Use

To use and test any of the heavily WIP methods, simply execute from the root directory: (At the moment we only are testing things)

	bash MatrixOps/scripts/build_test.sh

And then to execute the code without any parameters (parameters are WIP, also again we only have tests atm) run from root:

	#for vec test class 
	./test_vec

	#for matrix text class
	./test_matrix

    #
    ./unit_test

Current Plans

We decided to not use inheritance, there isn't enough overlap between the classes without just implementing everything as a Matrix, so, there will be no more abstract tensor here.

classDiagram
    Matrix : -size_t row
    Matrix : -size_t col
    Matrix : -vector[vector[T]] matrix_vals
	
    Matrix : +sum()
    Matrix : +diff()
    Matrix : +scale()
    Matrix : +prod()
    Matrix : +norm()
    Matrix : + (Uniform Distribution [a,b] Constructor) Matrix(size_t row, size_t col, T a, T b)
    Matrix : + (Random Integer Constructor on [a,b]) Matrix(size_t row, size_t col, int a, int b)
    Matrix : + (Identity Constructor) Matrix(size_t n)
    Matrix : + (Zero Constructor) Matrix(size_t row, size_t col)
    Matrix : + (2d Vec Constructor) Matrix(vector[vector[T]] info)
	
	
    Matrix : +row()
    Matrix : +col()
    Matrix : +frobenius()
    Matrix : +SVD()
    
    Vector : -size
    Vector : -vector[T] vector_vals
    
    Vector : + (Uniform Distribution [a,b] Constructor) Vector(size_t size, T a, T b)
    Vector : + (Random Integer Constructor on [a,b]) Vector(size_t size, int a, int b)
    Vector : + (Identity Constructor) Vector(size_t size, size_t position)
    Vector : + (Zero Constructor) Vector(size_t size)
    Vector : + (Vec Constructor) Vector(vector[T] info)
	
	Vector : +sum()
    Vector : +diff()
    Vector : +scale()
    Vector : +prod()
    Vector : +norm()
    Vector : +size()
    Vector : +l1_norm()
    Vector : +lp_norm(int p)
    
    
    

Loading

Motivation

I wanted to provide an implementation of a lot of the algorithms I saw in my NLA course for fun with numpy. I hope to go further with a lot of the code here and potentially extend it to some lower level code I can put in a python wrapper. I included some TeX notes with everything as well since

---

TODO

Add Householder reflector code, decent matrix multiplication algorithms (could add strassen for fun it's not very good), QR factorization algos and SVD computation as well. Equality operator for vec and matrix, matrix vector products as well