martinferianc / patternrecognition-eie4

Pattern recognition techniques done as a coursework for EE4-68

👋 Hi there! I'm Martin

Website • Google Scholar • LinkedIn

---

Martin obtained an MEng in Electronic and Information Engineering from Imperial College London, London, UK in 2015. He is currently a PhD candidate in the Department of Electronic and Electrical Engineering at University College London. His research interests include Bayesian neural networks, deep learning , hardware acceleration and confidence calibration. He has hands-on experience from industrial/academic placements in different countries.

patternrecognition-eie4's Issues

Coursework 1

• Confusion matrix for the best result in q2
• Confusion matrices for the best results in q3
• Confusion matrix for q3b ensemble learning best case
• Talk about the rank and dimensionality of S_w and S_b
• Example success and failure case for q2a
• Ensemble learning
• Evaluation & Future work
• Comment code
• Time/memory relationships and computational trade-off for q2
• Q1 Pros and cons of each method

MARTIN

ALEX

• edit all equations in section 3.1 to be latex equations
• get between class seperation before and after LDA
• update for new optimisation problem in section 3.1
• explain need for PCA in section 3.2
• add table for results: bagging, parameter randomisation. And averaging + majority vote for both.

CW2 To-Do List:

ALEX

• NCA/RCA
• Neural Network

MARTIN
-[ ] Neural Network

• Evaluation & Discussion
• Add graphs to the report and confusion matrices
• Change the README.md.