λ³Έ κ°μλ TEAMLABκ³Ό Inflearnμ΄ ν¨κ» ꡬμΆν λ°μ΄ν° μ¬μ΄μΈμ€ κ³Όμ μ λ λ²μ§Έ κ°μμΈ λ°λ°λ₯ λΆν° μμνλ λ¨Έμ λ¬λ μ
λ¬Έ
μ
λλ€. λ°λ°λ₯λΆν° μμνλ λ¨Έμ λ¬λ μ
λ¬Έ
μ Part Iκ³Ό Part IIλ‘ κ΅¬μ±λμ΄ μμ΅λλ€.
λ³Έ κ°μλ TEAMLABκ³Ό Inflearnμ΄ ν¨κ» μ€λΉν WADIZ νλ©μ μ§μμ λ°μμ μλμμ΅λλ€. μλ λͺ©λ‘μ λν κ°μλ₯Ό κ°λ°ν μμ μ λλ€.
- λ°μ΄ν° κ³Όνμ μν νμ΄μ¬ μ λ¬Έ - κ°λ° μλ£
- Machnine Learning from Scratch with Python Part I - λ³Έκ³Όμ
- Machnine Learning from Scratch with Python Part II
λν κΈ°μ‘΄ K-MOOC κ³Όμ μ μλ λͺ©λ‘μ μ°Έκ³ νμκΈ° λ°λλλ€.
- κ°μ’λͺ : λ°λ°λ₯ λΆν° μμνλ λ¨Έμ λ¬λ μ λ¬Έ(Machine Learning from Scratch with Python)
- κ°μμλͺ : κ°μ²λνκ΅ μ°μ κ²½μ곡νκ³Ό μ΅μ±μ² κ΅μ ([email protected], Director of TEAMLAB)
- Facebook: Gachon CS50
- Email: [email protected]
- λ³Έ κ³Όμ μ λ¨Έμ λ¬λμ λν κΈ°μ΄κ°λ κ³Ό μ£Όμ μκ³ λ¦¬μ¦λ€μ λν΄ μ΄ν΄νκ³ κ΅¬ννλ κ²μ λͺ©μ μΌλ‘ ν¨
- λ³Έ κ³Όμ μ ν΅ν΄ μκ°μλ λ°μ΄ν° κ³Όνμμ μ¬μ©λλ λ€μν μ©μ΄μ λν κΈ°λ³Έμ μΈ μ΄ν΄λ₯Ό ν μ μμ
- λ³Έ κ³Όμ μ κΈ°λ³Έμ μΈ κ΅¬μ±μ μκ³ λ¦¬μ¦μ λν μ€λͺ
,
Numpyλ₯Ό μ¬μ©ν ꡬν
,Scikit-Learnμ μ¬μ©ν ν¨ν€μ§ νμ©
μΌλ‘ μ΄λ£¨μ΄ μ Έ μμ - μκ°μλ λ¨Έμ λ¬λμμ μ£Όλ‘ μ¬μ©λλ μκ³ λ¦¬μ¦μ ꡬννκΈ° μν΄ κ³ λ±νκ΅ μμ€μ ν΅κ³νκ³Ό μ νλμνμ μ΄ν΄κ° νμν¨
- μκ°μλ λ³Έ κ³Όμ μ ν΅ν΄ Numpy, Pandas, Matplotlib, Scikit-Learn λ± λ°μ΄ν° λΆμμ μν κΈ°λ³Έμ μΈ νμ΄μ¬ ν¨ν€μ§λ₯Ό μ΄ν΄νκ²λ¨
- Course Overvier = κ°μμμ
- Machine learning overview -
κ°μμμ, κ°μμλ£ - An understanding of the data keywords - κ°μμμ, κ°μμλ£
- How to learn machine learning - κ°μμμ, κ°μμλ£
- Types of machine learning - κ°μμμ, κ°μμλ£
- Data era: In a perspective of business - κ°μμμ, κ°μμλ£
- Environment setup
- Python ecosystem for machine learning - κ°μμμ, κ°μμλ£
- How to use Jupyter Notebook - κ°μμμ, κ°μμλ£
- μ°Έκ³ μλ£
- κ°μνκ²½κ³Ό Package νμ©νκΈ° - κ°μμμ, κ°μμλ£
- Chapter Intro - κ°μμμ, κ°μμλ£λͺ¨μ, μ½λ
- The concepts of a feature - κ°μμμ, κ°μμλ£
- Data types - κ°μμμ, κ°μμλ£
- Loading data with pandas - κ°μμμ, κ°μμλ£
- Representing a model with numpy - κ°μμμ, κ°μμλ£
- Lab: Simple Linear algebra concepts -
[κ°μμμ](), κ°μμλ£ - Lab: Simple Linear algebra codes -
κ°μμμ, κ°μμλ£ - Assignment: Linear algebra with pythonic code - PDF, κ°μμλ£
- Chapter Intro - κ°μμμ, κ°μμλ£, κ°μμ½λ, μ½λλ€μ΄λ‘λ
- Numpy overview - κ°μμμ
- ndarray - κ°μμμ
- Handling shape - κ°μμμ
- Indexing & Slicing - κ°μμμ
- Creation functions - κ°μμμ
- Opertaion functions - κ°μμμ
- Array operations - κ°μμμ
- Comparisons - κ°μμμ
- Boolean & fancy Index - κ°μμμ
- Numpy data i/o - κ°μμμ
- Assignment: Numpy in a nutshell - PDF, κ°μμλ£
- TF-KR 첫 λͺ¨μ: Zen of NumPy - λ°νμλ£, κ°μμμ (νμ±μ£Ό, 2016)
- Chapter Intro - κ°μμμ, κ°μμλ£, κ°μμ½λ, μ½λλ€μ΄λ‘λ
- Pandas overview - κ°μμμ
- Series - κ°μμμ
- DataFrame - κ°μμμ
- Selection & Drop - κ°μμμ
- Dataframe operations - κ°μμμ
- lambda, map apply - κ°μμμ
- Pandas builit-in functions - κ°μμμ
- Lab Assignment: Build a matrix - PDF, κ°μμλ£
- Chapter Intro - κ°μμλ£, κ°μμ½λ, μ½λλ€μ΄λ‘λ
- Groupby I - κ°μμμ
- Groupby II - κ°μμμ
- Casestudy - κ°μμμ
- Pivot table & Crosstab - κ°μμμ
- Merge & Concat - κ°μμμ
- Database connection & Persistance - κ°μμμ
- Chapter overview - κ°μμλ£, κ°μμ½λ, μ½λλ€μ΄λ‘λ
- Matplotlib overview - κ°μμμ, κ°μμλ£
- Basic functions & operations - κ°μμμ
- Graph - κ°μμμ
- Matplotlib with pandas - κ°μμμ
- Data Cleaning Problem Overview - κ°μμμ, κ°μμλ£
- Missing Values - κ°μμμ
- Categorical Data Handling - κ°μμμ
- Feature Scaling - κ°μμμ
- Casestudy - KagglepProblems - κ°μμμ, κ°μμλ£
- Chapter overview - κ°μμλ£, κ°μμ½λ, μ½λλ€μ΄λ‘λ
- Linear regression overview -
κ°μμμ - Cost functions -
κ°μμμ - Normal equation -
κ°μμμ - Lab Assignment: Normal equation - PDF, κ°μμλ£
- Gradient descent approach -
κ°μμμ - Linear regression wtih gradient descent -
κ°μμμ - Linear regression implementation wtih Numpy -
κ°μμμ - Multivariate linear regression models -
κ°μμμ - Performance measure for a regression model -
κ°μμμ - Linear regression implementation wtih scikit-learn -
κ°μμμ - Lab Assignment: Gradient descent - PDF, κ°μμλ£
- Chapter overview - κ°μμλ£, κ°μμ½λ, μ½λλ€μ΄λ‘λ
- Stochastic gradient descent -
κ°μμμ - SGD implementation issues -
κ°μμμ - Overfitting and regularization overview -
κ°μμμ - Regularization - L1, L2 -
κ°μμμ - sklearn Linear Model family -
κ°μμμ - Polynomial regression -
κ°μμμ - Sampling method -
κ°μμμ - Kaggle project : Bike demand -
κ°μμμ
- Chapter overview -
κ°μμμ, κ°μμλ£, κ°μμ½λ - Logistic regression overview -
κ°μμμ - Sigmoid function -
κ°μμμ - Cost function -
κ°μμμ - Logistic regression implementation with numpy -
κ°μμμ - Maximum Likelyhood Estimation -
κ°μμμ - Logistic regression with scikit-learn -
κ°μμμ - Confusion matrix -
κ°μμμ - Performance metrics for classification -
κ°μμμ - ROC curve & AUC -
κ°μμμ
- Chapter overview - κ°μμλ£, κ°μμ½λ
- Multiclass Classification overview -
κ°μμμ - Softmax function #1 -
κ°μμμ - Softmax function #2 -
κ°μμμ - Softmax regression with numpy -
κ°μμμ - Performance metrics for classification -
κ°μμμ - Multiclass classification with scikit-learn -
κ°μμμ
- Chapter overview -
κ°μμμ, κ°μμλ£, κ°μμ½λ - Probability overview -
κ°μμμ - Bayes' theorem -
κ°μμμ - Single variable bayes classifier -
κ°μμμ - Navie bayesian Classifier -
κ°μμμ - NB Classifier Implementation -
κ°μμμ - Multinomial Naive Bayes -
κ°μμμ - Gaussian Naive Bayes -
κ°μμμ - NB classifier with sklearn -
κ°μμμ - 20news group classifaication -
κ°μμμ#1,#2
- Text hadnling Lab: News categorization -
κ°μμμ#1,#2, κ°μμλ£
- Chapter overview -
κ°μμμ, κ°μμλ£, κ°μμ½λ - Decision tree overview -
κ°μμμ - The concept of entropy -
κ°μμμ - The algorithme of growing decision tree -
κ°μμμ - ID3 & Information gain -
κ°μμμ - CART & Gini Index -
κ°μμμ - Tree pruning -
κ°μμμ - Decision Tree with sklearn -
κ°μμμ - Handling a continuous attribute -
κ°μμμ - Decision Tree for Regression -
κ°μμμ
- Chapter intro - κ°μμμ, κ°μμλ£, κ°μμ½λ
- Ensemble model overview -
κ°μμμ - Voting classifier -
κ°μμμ - Bagging -
κ°μμμ - Random Forest -
κ°μμμ - AdaBoost -
κ°μμμ - Gradient boosting -
κ°μμμ#1,μμ#2 - XGBoost & LightGBM -
κ°μμμ - Installation guide on Windows -
κ°μμμ, κ°μμλ£ - Stacking -
κ°μμμ
- Overview - κ°μμμ, κ°μμλ£, κ°μμ½λ
- Feature Engineering #1: Generation - κ°μμμ
- Feature Engineering #2: Statics - κ°μμμ
- Feature Engineering #3: Model based - κ°μμμ
- Feature Engineering #4: Iterative - κ°μμμ
- Imbalanced dataset #1 - κ°μμμ
- Imbalanced dataset #2 - κ°μμμ
- Hyperparmeter searching with Distributed Machines - κ°μμμ
- AutoML - κ°μμμ
- Machine Learning (Couera) by Andrew Ng
- λͺ¨λλ₯Ό μν λ₯λ¬λ by Sung Kim
- C++λ‘ λ°°μ°λ λ₯λ¬λ by Sung Kim
- Machine Learning From Scratch[https://github.com/eriklindernoren/ML-From-Scratch]
- Reading materials
- λ°λ°λ₯λΆν° μμνλ λ°μ΄ν° κ³Όν(μ‘°μ 그루μ€, 2016)
- νμ΄μ¬ λ¨Έμ λ¬λ(μΈλ°μ€ν°μ λΌμμΉ΄, 2016)
- Hands-On Machine Learning with Scikit-Learn and TensorFlow(AurΓ©lien GΓ©ron, 2017, PDF)
- Data Mining: Concepts and Techniques(Jiawei Han, Micheline Kamber and Jian Pei , 2011, PDF)
- Supplementary textbooks
- νμ΄μ¬ λΌμ΄λΈλ¬λ¦¬λ₯Ό νμ©ν λ°μ΄ν° λΆμ(μ¨μ€ λ§₯ν€λ, 2013)
- λ¨Έμ λ¬λ μΈ μ‘μ (νΌν° ν΄λ§ν΄, 2013)
- λ°μ΄ν° κ³Όν μ λ¬Έ(λ μ΄μ² μνΈ | μΊμ μ€λ, 2014)
- λ¨Έμ λ¬λ μΈ νμ΄μ¬(λ§μ΄ν΄ 보μΈμ¦, 2015)
- λ¨Έμ λ¬λ μ΄λ‘ μ λ¬Έ(λμΉ΄μ΄ μμΈ μ§, 2016)
- μ
λ¬Έ μμ€μ ν΅κ³ν
- μΈμμμ κ°μ₯ μ¬μ΄ ν΅κ³ν(κ³ μ§λ§ νλ‘μ ν€, 2009)
- μΈμμμ κ°μ₯ μ¬μ΄ λ² μ΄μ¦ν΅κ³νμ λ¬Έ(κ³ μ§λ§ νλ‘μ ν€, 2017)
- νλ₯ κ³Όν΅κ³(νμλνκ΅ μ΄μν κ΅μ, 2014)
- Reading Materials: Data Science from Scratch - Ch.5, Ch.6, Ch.7
- κ³ κ΅ μ΄κ³Ό μμ€μ μ νλμν (Matrixμ Vectorμ κΈ°λ³Έκ°λ
μ Review νμ)
- Essence of linear algebra(3Blue1Brown, 2017)
- Linear Algebra(Khan Academy)
- μ νλμν(νμλ μ΄μν κ΅μ, 2013) - Advance Course
- Reading Materials - Data Science from Scratch - Ch.4
- κ³ κ΅ μ΄κ³Ό μμ€μ λ―Έμ λΆν (κ°λ
μ λν μ΄ν΄ νμ)
- Essence of calculus(3Blue1Brown, 2017)
- νμ΄μ¬ κΈ°μ΄
- λ°μ΄ν° κ³Όνμ μν νμ΄μ¬ μ λ¬Έ (TEAMLAB, 2017)
- Git
- Pro Git (μ€μΊ μ€μ½ | λ²€ μ€νΈλΌμ, 2016)
- Git & Github (TEAMLAB, 2016)
- Git κ°μ (μνμ½λ©, 2014)
λ¨Έμ λ¬λ, λ¨Έμ λ¬λ κ°μ’, λ¨Έμ λ¬λ μ λ¬Έ, Machine Leaning μ λ¬Έ, Machione Learning κ°μ’, λ¨Έμ λ¬λ κ°μ, λ¨Έμ λ¬λ MOOC, Linear Regression, μ ννκ·, μμλΈ, λ₯λ¬λ, λ΄λ΄λ·, νμ΄μ¬, python, νμ΄μ¬ μ λ¬Έ, νμ΄μ¬ κ°μ’, Python μ λ¬Έ, νμ΄μ¬ κ°μ’, νμ΄μ¬ κ°μ, Python κ°μ, Python MOOC, λ°μ΄ν°λ§μ΄λ, Data mining, κ°μ²λ μ΅μ±μ² , μ΅μ±μ² κ΅μ, νλ‘κ·Έλλ° μ λ¬Έ, νλ‘κ·Έλλ° κ°μ’, μ½λ©, μ½λ© μ λ¬Έ, λ°μ΄ν° κ³Όν, λ°μ΄ν° μ¬μ΄μΈμ€, Data science