randomize answer
Browse files- dataset/conceptnet_relational_similarity/test.jsonl +0 -0
- dataset/conceptnet_relational_similarity/valid.jsonl +0 -0
- dataset/nell_relational_similarity/test.jsonl +0 -0
- dataset/nell_relational_similarity/valid.jsonl +0 -0
- dataset/t_rex_relational_similarity/test.jsonl +0 -0
- dataset/t_rex_relational_similarity/valid.jsonl +0 -0
- shuffle_candidates.py +25 -0
dataset/conceptnet_relational_similarity/test.jsonl
CHANGED
|
The diff for this file is too large to render.
See raw diff
|
|
|
dataset/conceptnet_relational_similarity/valid.jsonl
CHANGED
|
The diff for this file is too large to render.
See raw diff
|
|
|
dataset/nell_relational_similarity/test.jsonl
CHANGED
|
The diff for this file is too large to render.
See raw diff
|
|
|
dataset/nell_relational_similarity/valid.jsonl
CHANGED
|
The diff for this file is too large to render.
See raw diff
|
|
|
dataset/t_rex_relational_similarity/test.jsonl
CHANGED
|
The diff for this file is too large to render.
See raw diff
|
|
|
dataset/t_rex_relational_similarity/valid.jsonl
CHANGED
|
The diff for this file is too large to render.
See raw diff
|
|
|
shuffle_candidates.py
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
import json
|
| 2 |
+
from glob import glob
|
| 3 |
+
from random import shuffle, seed
|
| 4 |
+
|
| 5 |
+
target = ['nell_relational_similarity', 'conceptnet_relational_similarity', 't_rex_relational_similarity']
|
| 6 |
+
|
| 7 |
+
for t in target:
|
| 8 |
+
for i in glob(f"dataset/{t}/*.jsonl"):
|
| 9 |
+
with open(i) as f:
|
| 10 |
+
tmp = [json.loads(x) for x in f.read().split('\n') if len(x) > 0]
|
| 11 |
+
for n, y in enumerate(tmp):
|
| 12 |
+
answer_original = y['choice'][y['answer']]
|
| 13 |
+
choice = y['choice']
|
| 14 |
+
seed(n)
|
| 15 |
+
shuffle(choice)
|
| 16 |
+
index = list(range(len(y['choice'])))
|
| 17 |
+
seed(n)
|
| 18 |
+
shuffle(index)
|
| 19 |
+
answer = index.index(y['answer'])
|
| 20 |
+
assert answer_original == choice[answer]
|
| 21 |
+
y['choice'] = choice
|
| 22 |
+
y['answer'] = answer
|
| 23 |
+
|
| 24 |
+
with open(i, 'w') as f:
|
| 25 |
+
f.write("\n".join([json.dumps(x) for x in tmp]))
|