100 Days Challenge Day 1 - Linear Regression, Logistic Regression and Neural Networks

 

100 Days Challenge - Day 1.

Linear Regression, Logistic Regression and Neural Networks.

Revised Week 1 to Week 4 of Andrew Ng's Machine Learning course on Coursera which I had already completed a few months ago.

Topics covered include:

Introduction to Machine Learning

• Supervised
• Unsupervised

Linear Regression in One Variable

• Model Representation
• Cost Function
• Gradient Descent

 Linear Regression in Multiple Variables

• Model Representation
• Gradient Descent
• Feature Scaling
• Learning Rate
• Polynomial Regression
• Normal Equation
• Normal Equation vs Gradient Descent
• Normal Equation Non-Invertibility

 Logistic Regression

• Hypothesis Representation
• Sigmoid Function
• Decision Boundary (Linear, Non-Linear)
• Cost Function and Gradient Descent
• One vs All Multi-class Classification

 Regularization

• Overfitting
• Correction in Cost Function (Regularization)
• Gradient Descent
• Normal Equation and Invertibility

Neural Networks

• Examples and Intuitions
• Multi-class Classification