How to turn a binary matrix into a data.frame in R? lpSolveAPI

问题

library(lpSolveAPI)
lprec1 <- make.lp(0,nrow(df) 

  add.constraint(lprec1, c(1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0), "<=", as.numeric(ads1)) 
  add.constraint(lprec1, c(0,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0), "<=", as.numeric(ads2)) 
  add.constraint(lprec1, c(0,0,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0), "<=", as.numeric(ads3)) 
  add.constraint(lprec1, c(0,0,0,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0), "<=", as.numeric(ads4)) 
  add.constraint(lprec1, c(0,0,0,0,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0), "<=", as.numeric(ads5)) 
  add.constraint(lprec1, c(0,0,0,0,0,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0), "<=", as.numeric(ads6)) 
  add.constraint(lprec1, c(0,0,0,0,0,0,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0), "<=", as.numeric(ads7)) 
  add.constraint(lprec1, c(0,0,0,0,0,0,0,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0), "<=", as.numeric(ads8)) 
  add.constraint(lprec1, c(0,0,0,0,0,0,0,0,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0), "<=", as.numeric(ads9)) 
  add.constraint(lprec1, c(0,0,0,0,0,0,0,0,0,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0), "<=", as.numeric(ads10)) 
  add.constraint(lprec1, c(0,0,0,0,0,0,0,0,0,0,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0), "<=", as.numeric(ads11)) 
  add.constraint(lprec1, c(0,0,0,0,0,0,0,0,0,0,0,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0), "<=", as.numeric(ads12)) 
  add.constraint(lprec1, c(0,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0), "<=", as.numeric(ads13)) 
  add.constraint(lprec1, c(0,0,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0), "<=", as.numeric(ads14)) 
  add.constraint(lprec1, c(0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0), "<=", as.numeric(ads15)) 
  add.constraint(lprec1, c(0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0), "<=", as.numeric(ads16)) 
  add.constraint(lprec1, c(0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0), "<=", as.numeric(ads17)) 
  add.constraint(lprec1, c(0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0), "<=", as.numeric(ads18)) 
  add.constraint(lprec1, c(0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0), "<=", as.numeric(ads19)) 
  add.constraint(lprec1, c(0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0,0,0,0,0,0,0,0,0,0,0), "<=", as.numeric(ads20)) 
  add.constraint(lprec1, c(0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0,0,0,0,0,0,0,0,0,0), "<=", as.numeric(ads21)) 
  add.constraint(lprec1, c(0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0,0,0,0,0,0,0,0,0), "<=", as.numeric(ads22)) 
  add.constraint(lprec1, c(0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0,0,0,0,0,0,0,0), "<=", as.numeric(ads23)) 
  add.constraint(lprec1, c(0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0,0,0,0,0,0,0), "<=", as.numeric(ads24)) 
  add.constraint(lprec1, c(0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0,0,0,0,0,0), "<=", as.numeric(ads25)) 
  add.constraint(lprec1, c(0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0,0,0,0,0), "<=", as.numeric(ads26)) 
  add.constraint(lprec1, c(0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0,0,0,0), "<=", as.numeric(ads27)) 
  add.constraint(lprec1, c(0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0,0,0), "<=", as.numeric(ads28)) 
  add.constraint(lprec1, c(0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0,0), "<=", as.numeric(ads29))
  add.constraint(lprec1, c(0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0), "<=", as.numeric(ads30)) 
  add.constraint(lprec1, c(0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,0,0), "<=", as.numeric(ads31))
  add.constraint(lprec1, c(0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,0), "<=", as.numeric(ads32)) 
  add.constraint(lprec1, c(0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1), "<=", as.numeric(ads33)) 
solve(lprec1) 
print(get.variables(lprec1)) 
print(get.objective(lprec1))

I turned all my ads vectors into a single data.frame df$ads, is there any way to transform this matrix too? I tried using

add.constraint(lprec1, diag(nrow (df))), "<=", as.vector(df$ads))
solve(lprec1)

but lpSolveAPI recognizes that length is different:Error in add.constraint(lprec1, diag(nrow(df), "<=", df$ads) : the length of ‘xt’ is not equal to the number of decision variables in the model but there are 33 decision variables and the nrow(df) are 33... Is there any way to binarize without having to make this matrix?

length(diag(nrow(df))) = 361

The length is different sizes, is there any way to turn these vectors into a single data.frame with length = 33?

回答1:

Here is an option, where we create a list of vectors, loop through the sequence of list and assign the constraints

a <- as.vector(diag(5))
lst1 <- asplit(matrix(a, ncol = 5, byrow = TRUE), 1)

library(lpSolveAPI)
lprec1 <- make.lp(0, length(lst1)) 
ads <- c(0, 5, 1, -1, 0)
for(i in seq_along(lst1)) add.constraint(lprec1, lst1[[i]], "<=", ads[i])
solve(lprec1)
#[1] 2

来源：https://stackoverflow.com/questions/59491515/how-to-turn-a-binary-matrix-into-a-data-frame-in-r-lpsolveapi

标签

optimization

lpsolve