-
Notifications
You must be signed in to change notification settings - Fork 1
/
05-build-queries.py
71 lines (51 loc) · 1.58 KB
/
05-build-queries.py
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
#!/usr/bin/env python
# coding: utf-8
# In[ ]:
import os
import json
import math
import random
import argparse
parser = argparse.ArgumentParser()
parser.add_argument("dataset")
args = parser.parse_args()
dataset = args.dataset
def nCr(n,r):
f = math.factorial
return f(n) // f(r) // f(n-r)
cur_dir = os.path.join(os.getcwd())
# In[ ]:
good_gold_set = {}
for filename in os.listdir('data/eval/filter_sets/{}/'.format(dataset)):
with open('data/eval/filter_sets/{}/'.format(dataset)+filename, 'r') as fin:
setname = filename.split('.')[0]
data = fin.readlines()
ents = []
for line in data:
ents.append(line.strip('\n'))
good_gold_set[setname] = ents
# In[ ]:
ttl = 0
print("{} sets".format(len(good_gold_set)))
for st in good_gold_set:
print("===={}====".format(st))
if not os.path.exists('data/eval/queries/{}'.format(dataset)):
os.makedirs('data/eval/queries/{}'.format(dataset))
with open('data/eval/queries/{}/{}.query'.format(dataset,st), 'w') as fout:
c = 0
for i in range(2,6):
if dataset == "ap89":
leng = min(nCr(len(good_gold_set[st]),i), 50)
else:
leng = min(nCr(len(good_gold_set[st]),i), 100)
c += leng
samples = []
while len(samples) < leng:
tmp = tuple(sorted(random.sample(good_gold_set[st],i)))
if tmp not in samples:
samples.append(tmp)
fout.write(str(samples)+"\n")
ttl += c
print(c)
# In[ ]:
print(ttl)