Coursera - Machine Learning

This repo contains python code implementation of coursera machine learning course by Prof. Andrew Ng. The original course was done in MATLAB

All homework exercises on June 6th, 2020
Coded in jupyter notebook, all functions are defined inside each notebook file

Syllabus

Week 1

• Introduction to Machine Learning
  • Supervised Learning
  • Unsupervised Learning
• Linear Regression with One Variable
  • Cost Function
  • Gradient Descent
• Linear Algebra Review

Week 2

• Linear Regression with Multiple Features
  • Feature Scaling
  • Learning Rate
  • Features and Polynomial Regression
  • Normal Equation
• Octave/ Matlab Tutorial

Week 3

• Logistic Regression
  • Sigmoid Function
  • Simplified Cost Function and Gradient Descent
  • One-vs-all Classification
• Regularization
  • Overfitting
  • Regularized Linear Regression
  • Regularized Logistic Regression

Week 4

• Neural Network
  • Introduction
  • Neurons and Brain
  • Neuron Model: Logistic Unit (Perceptron)
  • Forward Progaion: Vectorized Implementation
  • Multiple Output Units

Week 5

• Neural Network: Learning
  • Cost Function
  • Backpropagation Algorithm
  • Gradient Checking
  • Random Initialization

Week 6

• Advice for Machine Learning Implementation
  • Hypothesis Evaluation
  • Train/Validation/Test Error Analysis
  • Bias vs Variance Diagnosis
  • Learning Curve

Week 7

• Support Vector Machine
  • Optimization Objective
  • Large Margin Intuition
  • Linear and Gaussian Kernels

Week 8

• Clustering
  • Unsupervised Learning
  • K-Means Algorithm
  • Choosing Number of Cluster
• Dimensionality Reduction
  • Data Compression & Visualization
  • Principal Component Analysis (PCA)
  • Reconstruction from Compressed Representation
  • Choosing Number of PCA

Week 9

• Anomaly Detection
  • Gaussian Distribution
  • Multivariate Gaussian Distribution
• Recommender System
  • Problem Formulation
  • Content-based Recommendations
  • Collaborative Filtering
  • Vectorization: Low Rank Matrix
  • Mean Normalization

Week 10

• Large Scale Machine Learning
  • Learning with Large Dataset
  • Stochastic Gradient Descent
  • Mini-batch Gradient Descent
  • SGD Convergence
  • Online Learning
  • Map-reduce and Data Parallelism

Week 11

• Application Example
  • Photo OCR
  • Sliding Windows
  • Artificial Data Synthesis

Homework to Each Week Content

Week	Corresponding Homework	Content
1	-	Introduction to ML
2	ex1-Linear_Regression	Linear Regression
3	ex2-Logistic_Regression	Logistic Regression
4	ex3-Multi_Class_Classification_and_NN	Multi Class Classification & NN
5	ex4-Neural_Network_Learning	Neural Network Learning
6	ex5-Regularized_Linear_Regression	Advice for Machine Learning Implementation
7	ex6-Support_Vector_Machine	Support Vector Machine
8	ex7-K_means_Clustering_and_PCA	Clustering & Dimensionality Reduction
9	ex8-Anomaly_Detection_and_Reccomendation	Anomaly Detection & Recommender System
10	-	Large Scale Machine Learning
11	-	Application Example: Photo OCR

Note

The develop folder contains my experiment of exercise 1 with MATLAB, this might be deleted later, the purpose is to understand how mesh function works