tpe
This crate provides a hyperparameter optimization algorithm using TPE (Tree-structured Parzen Estimator).
Examples
Minimize the result of a quadratic function
An example optimizing a simple quadratic function which has one numerical and one categorical parameters.
use SeedableRng as _;
let choices = ;
let mut optim0 =
new;
let mut optim1 =
new;
let mut best_value = INFINITY;
let mut rng = from_seed;
for _ in 0..100
assert_eq!;
kurobako
benchmark
There is an example examples/tpe-solver.rs which implements
the kurobako
solver interface, so you can run a benchmark using TPE as follows:
$ PROBLEMS=$(kurobako problem-suite sigopt auc)
$ SOLVERS="$(kurobako solver command -- cargo run --release --example tpe-solver) $(kurobako solver optuna)"
$ kurobako studies --solvers $SOLVERS --problems $PROBLEMS --repeats 30 --budget 80 | kurobako run > result.json
$ cat result.json | kurobako report > report.md
References
Please refer to the following papers about the details of TPE: