Commits

Show all
Author Commit Message Labels Comments Date
Aleksey Khudyakov
Merge branch 'oldmaster' Commit which were missing after fixing problems with HG<->git mirroring Conflicts: .hgtags statistics.cabal
Bookmarks
master
Aleksey Khudyakov
Make cumulative and complCumulative more accurate for Laplace distr.
Mihai Maruseac
Add tests for Laplace Distribution. All tests passed: Tests for: LaplaceDistribution: C.D.F. sanity: [OK, passed 100 tests] CDF limit at +inf: [OK, passed 100 tests] CDF limit at -inf: [OK, passed 100 tests] CDF at +inf = 1: [OK, passed 100 tests] CDF at -inf = 1: [OK, passed 100 tests] CDF is nondecreasing: [OK, passed 100 tests] 1-CDF is correct: [OK, passed 100 tests] Binary OK: [OK, passed 100 …
Mihai Maruseac
Add Laplace Distribution. The Laplace Distribution (also known as double-exponential distribution) arises from Laplace's first error law (the distribution of the absolute deviations from the mean) and its relationship with the Gaussian Distribution (normal distribution) is the same as the relationship between L1 curve fitting (least absolute deviations) and L2 curve fitting (least squares). The distribution is widely used…
Bryan O'Sullivan
Added tag 0.13.2.3 for changeset 4302388f1acb
Bryan O'Sullivan
Bump version to 0.13.2.3
Tags
0.13.2.3
Bryan O'Sullivan
Restore vector 0.10 dependency
Bryan O'Sullivan
Added tag 0.13.2.2 for changeset e9452e9e69ce
Bryan O'Sullivan
Bump version to 0.13.2.2
Tags
0.13.2.2
Bryan O'Sullivan
Downgrade dependency on vector to 0.9
Aleksey Khudyakov
Fix documentation it's plain wrong
Aleksey Khudyakov
Add test for Spearman correlation. We probably need to add some unit tests for it too
Aleksey Khudyakov
Fix bug in ranking algorithm
Aleksey Khudyakov
Add new test for Pearson correlation
Aleksey Khudyakov
Better naming sigma usually corresponds to standard deviation
Aleksey Khudyakov
Remove tests for Pearson's and Spearman's correlation It's not clear how reference values were obtained and test failure provides no insight into nature of failure
Aleksey Khudyakov
To properl normalize correlation we need to use ML variabne
Aleksey Khudyakov
Implement symmetric generate
Aleksey Khudyakov
Add unsafeNew for mutable matrices
Aleksey Khudyakov
Rename function and refactor a bit
Aleksey Khudyakov
Merge modules S.Correpation.{Peasron,Spearman} to S.Correlation
Aleksey Khudyakov
Drop diagRect and implement diag directly
Aleksey Khudyakov
Add function to generate matrices from indices
Aleksey Khudyakov
Check that matrix will be correct in fromRows/fromCols
Aleksey Khudyakov
Refactor toLists/toRows
Aleksey Khudyakov
Reorder code a bit
Aleksey Khudyakov
Use correlation from S.Sample
Aleksey Khudyakov
Style
Aleksey Khudyakov
Refactor rank for unsorted vector
Aleksey Khudyakov
Improve documentation of rank
  1. Prev
  2. Next