| Title: | Quickly construct and rate large binary pairwised comparisons |
|---|---|
| Description: | A collection of functions for constructing large pairwised comparisons and rating them using Elo rating system with supporting parallel processing. The method of random sample pairs is based on Reservoir Sampling proposed by JVitter (1985) <doi:10.1145/3147.3165>. |
| Authors: | Xiaohao Yang [aut, cre, cph] |
| Maintainer: | Xiaohao Yang <[email protected]> |
| License: | GPL-3 |
| Version: | 1.0 |
| Built: | 2025-01-24 04:10:34 UTC |
| Source: | https://github.com/billbillbilly/quickpwcr |
ramdonly pair players using reservoir sampling method.
m_elo( df_pw, wl, elo_randomisations = 500, initial_rating = 0, k = 100, cores = 1 )m_elo( df_pw, wl, elo_randomisations = 500, initial_rating = 0, k = 100, cores = 1 )
df_pw |
dataframe, indicating a dataframe that includes the columns of winners and losers. |
wl |
vector, indicating the column names of winners and losers (c('w', 'l')). The first column name is for winners and the second column name is for losers. |
elo_randomisations |
numeric, indicating the number of interactions that the Elo rating system is run with the randomized pairwise comparisons |
initial_rating |
The initial rating of the players at the beginning |
k |
numeric, the K-factor determines the amount of change to the updated ratings |
cores |
numeric, indicating the number of CUP cores to be used for parallel processing |
dataframe
Glickman, M. E., & Jones, A. C. (1999). Rating the chess rating system. CHANCE-BERLIN THEN NEW YORK-, 12, 21-28.
df <- data.frame(a = c(1,6,0,4,'a','v',9,'n'), b = c('w',3,5,2,'d','j',8,'p')) pw <- quickPWCR::m_elo(df_pw = df, wl = c('a', 'b'), elo_randomisations = 100, initial_rating = 1000, k = 100, cores = 1)df <- data.frame(a = c(1,6,0,4,'a','v',9,'n'), b = c('w',3,5,2,'d','j',8,'p')) pw <- quickPWCR::m_elo(df_pw = df, wl = c('a', 'b'), elo_randomisations = 100, initial_rating = 1000, k = 100, cores = 1)
ramdonly pair players using reservoir sampling method.
randompair(players, k, cores = 1)randompair(players, k, cores = 1)
players |
vector, indicating a list of players. |
k |
numeric, indicating how many players each player will be paired with. |
cores |
numeric, indicating the number of CUP cores to be used for parallel |
dataframe
JVitter, J. S. (1985). Random sampling with a reservoir. ACM Transactions on Mathematical Software (TOMS), 11(1), 37-57.
players <- c(1, 'a', 'c', 4, 7, 2, 'w', 'y', 3, 0, 8) pw <- quickPWCR::randompair(players = players, k = 3)players <- c(1, 'a', 'c', 4, 7, 2, 'w', 'y', 3, 0, 8) pw <- quickPWCR::randompair(players = players, k = 3)