JGS594 Lecture 05

Software Engineering for Machine Learning
Coding
(202201)

Transcript

1. jgs
   SER 594
   Software Engineering for
   Machine Learning
   Lecture 05: Assignment 01
   Dr. Javier Gonzalez-Sanchez
   javiergs@asu.edu
   javiergs.engineering.asu.edu | javiergs.com
   PERALTA 230U
   Office Hours: By appointment
   

   View Slide

2. jgs
   Previously …
   

   View Slide

3. Javier Gonzalez-Sanchez | SER 594 | Spring 2022 | 3
   jgs
   Next
   From
   Scratch
   

   View Slide

4. Javier Gonzalez-Sanchez | SER 594 | Spring 2022 | 4
   jgs
   Neural Network
   

   View Slide

5. Javier Gonzalez-Sanchez | SER 594 | Spring 2022 | 5
   jgs
   Code
   https://github.com/javiergs/Medium/blob/main/NeuralNetwork/BasicNeuralNetwork.java
   

   View Slide

6. jgs
   Javier Gonzalez-Sanchez | SER332 | Spring 2018 | 6
   …
   Go for a value < 0.05
   

   View Slide

7. Javier Gonzalez-Sanchez | SER 594 | Spring 2022 | 7
   jgs
   Constructor
   

   View Slide

8. jgs
   Javier Gonzalez-Sanchez | SER332 | Spring 2018 | 8
   

   View Slide

9. Javier Gonzalez-Sanchez | SER 594 | Spring 2022 | 9
   jgs
   Main
   10,000
   4
   1
   

   View Slide

10. Javier Gonzalez-Sanchez | SER 594 | Spring 2022 | 10
    jgs
    Backward
    Forward Propagation
    Error
    Forward Propagation
    Error
    Forward Propagation
    Error
    Main
    Forward
    Propagation
    Error
    MSE
    10,000 times
    Backward
    Backward
    Backward
    Propagation
    

    View Slide

11. Javier Gonzalez-Sanchez | SER 594 | Spring 2022 | 11
    jgs
    Forward Propagation
    

    View Slide

12. jgs
    Javier Gonzalez-Sanchez | SER332 | Spring 2018 | 12
    

    View Slide

13. Javier Gonzalez-Sanchez | SER 594 | Spring 2022 | 13
    jgs
    Backward Propagation (Part 1 of 2)
    

    View Slide

14. Javier Gonzalez-Sanchez | SER 594 | Spring 2022 | 14
    jgs
    Backward Propagation (Part 2 of 2)
    

    View Slide

15. jgs
    Assignment 01
    

    View Slide

16. Javier Gonzalez-Sanchez | SER 594 | Spring 2022 | 16
    jgs
    Instructions
    § Use the code from
    https://github.com/javiergs/Medium/blob/main/NeuralNetwork/BasicNeuralNetwork.java
    § Write a paper addressing the following questions
    

    View Slide

17. Javier Gonzalez-Sanchez | SER 594 | Spring 2022 | 17
    jgs
    Questions
    1. Can you reuse that code and train a neural network for identifying AND
    operator? What is the Error after 10,000 iterations of training?
    2. Can you reuse that code and train a neural network for identifying OR
    operator? What is the Error after 10,000 iterations of training?
    3. Compare with the provided error for XOR.
    Explain the difference or similarities
    

    View Slide

18. Javier Gonzalez-Sanchez | SER 594 | Spring 2022 | 18
    jgs
    Questions
    For the error calculation. You will notice this code
    meanSquareError += Math.pow (err, 2)/2;
    double err1 = meanSquareError / (network.nodesOutputLayer+network.nodesHiddenLayer-1);
    4. Is there a problem there?
    5. Could you identify and explain or fix?
    

    View Slide

19. Javier Gonzalez-Sanchez | SER 594 | Spring 2022 | 19
    jgs
    Questions
    6. What is the size of the array arrayBias?
    Is that, OK?
    Should it or Could it Be smaller?
    If so, can you fix the code? It is worthy to make it smaller.
    

    View Slide

20. Javier Gonzalez-Sanchez | SER 594 | Spring 2022 | 20
    jgs
    Questions
    7. Plot the values of the eight W values
    Describe/notice how they change.
    

    View Slide

21. Javier Gonzalez-Sanchez | SER 594 | Spring 2022 | 21
    jgs
    Questions
    8. Express the source code from lines 64 to 75 as an equation using the
    names on the picture
    9. Express the source code from lines 77 to 88 as equations using the
    names on the picture
    10. Express the source code from lines 90 to 93 as equations using the
    names on the picture
    

    View Slide

22. Javier Gonzalez-Sanchez | SER 594 | Spring 2022 | 22
    jgs
    Extra
    If you are familiar with derivatives,
    § Can you explain the equations from the previous question (where they come
    from)
    Hints:
    o derivative of sigmoid function;
    o partial derivative of error function with respect to Wi
    o Derivative of a constant
    o Derivative of (Ax + Bx + Cx + …)
    

    View Slide

23. Javier Gonzalez-Sanchez | SER 594 | Spring 2022 | 23
    jgs
    Picture
    

    View Slide

24. Javier Gonzalez-Sanchez | SER 594 | Spring 2022 | 24
    jgs
    Questions
    

    View Slide

25. jgs
    Summary
    

    View Slide

26. Javier Gonzalez-Sanchez | SER 594 | Spring 2022 | 26
    jgs
    Data Scientist vs Engineers
    

    View Slide

27. Javier Gonzalez-Sanchez | SER 594 | Spring 2022 | 27
    jgs
    Roles
    

    View Slide

28. Javier Gonzalez-Sanchez | SER 594 | Spring 2022 | 28
    jgs
    Use
    

    View Slide

29. Javier Gonzalez-Sanchez | SER 594 | Spring 2022 | 29
    jgs
    Use
    

    View Slide

30. Javier Gonzalez-Sanchez | SER 594 | Spring 2022 | 30
    jgs
    Use
    

    View Slide

31. jgs
    SER 594 Software Engineering for Machine Learning
    Javier Gonzalez-Sanchez, Ph.D.
    javiergs@asu.edu
    Spring 2022
    Copyright. These slides can only be used as study material for the class CSE205 at Arizona State University.
    They cannot be distributed or used for another purpose.
    

    View Slide