Machine Learning Material

Materials or code that I have used to learn machine/deep learning.

Project Structure

1. src/: source code
2. Unittest/: unit test for source code
3. notebooks/: jupyter notebooks
4. scripts/: python scripts
5. data/: training data

Refer to cookiecutter-data-science-structure for machine learning project structure

Machine Learning Algorithms Code Implementation

• Traditional Machine Learning
  • Generalized Linear Model
    • Linear regression (Lasso, Ridge)
    • logistic regression
    • Linear discriminant analysis
  • Decision tree
    • ID3 decision tree
    • C4.5 decision tree
    • CART
    • Random Forest
    • Gradient Boosted Tree
  • SVM
    • Linear SVM (Binary classification)
    • Kernel SVM (Binary classification)
  • KNN
  • Naive Bayes
  • Clustering
    • K-means
    • Hierarchical Clustering
• Deep Learning
  • Neural Network
  • Sequence Model
    • RNN
    • LSTM
    • GRU
    • Attention
    • seq2seq
    • transformer
  • CNN:
    • AlexNet
    • VGG11
    • NiN
    • GoogLeNet
    • ResNet18
    • DenseNet
• (Deep) Reinforcement Learning
  • Multi-bandit
    • Contextual multi-bandit
  • Q learning
  • SARSA
  • Deep Q-learning
  • Policy gradient
  • Actor-Critic
  • ....
• Representation Learning
  • Word2Vector
• Application
  • Recommendation
    • Deep Semantic Similarity Model(DSSM)
    • YoutubeDNN
    • MIND
    • Matrix Factorization (MF)
    • Factorization Machine (FM)
    • Wide & deep
    • DeepFM

Course

1. Multi-agent System
2. Udacity - Machine Learning DevOps Nanodegree Program
3. Tensoflow: avanced technique specialization

Competition codes:

1. Personalize Expedia Hotel Searches - ICDM 2013
2. Amazon KDD Cup 2023 - Build Multilingual Recommendation System

More learning codes

1. Kaggle
   1. categorical feature with LR, XGBoost and CatBoost

How to run the code

Step 1: environment setup