Speed up critical parts

jacobnzw / SSMToybox

Nonlinear Sigma-Point Kalman Filters based on Bayesian Quadrature

MIT License

12 stars 0 forks source link

Open jacobnzw opened 5 years ago

jacobnzw commented 5 years ago

Numba might fit my needs.

👉 RBFStudent expectations 👉 utils.maha() 👉 RBF.eval()

jacobnzw commented 5 years ago

According to preliminary results Numba actually makes things worse. It might work if I tune batch_size.