Commit 3eb0c388 authored by Médéric Boquien's avatar Médéric Boquien
Browse files

Implement a cache for the computation of the optimal grid when merging two...

Implement a cache for the computation of the optimal grid when merging two grids. As the best_grid() function takes numpy arrays as arguments, it makes caching difficult as they are not hashable. To solve the problem we define a helper function to do the memoisation based on the size, min, and max values of each of the input arrays.
parent 38e7f35b
......@@ -40,7 +40,38 @@ def nu_to_lambda(frequency):
return 1.e-9 * c / frequency
def best_grid_memoise(f):
Memoisation helper for the best_grid() function. Given that best_grid takes
numpy arrays that are not hashable as input, we cannot use the standard
memoisation functions. Here we use the array sizes, minimum and maximum
values as hashes.
f: function
Function to memoise.
best_grid_helper: function
Meomoised best_grid function
memo = {}
def best_grid_helper(x,y):
sx = x.size
minx = x[0]
maxx = x[-1]
sy = y.size
miny = y[0]
maxy = y[-1]
if (sx, minx, maxx, sy, miny, maxy) not in memo:
memo[(sx, minx, maxx, sy, miny, maxy)] = f(x, y)
return memo[(sx, minx, maxx, sy, miny, maxy)]
return best_grid_helper
def best_grid(wavelengths1, wavelengths2):
Return the best wavelength grid to regrid to arrays
......@@ -61,14 +92,13 @@ def best_grid(wavelengths1, wavelengths2):
Array containing all the wavelengths found in the input arrays.
wl = np.concatenate((wavelengths1, wavelengths2))
flag = np.ones(len(wl), dtype=bool)
np.not_equal(wl[1:], wl[:-1], out=flag[1:])
return wl[flag]
def luminosity_to_flux(luminosity, dist):
Convert a luminosity (or luminosity density) to a flux (or flux density).
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment