calculate the parameters.
You will learn about Linear Regression-Gradient Descent using Mean Squared Error Cost Function.
We will show you how Linear Regression parameters are calculated using a simple example.
You will learn about Logistic Regression: Classification problem.
You will learn about Decision Boundary and Sigmoid Function.
We will continue our discussion about Decision Boundary and Sigmoid Function.
We will discuss Gradient Descent using Mean Squared Error Cost Function and the problems.
You will learn the concept of Entropy for a probability distribution.
You will learn the concept of Cross-Entropy for a probability distribution.
We will continue with the concept of Cross-Entropy for a probability distribution.
We will explain Cross-Entropy as a Cost-Function.
You will learn Gradient Descent with Cross-Entropy Cost Function.
We will explore the idea of Multiclass Classification using a binary classifier.
Introduction to the Logical Operators.
You will learn how to model Logical Operators using Perceptron(s).
You will develop a basic understanding of Biological Neuron.
You will develop a basic understanding of the math/numbers behind an image.
We will start the topic of Multiclass Classification using Neural Network .
We will work on the steps to calculate weights using the back-propagation technique.
We will continue on the steps to calculate weights using the back-propagation technique for inner layers.
We will continue on the steps to calculate weights using the back-propagation technique for inner layers.
Changing the activation function from Sigmoid to ReLU for inner layers and SIgmoid to Softmax for the output layer.
ReLU and Softmax explained.
You will learn how to setup Google Colab. and mount Google Drive.
Hands-on Deep Neural Network (DNN) based image classification using Google Colab. & TensorFlow.
Continuation of hands-on Deep Neural Network (DNN) based image classification using Google Colab. & TensorFlow.
Introduction to Convolution Neural Networks (CNN).
Explained the concept of Convolution Neural Networks (CNN) Architecture, Feature Extraction, Filters, and Pooling Layer.
Hands-on Convolution Neural Networks (CNN) based image classification using Google Colab & TensorFlow.
Explained Overfitting and Underfitting problems and how to address that issue.
Explained Regularization, Dropout, and Early-Stopping to address overfitting and underfitting problems.
Hands-on example to fix overfitting and underfitting problems.
If you are interested in Machine Learning, Neural Networks, Deep Learning, Deep Neural Networks (DNN), and Convolution Neural Networks (CNN) with an in-depth and clear understanding, then this course is for you.
Topics are explained in detail. Concepts are developed progressively in a step by step manner. I sometimes spent more than 10 minutes discussing a single slide instead of rushing through it. This should help you to be in sync with the material presented and help you better understand it.
The hands-on examples are selected primarily to make you familiar with some aspects of TensorFlow 2 or other skills that may be very useful if you need to run a large and complex neural network job of your own in the future.
Hand-on examples are available for you to download.
Please watch the first two videos to have a better understanding of the course.
TOPICS COVERED
What is Machine Learning?
Linear Regression
Steps to Calculate the Parameters
Linear Regression-Gradient Descent using Mean Squared Error (MSE) Cost Function
Logistic Regression: Classification
Decision Boundary
Sigmoid Function
Non-Linear Decision Boundary
Logistic Regression: Gradient Descent
Gradient Descent using Mean Squared Error Cost Function
Problems with MSE Cost Function for Logistic Regression
In Search for an Alternative Cost-Function
Entropy and Cross-Entropy
Cross-Entropy: Cost Function for Logistic Regression
Gradient Descent with Cross Entropy Cost Function
Logistic Regression: Multiclass Classification
Introduction to Neural Network
Logical Operators
Modeling Logical Operators using Perceptron(s)
Logical Operators using Combination of Perceptron
Neural Network: More Complex Decision Making
Biological Neuron
What is Neuron? Why Is It Called the Neural Network?
What Is An Image?
My “Math” CAT. Anatomy of an Image
Neural Network: Multiclass Classification
Calculation of Weights of Multilayer Neural Network Using Backpropagation Technique
How to Update the Weights of Hidden Layers using Cross Entropy Cost Function
Hands On
Google Colab. Setup and Mounting Google Drive (Colab)
Deep Neural Network (DNN) Based Image Classification Using Google Colab. & TensorFlow (Colab)
Introduction to Convolution Neural Networks (CNN)
CNN Architecture
Feature Extraction, Filters, Pooling Layer
Hands On
CNN Based Image Classification Using Google Colab & TensorFlow (Colab)
Methods to Address Overfitting and Underfitting Problems
Regularization, Data Augmentation, Dropout, Early Stopping
Hands On
Diabetes prediction model development (Colab)
Fixing problems using Regularization, Dropout, and Early Stopping (Colab)
Hands On: Various Topics
Saving Weights and Loading the Saved Weights (Colab)
How To Split a Long Run Into Multiple Smaller Runs
Functional API and Transfer Learning (Colab)
How to Extract the Output From an Intermediate Layer of an Existing Model (Colab), and add additional layers to it to build a new model.