utmi-pattern-information

Assignments of class in UTokyo "Pattern Information".

http://www.mi.t.u-tokyo.ac.jp/harada/lectures/pattern/patternreport20150116.pdf

Report

• View Report
• Download Report

Summary

• Assignment1: Windrow-Hoff Algorithm to get decisive surface.
• Assignment2: Use pseudo-inverse to map data to label.
• Assignment3: K-Nearest Neighbors.
• Assignment4: Probability generation model.
• Challenge1: PCA and FisherLDA.
• Challenge2: LogisticRegression for 2 classes.
• Challenge3: Emotion discrimination.

Codes

Execute below to get output in output/ dir.

Python code

$ cd code
$ python kadai1.py       # Assignment1
$ python kadai2.py       # Assignment2
$ python kadai3_1.py     # Assignment3
$ python kadai3_2.py     # Assignment3
$ python kadai4_1.py     # Assignment4
$ python kadai4_2.py     # Assignment4
$ python kadai4_3.py     # Assignment4
$ python challenge1.py   # Challenge1
$ python challenge2.py   # Challenge2
$ python challenge3.py   # Challenge3

C++ code

$ cd code/cpp
$ make
$ ./kadai1

Euslisp code

$ cd code/euslisp
$ eus kadai1.lisp