Code

Sparse2#686

Code

Open

kartikv wants to merge 42 commits intoflintlib:mainflintlib/flint:mainfrom kartikv:sparse2kartikv/flint:sparse2Copy head branch name to clipboard

+26,408-244

Commits

Commits on May 24, 2021

• Another lost function

  

  kartikv

  authored and

  root

  committed
  3df6e6f

  Copy full SHA for 3df6e6f

• Initial code for sparse matrices mod limb size integers, just construction and arithmetic for starters

  

  kartikv

  authored and

  root

  committed
  320ed32

  Copy full SHA for 320ed32

• Basic lanczos, assumes non-symmetric matrix, returns any pseudosolution (x s.t. AtAx = Ab)

  

  kartikv

  authored and

  root

  committed
  002ed93

  Copy full SHA for 002ed93

• Added sparse vector class to nmod, changed sparse matrix class to use it for underlying, added (untested) LU decomposition

  

  kartikv

  authored and

  root

  committed
  31cbf65

  Copy full SHA for 31cbf65

• change #cmakedefine to #cmakedefine01 for HAVE_CPU_SET_T

  

  mahrud

  authored and

  root

  committed
  07b61f2

  Copy full SHA for 07b61f2

• Rebased and merged

  

  kartikv

  authored and

  root

  committed
  05b52a8

  Copy full SHA for 05b52a8

• Added sparse vector class to nmod, changed sparse matrix class to use it for underlying, added (untested) LU decomposition

  

  kartikv

  authored and

  root

  committed
  8f02984

  Copy full SHA for 8f02984

• Spacing and cuddling fixed

  

  kartikv

  authored and

  root

  committed
  6de1379

  Copy full SHA for 6de1379

• Lost a function in the last commit, fixed

  

  kartikv

  authored and

  root

  committed
  18f39a0

  Copy full SHA for 18f39a0

• Another lost function

  

  kartikv

  authored and

  root

  committed
  693533b

  Copy full SHA for 693533b

• Working LU code (with test) and direct solving

  

  kartikv

  authored and

  root

  committed
  070ec4d

  Copy full SHA for 070ec4d

• Added code to construct reduced row echelon form and associated solver, another construction option, and (untested) inversion code

  

  kartikv

  authored and

  root

  committed
  8111815

  Copy full SHA for 8111815

• Now with additional utilities, more correct basic functions, and nullspace and inversion functions

  

  kartikv

  authored and

  root

  committed
  502acdf

  Copy full SHA for 502acdf

• Added nullvector functions for Lanzcos, everything for basic Wiedemann

  

  kartikv

  authored and

  root

  committed
  ed47162

  Copy full SHA for ed47162

• Created sparse vector and matrix utilities for all the fq variants

  

  kartikv

  authored and

  root

  committed
  2cb2ea8

  Copy full SHA for 2cb2ea8

• Fixed spacing

  

  kartikv

  authored and

  root

  committed
  69a9be7

  Copy full SHA for 69a9be7

• Fixed bug in multiply, block Lanczos now works for solving (but not for nullspace)

  

  kartikv

  authored and

  root

  committed
  23d2c4e

  Copy full SHA for 23d2c4e

• Now with working block Wiedemann (and nullspace for both block methods)

  

  kartikv

  authored and

  root

  committed
  51b5584

  Copy full SHA for 51b5584

• Added rst documentation for sparse vectors and matrices

  

  kartikv

  authored and

  root

  committed
  cd80b7a

  Copy full SHA for cd80b7a

• Most of the desired code for sparse integer linear algebra, except for smith normal form

  

  kartikv

  authored and

  root

  committed
  2e35dec

  Copy full SHA for 2e35dec

• Fixed spacing issues

  

  kartikv

  authored and

  root

  committed
  7323a4b

  Copy full SHA for 7323a4b

• Cleaned up some tests

  

  kartikv

  authored and

  root

  committed
  8c07eec

  Copy full SHA for 8c07eec

• Changed mods -> smod, need to change function name later

  

  kartikv

  authored and

  root

  committed
  f8f4b76

  Copy full SHA for f8f4b76

• Initial code for sparse matrices mod limb size integers, just construction and arithmetic for starters

  

  kartikv

  authored and

  root

  committed
  e0448b7

  Copy full SHA for e0448b7

• Added sparse vector class to nmod, changed sparse matrix class to use it for underlying, added (untested) LU decomposition

  

  kartikv

  authored and

  root

  committed
  c451471

  Copy full SHA for c451471

• Rebased and merged

  

  kartikv

  authored and

  root

  committed
  3907bdc

  Copy full SHA for 3907bdc

• Added sparse vector class to nmod, changed sparse matrix class to use it for underlying, added (untested) LU decomposition

  

  kartikv

  authored and

  root

  committed
  d618a89

  Copy full SHA for d618a89

• Spacing and cuddling fixed

  

  kartikv

  authored and

  root

  committed
  5780424

  Copy full SHA for 5780424

• Lost a function in the last commit, fixed

  

  kartikv

  authored and

  root

  committed
  70ae8c1

  Copy full SHA for 70ae8c1

• Added code to construct reduced row echelon form and associated solver, another construction option, and (untested) inversion code

  

  kartikv

  authored and

  root

  committed
  83d12cf

  Copy full SHA for 83d12cf

• Now with additional utilities, more correct basic functions, and nullspace and inversion functions

  

  kartikv

  authored and

  root

  committed
  f7e1519

  Copy full SHA for f7e1519

• Added nullvector functions for Lanzcos, everything for basic Wiedemann

  

  kartikv

  authored and

  root

  committed
  36cef5f

  Copy full SHA for 36cef5f

• Created sparse vector and matrix utilities for all the fq variants

  

  kartikv

  authored and

  root

  committed
  e2bc114

  Copy full SHA for e2bc114

• Fixed spacing

  

  kartikv

  authored and

  root

  committed
  559f691

  Copy full SHA for 559f691

• Fixed bug in multiply, block Lanczos now works for solving (but not for nullspace)

  

  kartikv

  authored and

  root

  committed
  ff58185

  Copy full SHA for ff58185

• Now with working block Wiedemann (and nullspace for both block methods)

  

  kartikv

  authored and

  root

  committed
  7638bff

  Copy full SHA for 7638bff

• Added rst documentation for sparse vectors and matrices

  

  kartikv

  authored and

  root

  committed
  58cee49

  Copy full SHA for 58cee49

• Most of the desired code for sparse integer linear algebra, except for smith normal form

  

  kartikv

  authored and

  root

  committed
  aae4174

  Copy full SHA for aae4174

• Fixed spacing issues

  

  kartikv

  authored and

  root

  committed
  fd54b32

  Copy full SHA for fd54b32

• Added copyrights, resizing, and some asserts

  

  root

  authored and

  root

  committed
  8c6b44a

  Copy full SHA for 8c6b44a

• Removed some duplicate functions, re-added templates to cmake list

  

  root

  authored and

  root

  committed
  af5c2b5

  Copy full SHA for af5c2b5

Commits on May 26, 2021

• Added fmpz_mod_mat to cmakelists

  

  root

  authored and

  root

  committed
  6175549

  Copy full SHA for 6175549