Machine Learning (ML) Tutorial -ML as a tool for Scientists

機械学習を研究室のメンバーにインストラクションする会を企画した時の資料です．中身はほとんどありませんのでご注意ください．
経緯はこちら．https://link.medium.com/Xq3gXhwVd5

Transcript

1. Machine Learning (ML) Tutorial
   - ML as a tool for Scientists
   200316
   @oyoroco
   

   View Slide

2. Machine Learning (ML)
   Output
   ● Prediction
   Input
   ● Data
   

   View Slide

3. Machine Learning (ML)
   Output
   ● Prediction
   ● Score
   Input
   ● Data (features)
   ● Task
   ○ Target
   ○ Regression or
   Classification
   ○ Metrics
   

   View Slide

4. ML as a tool
   Output
   ● Prediction
   ● Score
   Input
   ● Data (features)
   ● Task
   ○ Target
   ○ Regression or
   Classification
   ○ Metrics
   Auto ML
   1. Define a good task
   2. Validate results
   3. Accelerate your work!
   

   View Slide

5. Today’s goal
   1. Make a simple ML model
   2. Help your decision whether machine learning
   can be used for your works
   

   View Slide

6. Ref. https://gihyo.jp/book/2019/978-4-297-10843-4
   ML flow chart
   1. Exploratory Data Analysis (EDA)
   2. Task & Metrics
   3. Feature engineering
   4. Modeling
   5. Model validation
   6. (Model tuning)
   7. (Ensemble)
   

   View Slide

7. 1. Exploratory Data Analysis (EDA)
   ● Contents of the data
   ● Prediction target
   

   View Slide

8. 2. Task & Metrics
   Task
   ● Regression
   ● Classification
   ○ Binary
   ○ Multi
   Evaluation metrics
   ● Regression
   ○ Root mean square error: RMSE
   ○ Mean square error: MAE
   ● Classification
   ○ Confusion matrix
   ○ Log loss
   ○ AUC
   Objective function
   ● Regression
   ○ RMSE
   ● Classification
   ○ Log loss cf.) Gradient descent, Differentiable
   

   View Slide

9. 3. Feature engineering
   ● Missing values (NaN)
   ● Standardization (for regularization)
   ● Categorical features
   ○ One-hot encoding
   ○ Label encoding
   ● Dimension reduction
   ○ Principal component analysis: PCA
   

   View Slide

10. 4. Modeling
    ● Choose model (& hyper parameters)
    ■ model = Model(params)
    ● Training
    ○ Data (Features), Target
    ■ model.fit(train_x, train_y)
    ● Prediction
    ○ Test data
    ■ pred = model.predict(test_x)
    

    View Slide

11. 4. Modeling: Model
    ● Linear model
    ● k-nearest neighbor algorithm: kNN
    ● Random forest
    ○ Decision tree + bagging
    ● Neural network: NN
    ● Gradient boosting decision tree: GBDT
    ○ Decision tree + Gradient boosting
    

    View Slide

12. 5. Model validation
    Training data Test data
    Validation data
    All data
    https://scikit-learn.org/stable/modules/cross_validation.html
    

    View Slide

13. Example:
    ● 当日はここから実際の実験データを題材に，スライドとnotebookで
    実際にコードを動かしながらデモをしました．
    ● タスクは2値分類で，評価指標はAUC．
    ● データ数~1600, 特徴量~200
    ● モデルはlogistic回帰とLGBM
    ● 流れは，EDA -> タスクの設定(+AUCの説明) -> モデリング(logistic
    回帰の説明) -> train/predictののち，結果を踏まえ，もう一度EDA
    をしてから特徴量エンジニアリング -> 実験を行いました．
    

    View Slide

14. ML as a tool
    1. Define a good task
    2. Validate results
    3. Accelerate your work!
    

    View Slide