1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
|
# todo combine benchmarks of scorers into common code base
from __future__ import annotations
import timeit
import pandas as pd
def benchmark(name, func, setup, lengths, count):
print(f"starting {name}")
start = timeit.default_timer()
results = []
for length in lengths:
test = timeit.Timer(func, setup=setup.format(length, count))
results.append(min(test.timeit(number=1) for _ in range(7)) / count)
stop = timeit.default_timer()
print(f"finished {name}, Runtime: ", stop - start)
return results
setup = """
from rapidfuzz import fuzz as rfuzz
from fuzzywuzzy import fuzz
import string
import random
random.seed(18)
characters = string.ascii_letters + string.digits + string.whitespace + string.punctuation
a = ''.join(random.choice(characters) for _ in range({0}))
b_list = [''.join(random.choice(characters) for _ in range({0})) for _ in range({1})]
"""
lengths = list(range(1, 512, 2))
count = 1000
def scorer_benchmark(funcname):
time_rapidfuzz = benchmark("rapidfuzz", f"[rfuzz.{funcname}(a, b) for b in b_list]", setup, lengths, count)
time_fuzzywuzzy = benchmark("fuzzywuzzy", f"[fuzz.{funcname}(a, b) for b in b_list]", setup, lengths, count)
results = pd.DataFrame(
data={
"length": lengths,
"rapidfuzz": time_rapidfuzz,
"fuzzywuzzy": time_fuzzywuzzy,
}
)
results.to_csv(f"results/{funcname}.csv", sep=",", index=False)
scorer_benchmark("ratio")
scorer_benchmark("partial_ratio")
scorer_benchmark("token_sort_ratio")
scorer_benchmark("token_set_ratio")
scorer_benchmark("partial_token_sort_ratio")
scorer_benchmark("partial_token_set_ratio")
scorer_benchmark("WRatio")
# token_ratio is unique to RapidFuzz
time_token_ratio = benchmark(
"token_ratio",
"[rfuzz.token_ratio(a, b, processor=None) for b in b_list]",
setup,
lengths,
count,
)
# this gets very slow, so only benchmark it for smaller values
time_token_ratio_simple = benchmark(
"fuzzywuzzy",
"[max(rfuzz.token_sort_ratio(a, b, processor=None), rfuzz.token_set_ratio(a, b, processor=None)) for b in b_list]",
setup,
lengths,
count,
)
results = pd.DataFrame(
data={
"length": lengths,
"token_ratio": time_token_ratio,
"max(token_sort_ratio, token_set_ratio)": time_token_ratio_simple,
}
)
results.to_csv("results/token_ratio.csv", sep=",", index=False)
# partial_token_ratio is unique to RapidFuzz
time_partial_token_ratio = benchmark(
"token_ratio",
"[rfuzz.partial_token_ratio(a, b, processor=None) for b in b_list]",
setup,
lengths,
count,
)
# this gets very slow, so only benchmark it for smaller values
time_partial_token_ratio_simple = benchmark(
"fuzzywuzzy",
(
"[max(rfuzz.partial_token_sort_ratio(a, b, processor=None), "
"rfuzz.partial_token_set_ratio(a, b, processor=None)) for b in b_list]"
),
setup,
lengths,
count,
)
results = pd.DataFrame(
data={
"length": lengths,
"partial_token_ratio": time_partial_token_ratio,
"max(partial_token_sort_ratio, partial_token_set_ratio)": time_partial_token_ratio_simple,
}
)
results.to_csv("results/partial_token_ratio.csv", sep=",", index=False)
|
