Coding the Matrix Resources

This page links to resources for Coding the Matrix. The links are organized by chapter and section; as a consequence, there might be multiple links to the same resource.

Please do not make these materials publicly available elsewhere, and do not make your solutions public. Don't deprive others of the pleasure and challenge of solving the problems, and don't place temptation in the path of others.

The Function (and other mathematical and computational preliminaries)

Lab: Python—modules and control structures—and inverse index

The Field

Playing with ℂ
- plotting.py
- image.py
- img01.png
- png.py Module used by image.py
Playing with GF(2)
- GF2.py

The Vector

Dictionary-based representations of vectors
- vecutil.py
Our implementation of Vec
- vec.py
Solving a triangular system of linear equations
- triangular.py
Lab: Comparing voting records using dot-product
- voting_record_dump109.txt
- US_Senate_voting_data_109.txt This is the same as voting_record_dump109.txt, just with a more informative name
- UN_voting_data.txt A different dataset that can be used for the same purpose
Problems
- vec.py The tests formerly located in test_mat.py have been added to this file as doctests.
- test_vec.py This file is not needed since its doctests have been added to vec.py.

The Vector Space

The Matrix

What is a matrix?
- matutil.py
Matrix-vector and vector-matrix multiplication in terms of linear combinations
- solver.py
Lab: Error-correcting codes
- bitutil.py
Lab: Transformations in 2D geometry
- image_mat_util.py which uses...
- png.py
Problems
- mat.py The tests formerly located in test_mat.py have been added to this file as doctest.
- mat_sparsity.py This file contains some simple doctests that should take very little time if your implementations of transpose, matrix-vector multiplication, vector-matrix multiplication, and matrix-matrix multiplication exploit sparsity.
- UN_voting_data.txt

The Basis

Perspective rendering
- plotting.py
- vecutil.py
Lab: Perspective rectification
- board.png
- image_mat_util.py which uses...
- png.py
- cit.png
Problems
- matutil.py
- solver.py

Dimension

Dimension and rank
- independence.py
- solver.py
The annihilator
- solver.py
Problems
- solver.py
- triangular.py

Gaussian elimination

Gaussian elimination over GF(2)
- gaussian_examples.py
Using Gaussian elimination for other problems
- cracking_rand.py Some code for cracking Python's rand function
- echelon.py
Lab: Threshold Secret-Sharing
- bitutil.py
Lab: Factoring integers
- factoring_support.py

The Inner Product

Orthogonality
- orthogonalization.py
Lab: machine learning The following .data files were derived from data obtained from the University of Wisconsin Hospitals, Madison from Dr. William H. Wolberg.
- train.data File with data on which to train your classifier
- validate.data File with data on which to test your classifier
- cancer_data.py The procedure returns a pair (A, b) consisting of a P-by-D matrix A and a P-vector b where P is a set of patient IDs. This differs slightly from the spec given in this chapter of Edition 0.

Orthogonalization

Projection orthogonal to multiple vectors
- orthogonalization.py
Problems

Special Bases

Discrete Fourier transform
- fft.py
Lab: Using wavelets for compression
- image.py
- png.py used by image.py
- flag.png
- Dali.png

The Singular Value Decomposition

Closest dimension-k vector space
- svd.py
- game_results.csv
Lab: Digits
Lab: Eigenfaces
- faces.zip
- unclassified.zip
- image.py
- png.py used by image.py
- eigenfaces.py

The Eigenvector

Lab: Pagerank

The Linear Program

The simplex algorithm
- simplex.py
Lab: Learning through linear programming
- train.data File with data on which to train your classifier
- validate.data File with data on which to test your classifier
- cancer_data.py The procedure returns a pair (A, b) consisting of a P-by-D matrix A and a P-vector b where P is a set of patient IDs. This differs slightly from the spec given in this chapter of Editions 0 and 1.