Rust tools for Decision-DNNF formulas, including translation, model counting and model enumeration.
Decdnnf-rs is a tool for manipulating Decision-DNNFs produced by d4. It allows queries such as counting models, searching for models under assumptions, or enumeration of solutions, for which it excels. It also provides a tool for translating formulas generated by d4 into the format generated by c2d.
Decdnnf-rs requires a recent version of the Rust toolchain (>= 1.72.1). See rust-lang.org for more information on how to install Rust.
To build from source, run cargo build --release to compile the binary. It will be set in the target/release directory.
The decdnnf-rs tool expects a subcommand. To get the list, just invoke the command with the help flag.
decdnnf_rs -hIn the same way, you can obtain the list of expected and optional arguments by adding the help flag after the subcommand.
decdnnf_rs model-counting -hSome options are common to most commands, like the ones dedicated to input file and logging level.
Another one of interest is --n-vars.
Since the output format of d4 (which is the default input format of decdnnf_rs) does not provide the number of variables of the problems, this number cannot be deduced if it is more important than the highest variable index in use.
Setting --n-vars allows to override the number of variables returned by the parser, which is set to the highest variable index.
By default, decdnnf_rs partially checks the correctness on the input Decision-DNNF, in particular the decomposability property of the disjunction and the determinism of the disjunction node.
However, these checks may consume some time.
In case you are sure of the correctness of the formula, you can skip these tests by adding the --do-not-check flag.
Use the translation command:
decdnnf_rs translation -i instance.nnfUse the model-counting command:
decdnnf_rs model-counting -i instance.nnfUse the model-enumeration command:
decdnnf_rs model-enumeration -i instance.nnfThis commands admits multiple options allowing to set the number of variables (in case it is higher than the highest index in the input formula), use a compact output or use an enumeration algorithm based on a decision tree.
Run decdnnf_rs model-enumeration -h for more information.
Use the direct-access command:
decdnnf_rs direct-access --index 5 -i instance.nnfThe order of the models is determined by the structure of the formula by default, and may change with new versions of the tool.
Adding the --lexicographic-order flag is an effective way to prevent this.
This flag instructs the tool to sort the models according to the lexicographic order, although this approach does result in a longer processing time.
Use the sampling command:
decdnnf_rs sampling -l 5 -i instance.nnfThe indices associated with the models work in the same way as for direct access.
Decdnnf-rs is developed at CRIL (Univ. Artois & CNRS). It is made available under the terms of the GNU GPLv3 license.