55ccfb9ebc
This parallels what we did for the Irisawa hexlet realization. The kaleidocycle tangent test comes out slightly weaker, because we no longer confirm that the realized configuration matches the initial guess. However, we still confirm that the configuration history only has one entry, which is equivalent as long as the configuration history starts with the initial guess and is updated after every optimization step.
30 lines
No EOL
1.1 KiB
Rust
30 lines
No EOL
1.1 KiB
Rust
use nalgebra::{DMatrix, DVector};
|
|
|
|
use dyna3::engine::{Q, examples::realize_kaleidocycle};
|
|
|
|
fn main() {
|
|
const SCALED_TOL: f64 = 1.0e-12;
|
|
let (config, tangent, success, history) = realize_kaleidocycle(SCALED_TOL);
|
|
print!("Completed Gram matrix:{}", config.tr_mul(&*Q) * &config);
|
|
print!("Configuration:{}", config);
|
|
if success {
|
|
println!("Target accuracy achieved!");
|
|
} else {
|
|
println!("Failed to reach target accuracy");
|
|
}
|
|
println!("Steps: {}", history.scaled_loss.len() - 1);
|
|
println!("Loss: {}\n", history.scaled_loss.last().unwrap());
|
|
|
|
// find the kaleidocycle's twist motion by projecting onto the tangent space
|
|
const N_POINTS: usize = 12;
|
|
let up = DVector::from_column_slice(&[0.0, 0.0, 1.0, 0.0]);
|
|
let down = -&up;
|
|
let twist_motion: DMatrix<_> = (0..N_POINTS).step_by(4).flat_map(
|
|
|n| [
|
|
tangent.proj(&up.as_view(), n),
|
|
tangent.proj(&down.as_view(), n+1)
|
|
]
|
|
).sum();
|
|
let normalization = 5.0 / twist_motion[(2, 0)];
|
|
print!("Twist motion:{}", normalization * twist_motion);
|
|
} |