@bchevalier,

Would you be up for helping me better document this library's functionality?
No rush at all, but I haven't done a great job of documenting what is in this library or how to use it very well. I figure we can kick around what all to include in this thread until we're comfortable moving over pieces to the readme.

For instance, I think you've made it possible to have free-variables (no

"duals" should be "dual" below (from p. 21 of the documentation):

## Not run:
library(magrittr)
result <- MIPModel() %>%
add_variable(x[i], i = 1:5) %>%
add_variable(y[i, j], i = 1:5, j = 1:5) %>%
add_constraint(x[i] >= 1, i = 1:5) %>%
set_bounds(x[i], lb = 3, i = 1:3) %>%
set_objective(0) %>%
solve_model(with_ROI("glpk"))
solution <- get_solution(result, x[i])
solution2 <- get_s

Scipy has a LP solver

It would be nice to have integration with it. Even nicer would be to be able to generate the simplex matrices so that research would be immediate.

BR,
Santiago

The title says it: We need a proper way of citing the tool and the documentation.
Currently, the citation in Zenodo includes pseudonames that are not worthy of use in journals, imho.

• Sexp format
• Human-readable, standard format writer
• MPS format
  • reader
  • writer
• Other solver formats (eg http://lpsolve.sourceforge.net/5.1/formulate.htm#Read%20the%20model%20from%20an%20ASCII%20file)

When calling sum on an empty vector, I got the panic from here:

https://github.com/jcavat/rust-lp-modeler/blob/ca7a3cd6ac30ed715a900c5049bc320f6328ad1d/src/dsl/variables.rs#L451

why can't we return 0 when the vector is empty?

Write some welcoming content on first page of docs ./docs/index.rst.

The binaries provided by cddlib currently don't have man pages that explain their usage. This keeps them from being exposed in /usr/bin/ on Debian, see https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=580890.

It would be nice if we could make the output of --help and --version comply with help2man so we could generate man pages automatically.

As cplex is a commercial software, maybe you could add a config option to choose scipy instead cplex for optimizacion.

just like https://github.com/wardbradt/peregrine

Thank you

Add a vignette(s) describing:

• what the maximum covering location problem is
• how to use maxcovr
• interpreting output and generating summaries and plots
• how to use maxcovr with cross validation
• Using commercial solvers such as Gurobi