A simple implementation of 1-D Kernel Density Estimation with Numba.
KDE_Numba is instantiated and called like scipy.stats.gaussian_kde (same syntax, same arguments). For large arrays, Numba provides a speed-up of one to two orders of magnitude compared to Scipy.
- numpy
- numba
- matplotlib for plotting
- scipy for comparison
- perfplot for benchmarking
Send comments and questions to ablancha@mit.edu