README.md

MareepLearning

✏️ My homeworks of NTU EE5184 Machine Learning [2020 spring] (by Prof. Hung-Yi Lee)

• Course website: http://speech.ee.ntu.edu.tw/~tlkagk/courses_ML20.html

• FB Group: https://www.facebook.com/groups/1099602297060276/

• Kaggle: https://www.kaggle.com/c/ml2020spring-hw[1~15]/

• Homework: https://docs.google.com/spreadsheets/d/1nHixDtBX3LOWm9BwwbiC6XE9y4XAP-L_Q4dYZjKbiR8/

• hw1=9 (kaggle overfit), hw2=10+1 (best report), hw3=10+1 (kaggle top5), hw4=10,
  hw5=10, hw6=10+1 (best report), hw7=10, hw9=10, hw11=10.5+1 (best report) = 99.5 (A+)

• Task:

  • Regression: hw1(Re),hw2(Re)
  • Binary Classification: hw2(Re), hw4(RNN),
  • Multi-Class Classification: hw3(CNN), hw9(AE)
  • Generation: hw11(GAN)
  • Technique: hw5(explainable), hw6(adversarialAtk), hw7(Compression), hw10(Anomaly)

• Network:

  • Linear: hw1(Re), hw2(Re)
  • DNN based: hw3(CNN), hw4(RNN)
  • CNN based: hw3(CNN), hw5(explainable), hw6(adversarialAtk), hw7(Compression), hw9(AE), hw10(Anomaly), hw11(GAN)
  • RNN based: hw4(RNN)

• Data:

  • Supervised: hw1(Re), hw2(Re), hw3(CNN), hw4(RNN)
  • Unsupervised: hw5(explainable), hw6(adversarialAtk), hw7(Compression), hw9(AE), hw10(Anomaly), hw11(GAN)