Profile results

gistfile1.txt

$ kernprof -lv scratch_4.py 


Line #      Hits         Time  Per Hit   % Time  Line Contents
==============================================================
    13                                           @profile
    14                                           def run():
    15         1           47     47.0      0.1      midpoints = k[:-1] + np.diff(k)/2
    16         1           43     43.0      0.1      idx_aux = np.searchsorted(sortedA, midpoints)
    17         1            1      1.0      0.0      idx = []
    18         1            1      1.0      0.0      count = 0
    19         1           16     16.0      0.0      final_indices = np.zeros(sortedA.shape, dtype=int)
    20         1            0      0.0      0.0      old_obj = None
    21        30           23      0.8      0.1      for obj in idx_aux:
    22        29           17      0.6      0.0          if obj != old_obj:
    23        29           14      0.5      0.0              idx.append((obj, count))
    24        29            8      0.3      0.0              old_obj = obj
    25        29           14      0.5      0.0          count += 1
    26         1            1      1.0      0.0      old_idx = 0
    27        30           20      0.7      0.0      for idx_A, idx_k in idx:
    28        29         4824    166.3     11.4          final_indices[old_idx:idx_A] = idx_k
    29        29           20      0.7      0.0          old_idx = idx_A
    30                                           
    31         1          504    504.0      1.2      final_indices[old_idx:] = len(k)-1
    32                                           
    33         1        36792  36792.0     86.9      indicesClosest = final_indices[inv_indices_sort]

scratch_4.py

import time
import numpy as np

A = np.random.random(3000000)
k = np.random.random(30)

indices_sort = np.argsort(A)
sortedA = A[indices_sort]

inv_indices_sort = np.argsort(indices_sort)
k.sort()

@profile  # <---- Remove for running
def run():
    midpoints = k[:-1] + np.diff(k)/2
    idx_aux = np.searchsorted(sortedA, midpoints)
    idx = []
    count = 0
    final_indices = np.zeros(sortedA.shape, dtype=int)
    old_obj = None
    for obj in idx_aux:
        if obj != old_obj:
            idx.append((obj, count))
            old_obj = obj
        count += 1
    old_idx = 0
    for idx_A, idx_k in idx:
        final_indices[old_idx:idx_A] = idx_k
        old_idx = idx_A

    final_indices[old_idx:] = len(k)-1

    indicesClosest = final_indices[inv_indices_sort]

t0 = time.time()
run()
print(time.time() - t0)