SLIDE 1
Multi-Class Logistic Regression 11/07/2018
Liyuan Liu Ph.D. Students in Analytics and Data Science Kennesaw State University
5-cross validation ROC plot Multi-class logistic regression
Multi-Class Logistic Regression 11/07/2018 Liyuan Liu Ph.D. - - PowerPoint PPT Presentation
Multi-Class Logistic Regression 11/07/2018 Liyuan Liu Ph.D. - - PowerPoint PPT Presentation
Multi-Class Logistic Regression 11/07/2018 Liyuan Liu Ph.D. Students in Analytics and Data Science Kennesaw State University Multi-class logistic regression 5-cross validation ROC plot Process Data Preparation Softmax Function Gradient
SLIDE 2
SLIDE 3
3
Process
Data Preparation Softmax Function Gradient Descent Model Training and testing Add L1 Regularization
SLIDE 4
4
Data Preparation
- 1. Separate raw data to X and Y.
- 2. Add Intercept.
- 3. Normalized X use min-max method.
- 4. One Hot EncodedY.
SLIDE 5
5
Softmax
1/(1+np.exp(-score)) (np.exp(score) / np.sum(np.exp(score))
SLIDE 6
6
Gradient Descent
Learning Rate: 0.01 Epoch: 3000
SLIDE 7
7
Get Prediction Value
Rule: Extract the indexhas the highest probability.
The argmax() only for compute accuracy
SLIDE 8
8
ROC Plot
Use numpy.ravel to flatten the array.¶
SLIDE 9
9
Result-5 Cross Validation
SLIDE 10
10
Result L1 Regularization-5 Cross Validation
Why Regularization? Reduce Over-fitting Problem.
SLIDE 11
11
Result L1 Regularization-5 Cross Validation
SLIDE 12
THANKS
Liyuan Liu: lliyuan@students.kennesaw.edu