stitches aims to empower everyone to solve mathematics problems with the most efficient use of the hardware they have available.
Create a new rust binary.
cargo new --bin <project_name>
Add the stitches library to Cargo.toml.
[dependencies]
stitches "*"
Copy the null_problem
example to your main.rs file.
cp /path/to/stitches/examples/null_problem.rs src/main.rs
Run with --release
.
cargo run --release
The goal of stitches can be more technically stated as putting your candidate checking in the hot path. We only use monomorphization to maximize the percent of CPU time spent calculating your problem.
If your problem requires searching a space we didn't think about, you can implement your own Space
struct and use the same infrastructure as our out-of-the-box spaces.
We're even planning to include sanity and performance checks to let you know if your implementation may have problems.
[x] Generic searching [x] Basic performance statistics [ ] Sanity/performance checks for Space implementation before starting search [ ] Save/resume search state to/from disk [ ] Multi-machine networked searching [ ] Time to completion for finite spaces
[x] Linear [ ] Tree [ ] Multidimensional linear