This repository documents my self-taught journey to learn Machine Learning from scratch. The goal is to gain a comprehensive understanding of machine learning, starting from the basic mathematical, data-structure and algorithm.
- Day1 (1 June 2023) Review my Linear Algebra, Calculus, Prob/Stat, Python and overview concept of ML
- Day2 Concept Learning : Find-S, List-Then-Eliminate (clustering)
- Day3 Candidate Elimination
- Day4-6 Focused on understanding and implementing the k-NN
- Day7 learn more numpy useful functions and methods
END OF WEEK 1
WEEK 2
- Day8++ I'm progressing through the 'Introduction to Machine Learning' course on coursera. Below, I've shared what I've learned so far.
- Linear regression with one variable
- Cost function
- Gradient Descent
- Multiple linear regression
- Vectorization
- Gradient Descent With Multiple Variables
- Feature Scaling and Learning Rate
- Sklearn
- GD
- normal equation
WEEK 3+4
- Classification , sigmoid function
- Logistic regression
- Decision Boundary
- Cost function
- Gradient Descent
- Scikit for logistic regression
- Overfitting
- Regularization
END COURSE 1 and start COURSE 2 : Advanced Learning Algorithms
WEEK 5
- Introduction of neural network
- Neurons and Layer
- implement with tensorflow (use Tensorflow 2.7.0 out of date)
- Vectorization in each layer
WEEK 6
- Tensorflow detail
- Relu
- Softmax
- Multiclass vs MultiLabel
- Optimization and Advance Optimization
- Back propagation
- computation graph
WEEK 7
- Bias and variance
- Learning curves
- Bias and variance in neaural networks
- Error analysis
- Skewed datasets
WEEK 8
- Dicision tree model
- Learning Process
- Information gain
- Regresion tree
- Tree ensembles
- Random forest algorithm
- XGBoost
start COURSE 3 : Unsupervised Learning, Recommenders, Reinforcement Learning
WEEK 9,10,11
- clustering
- K-means
- anomaly detection
- collaborative filtering
- content-based filtering
- reinforcement learning
- DQN
END COURSE 3 : and start NEWCOURSE : Deep Learning Specialization (5 sub-courses)